TWI225335B - Post-FFT scaling to reduce multiple effects and fine-frequency offset estimation and calculation and use to improve communication system performance - Google Patents

Abstract

Methods and apparatus for scaling at a receiver representations of training signals received from a transmitter, and for estimating and calculating frequency offset and for more accurately determining a channel estimate are presented. A receiver including an equalization circuit that has a scaling circuit to scale representations of signals received from a transmitter, the receiver comprising: a frequency domain transform unit that is to produce a frequency domain representation of at least one training symbol received at the receiver and a frequency domain representation of a data symbol; and a scaling circuit that is to scale the frequency domain representation of the at least one training symbol based upon a largest coefficient in the frequency domain representation of the at least one training symbol to produced a scaled frequency domain representation of the at least one training symbol.

Description

V. Description of the Invention (1) Field The present invention relates generally to communication systems and automatic frequency control. More specifically, the present invention relates to adjusting the representation of a signal to minimize the loss of information in the representation during the operation of the representation. The present invention also relates to adding a coarse frequency estimate to a fine frequency estimate obtained from more data than the coarse frequency estimate, and using the fine frequency estimate to improve the performance of a communication system. Background. The market for home networks is growing at an alarming rate. Service providers from the cable television, telephone communications and digital subscriber line markets are racing to deliver bulk services such as basic telephone services, Internet access and entertainment directly to customers. These services all require high-bandwidth networks capable of delivering 30Mbits / second or even more. The Institute of Electrical and Electronics Engineers (IEEE) 802.11a standard describes a cost-effective, reliable, high-performance local area network (LAN) technology to distribute this multimedia information in the home. A network operating according to the standard 802.11a will use the 5_GHz UNn (Unlicensed International Information Infrastructure) frequency band and can achieve data rates up to 54M bits / second, which is superior to wireless technologies based on other standards Significant improvement. The 802.11a standard has some unique and different advantages over other wireless standards in that it uses vertical frequency division multiplexing (OFDM) 'relative to the spread spectrum and it operates in a clean frequency band of 5 GHz. A technology that solves many problems associated with indoor wireless environments. Indoor environments such as homes and offices are difficult because wireless systems must deal with phenomena called multipath. Multipath comes from the wall,

V. Description of the invention (2) Multiple received wireless signals reflected by the ceiling, floor, lifting equipment, personnel and other objects. In addition, wireless systems must deal with another frequency phenomenon called, attenuation, because of the location of an object associated with the transceiver or a communication device (e.g., telephone, TV), which gives the device access to cable television, telephone, or the Internet. Access to cables or wires by network providers causes signal obstructions to occur. OFDM has been designed to deal with these phenomena, while at the same time using spectrum more efficiently than spread-frequency reads to significantly improve performance. Approved in 1999, the IEEE 802.1la standard significantly improves the performance of indoor wireless networks (54m bits / s vs. 11 Mbits / s). The ability of OFDM to handle multipath and attenuation comes from the characteristics of OFDM modulation. OFDM modulation basically transmits a large number of narrow frequency carriers at the same time, which are sometimes called subcarriers, each of which is modulated at a low data rate, but all of them are added to obtain a very high data rate. Figure 1a illustrates the frequency spectrum of multiple modulation subcarriers in an OFDM system. In order to obtain high spectral efficiency, the frequency response of the subcarriers is overlapping and orthogonal, so the name of OFDM is obtained. Each narrow frequency carrier can use a variety of different modulation formats such as binary phase shift keying (BPSK), quaternary phase shift keying (QPSK), and 90-degree phase difference amplitude modulation qAm (or differential equivalent). Tune it. Because the modulation rate on each sub-carrier is very low, each sub-carrier will experience gentle attenuation and be easy to equalize in a multipath environment, where coordinated modulation is used. The spectrum of the modulated subcarriers is not separated but overlapped. The reason why the information transmitted on the carrier can still be separated is the so-called orthogonal relationship given the method its name. The orthogonal relationship of the sub-carriers requires a sub-carrier to add 1a to the way that all other signals on the frequency at which the signal value is received by 1225335, the description of the invention (3) are zero. In order to preserve this orthogonality, it makes the following true ... 1. The synchronizer of the purifier and the transmitter. This means that they should assume the same 'modulation frequency and the same transmission time ratio (which is usually not the case. 2. Analog components of the transmitter and receiver, part of the high quality. H · s by the Place a protection room without information 3 " Consider the Xiluank channel. This means that some of the signals cannot be used for transmission

Send information. Due to bandwidth limitation and multi-path propagation, the transmission channel between the transmitter and receiver distort the signal being transmitted, causing internal symbol interference (ISI) receivers to identify this channel distortion (or channel estimation), and This effect is taken into account by equalizing the data by using channel estimation. -A method for determining the channel estimation involves the transmission of an adjustment sequence, i.e. a set of fixed data known to both the transmitter and the receiver. By examining how the known, fixed shell material is modified by the channel, real random data can be adjusted to improve information yield.

The channel estimation is performed by converting the time-domain samples of the adjustment sequence to the frequency domain to determine the spectrum of the adjustment sequence when receiving at the receiver. Since the adjustment sequence is known, the spectrum of the adjustment sequence when transmitted from the transmitter is obtained. The quotient of the spectrum of the adjustment sequence received at the receiver and the spectrum of the adjustment sequence sent from the transmitter is the channel estimation or conversion function of the channel. The channel estimate can be smoothed and inverted before it is used to adjust the frequency domain representation of the received data, which involves additional mathematical operations. Mathematical operations with limited precision numbers almost always cause information loss due to surrounds and other errors

6 V. Description of the invention (4) This loss of information is usually not very important. However, if the values in the sampled sequence of channel estimates are relatively small and the accuracy of the format in which the values are stored is relatively low, the smoothing and inversion operations may cause a considerable loss of information about the channel estimates. The loss of information is obvious enough to damage the successful reply of random data, reducing the output. Possible solutions to prevent too much information loss include representing values in floating-point format with a large number of bits to accommodate silk and large signals. Western point notation generally suffers from relatively high power consumption and relatively slow execution speeds. Using a large number of bits consumes a relatively large amount of hardware and power, and may not all meet the needs of large dynamic range and high accuracy. The fractional size that can be represented by the smallest bit in a number format is the accuracy of the format. The maximum number that can be represented is the dynamic range of the number. If the receiver and the transmitter are asynchronous in frequency as described above, the orthogonality of the subcarriers is compromised and the data added to the subcarriers may not be correctly recovered due to the internal carrier interference. Figure lb illustrates the effect of the lack of synchronization in the frequency domain of multiple carriers. The dotted line shows where the spectrum of the subcarrier should be, and the solid line shows where the spectrum is due to lack of synchronization. Because the receiver and the transmitter need to be synchronized for reliable OFDM communication to occur, in fact they are not the case, so the frequency offset between the receiver and the transmitter must be compensated. Offsets can occur due to the inherent inaccuracies of the synthesizers and crystals in the transmitter and receiver, and can drift due to temperature or other reasons. The offset can be compensated at the receiver, but this method only produces a rough estimate of the actual offset. According to a method for compensating for offset, the analog signal received by the receiver is divided into three parts: a short-sequence symbol part, and a description of the invention (5) the timing symbol part and the data symbol part. Some of the short timing symbols in the short symbol section are used for automatic gain control and to detect symbol timing. Other short time series symbols are sampled and digitized and automatically correlated to produce a rough estimate of the offset. A rough estimate of the offset is then used to generate a digital period signal whose frequency is based on the rough estimate of the offset. The digital periodic signal is multiplied by a long-symbol digital sample, and the product is subjected to a fast Fourier transform to produce a channel estimate. When arriving, the digital carrier is also used to multiply the digital samples of the data symbols (digital data samples) to correct the offset. The product of the digital carrier and digital data samples can now be decoded. Because the short sign from which the frequency offset is obtained is relatively short, the estimate of the offset can be quite different from the actual offset. Therefore, there will be a covariance offset, which may cause the street spectrum of one carrier to overlap with the spectrum of other subcarriers. Because of the overlap, when recovering digital data samples, the data of the primary carrier can include interference from a nearby secondary carrier, reducing the output of the communication system. Furthermore, because there is a residual offset, the channel estimate is not a correct representation of the actual transfer function generated by the channel. As mentioned above, the existing solutions cannot provide a channel estimation representation that does not consume a relatively large amount of hardware and power, and it causes a significant loss of information when performing operations on the representation. As mentioned above, existing solutions cannot provide relatively good frequency or channel estimates between a receiver and a transmitter. It is therefore necessary to provide a solution that overcomes the disadvantages of existing solutions. Summary of the Invention Describes a method for scaling on a receiver to be received by a transmitter. 1225335 V. Description of the Invention (8) In the case of 'not well known operations, steps, functions and components to avoid obscuring the invention. Part of the description will use the terms common to those skilled in the art to present the main purpose of expressing the results to those skilled in the art, such as orthogonal multiplexing, fast Fourier transform (FFT), inverse fast Fourier transform (IFFT), autocorrelation 'Time, wave' delay, etc. When various individual steps are performed sequentially, different operations will be explained in a pair that is most helpful in understanding the present invention. However, the order of the description should not be understood as meaning that these actions must be performed in the order in which they are presented or even related to the order. Finally, repeated use, other embodiments, etc. do not necessarily refer to the same embodiment, although they may refer to the same embodiment. FIG. 2 illustrates a communication system according to an embodiment of the present invention. The system 200 includes an inlet 21, which is connected to a public switched telephone network (PSTN) via a cable (or cables), a cable television system, an Internet provider (ISP), or some other system. The entrance 21o includes a transceiver 210 'and an antenna 211. The home appliance 220 includes a transceiver 220 and an antenna 221. The home appliance 220 may be a television, computer, telephone, or other home appliances. The transceiver 210 provides the transceiver 220 for a wireless connection to a system connected to the portal 210. According to an embodiment, the transceivers 210 and 220 communicate according to the IEEE 802.1la standard. Therefore, each transceiver 21, and 220, including a receiver and a transmitter, exchanges information in the format of the 802.11a standard. In other examples, as noted below, the transceivers 21 0, and 220 may have some of the design features derived from the IEEE 802.11a standard. Figure 3 illustrates the packet structure required by the IEEE 802.11a standard to transmit and receive data between the two transceivers 11 5. Description of the invention (9). One of the transceivers 210, and 220, is designed to receive packets such as packet 300 and obtain timing information, data, and other information from the packets. For example, in the packet 300, the first 10 symbols (tl to t10), which are called short symbols, are repeated sequences used by a receiver to detect symbol timing and coarse carrier frequency offset. GI1 is a ring prefix of two long symbols T1 and T2, which is sometimes called a guard interval, because it is used as a rough internal symbol boundary to absorb multipath effects with Θ. Make GI1 long enough so that if the short symbol t10 goes through multiple paths, the symbol ti0 will be partially, and will be stuck in `` Gil without affecting the long symbol T2. T2 is used for channel estimation and fine symbol timing For adjustment. Because 0Fdm is extremely sensitive to the carrier frequency offset between the transmitter and receiver, the present invention uses T1 and T2 to provide continuous estimates (fine frequency offset estimation) to reduce the residual offset after any short symbols. According to an embodiment, each short symbol costs 0.8 μ8, all of which are 8 μ8 to perform automatic gain control (AGC) and coarse symbol timing and frequency offset estimation. According to an embodiment, GI1 takes 1.6 ps, which is about twice the amount of regular circular prefixes between data symbols. After the short symbol, when the relatively long GI1 is made to provide a sufficient buffer area to absorb errors in any symbol boundary, GI1 provides a multiple path that enables two long symbols T1 and T2 to be obtained without multiple paths from short symbols The approximate inter-symbol boundary of the effect. According to an embodiment ', when the data bits transmitted in T1 and T2 are known at the receiver, T1 and T2 each occupy approximately 3.2 ps and are used to obtain an estimate of the characteristics of the two channels. The two-channel estimates are combined and processed to form a reference channel estimate for subsequent data symbols. After the long symbol, the packet enters the data symbol. Each 12 1225335 V. Description of the invention (l4) The frequency difference generated based on the correlation of short symbols is called a coarse frequency offset estimate / frequency difference is passed to the signal generator 422, which generates A sine wave with a frequency difference of the output frequency. By causing the generator 422 to generate a sine wave having a frequency equal to the offset between the synthesizers, the mismatch between the synthesizers can be compensated. After correlating the short symbols and generating a rough offset estimate, the long symbols pass through the antenna 412 and AGC 413 and reach the mixer 414, where they are pulled down to the fundamental frequency or an intermediate frequency. According to one embodiment, the ADC 418 samples and digitizes at a rate of 20 million samples per second to generate 64 samples per long symbol. In one other embodiment, the ADC 418 samples each long symbol, which translates to a rate of 40 million samples / second. The mixer 420 multiplies the digital long symbol by a digital sine wave (digital periodic signal) generated by the generator 422. Because the sine wave generated by the generator 422 is based on a coarse frequency offset estimate on the output of the mixer 420, the adjusted samples will still have a residual offset. According to an embodiment, the output of the mixer 420 generated by the first long symbol is passed to a fast Fourier transform (FFT) unit, which performs the fast Fourier transform of the output and stores it in the memory 425. Similarly, a fast Fourier transform is performed on the output of the mixer 4 2 0 due to the second long symbol and stored in the memory 425. The averaging circuit 427 takes the conversion of each long symbol and averages them to provide the average of the conversion to the convolutional device 436. According to an embodiment, the output of the mixer 420 generated for each long symbol is Fourier-transformed separately. In addition, while the output of the mixer 420 is performing a fast Fourier transform according to an embodiment, it should be appreciated that other types of transforms known in the art (such as Hibbert transforms) can also be used to obtain the time domain table of the signal

17 1225335

5. Description of the invention (IS) method and convert it into a frequency domain table * method. The unit performing the time-domain to frequency-domain # conversion is referred to herein as a frequency-domain conversion unit.

The output of the averaging circuit 427 is its frequency domain representation when the two long symbols have been modified by the channel between the two transceivers. As described below, the two long symbol frequencies or representations can be used to generate the channel transfer function estimate (or channel estimate 4). The channel estimate can be inverted and used to flip the channel at the transceiver 21. Effect. Because the samples multiplied by the fast Fourier transform and the frequency is a sine wave based on a rough offset estimate, the frequency i or representation of the received number will get a residual offset. Therefore, the frequency domain representation produced by the averaging circuit 427 cannot be used to produce a correct representation of the actual channel transfer function until any residual offset is compensated. Any residual offset can be complemented by sampling with long symbols after generating a fine offset estimate

In order to produce a fine offset estimate, samples of the long symbols generated on the output of ADC 418 must first pass through column 426 and conjugate 428. Column 426 delays the digital sampling of the first long symbol of the two long symbols by the duration of one long symbol. The digital samples of the second long sign are changed to the complex conjugate produced by the complex conjugate 428. When a complex conjugate of each sample of the second long symbol is generated, it is multiplied by a digital multiplier 430 by the corresponding sample from column 426. The product of the multiplier 430 is added by the integrator 432. After the flicker product produced by the multiplier 430 has been added by the integrator 432, the wheel output of the integrator 432 is a complex value or a vector, which has between The estimated angle of the fine frequency offset. The frequency offset estimation generator 440 converts the vector output by the integrator 432 18 1225335

V. Description of the invention (l6)

The angle is divided by the duration of a long symbol, or more generally by the time between the beginnings of the two long symbols. The generator generates the remaining difference in frequency between the synthesizers in the transceivers 210 and 220. Because the digital long symbol samples have been multiplied by a frequency based signal based on the coarse offset estimation, the output of the generator 440 is the remaining frequency difference between the synthesizers in the transceivers 21o and 22o. The frequency difference between these synthesizers based on the correlation of long symbols is called a fine offset estimation. The fine offset estimate is passed to a signal generator 422, which generates a sine wave having a frequency equal to the sum of the fine frequency offset estimate and the coarse frequency offset estimate. By causing the generator 422 to generate a sine wave having a frequency equal to the residual offset between the synthesizers, the mismatch between the synthesizers can be further compensated. As described above, because the digital long symbol samples that have been fast Fourier transformed by the FFT unit 424 are multiplied by a signal having a frequency equal to the coarse offset estimate, the frequency domain representation of the received signal may not be converted by the channel The very accurate representation of the number of consecutive sending. This unmarried marriage

疋 Caused by the existence of a residual frequency offset. This residual frequency offset can be estimated and compensated using fine offset estimation. Because the frequency domain representation of the received signal is stored in the memory 425, the frequency domain representation of the received signal needs to be convolved with a frequency 0 domain representation of the signal with a frequency equal to the fine offset estimate f0 . The frequency domain representation of the windowed complex sine wave sampled for one-to-one finite periods has a general shape as a function-δίη (χ) / χ. On ® the frequency domain representation of a sine wave varies as a function of f0. According to an embodiment, the 'convolution 436 equates three samples of a frequency domain representation of a sine wave with a frequency equal to the fine offset estimate and the frequency of the received signal stored in the memory 425 19 1225335 V. Description of the invention (π )

Domain representation does convolution. The frequency domain compensator 434 retrieves three samples from the memory 438 of the frequency domain representation of a sine wave having a frequency equal to f0. In order to perform the convolution as quickly as possible, the memory 438 stores a table with correlated samples of the frequency domain representation of a sine wave with a frequency of f0 equal to many different values of fo. In order to obtain proper sampling, the compensator 434 first calculates the fine offset estimate f0 based on the output of the integrator 432, and then maps it to the table based on f0. In one embodiment, the compensator 434 obtains only the nearest term. In other embodiments, if the calculated fine offset estimate falls between two f0 values in the memory 438, the compensator 434 obtains a sample that is correlated with the two values. The compensator 434 then interpolates between each sample of one value and the corresponding sample of the other values to produce an interpolated sample value. The compensator 434 then provides the interpolated sample values to obtain the calculated fine offset estimate for the convolutional device 436. It then convolves the interpolated sample values with the frequency domain representation of the long symbol modified by the channel. . The output of convolution 436 is the frequency domain representation of the long symbols received at the receiver and adjusts the frequency offset between the transmitter and receiver. The output of the convolutional device 436 is then stored in the memory 441. In contrast to the memory 441 which stores the offset-adjusted frequency domain representation of the long symbols received on a receiver 400, the memory 442 generates a length which has been generated on the transceiver 210 'for transmission to the receiver 400. The frequency domain representation of the symbol is stored therein. The circuit 446 obtains the offset-adjusted frequency domain representation of the long symbol from the memory 441 and divides it by the frequency domain representation of the long symbol generated by the memory 442 on the transceiver 21o to generate a channel estimate. To be stored in the memory 448. In the above description, the memory material 2 stores the frequency-domain representation of the long symbol generated on the transceiver 210 while it should realize 20 V. Description of the invention (IS) = ::. Γ The memory in the embodiment 442 may place the time-domain table M_material of the long symbol generated on the transceiver ⑽. In another embodiment, the Fourier transform unit is placed in the memory 442 and the circuit 446 and converts the time-domain representation in the memory 442 into a frequency-domain representation suitable for the divider in the circuit 446. . 2 Purchase of 8t materials can be obtained by other circuits (not shown), and used to correct the illegality of the data symbols that arrive after the long symbol. The frequency domain representation of the frequency of the sine wave of the offset estimate-while sampling-in other embodiments, the compensator may store an equation for each sample. This equation describes how the complex value of the sample changes as a function of the precision, field offset, and estimate. After the compensator calculates the fine offset estimate, it compensates for n434 counts, etc. to determine the fine offset estimate for each sampling pair. After the compensator 434 supplies the sample value to the convolutional device 436, it Convolution with the frequency i or representation of the received signal stored in the memory 425. In the above description, the search table 438 stores only three samples for each fine offset estimate. At the same time, the actual number of samples stored for each fine offset estimate should be-not three Number and similar to design considerations' While storing three equations in the compensator in the description above, you should realize that the actual number of equations is a design consideration and may not be three, but equal to the required The number of samples. Figure 5 illustrates a receiver according to an embodiment of the invention. Receiver 1225335 V. Description of Invention (l9) The receiver 500 operates in a similar manner to the receiver 400. Therefore, it is not necessary to repeat the operating instructions of most components. The difference between the receiver 500 and the receiver 400 is the way in which the channel estimation is performed. Instead of Fourier transforming the output of the mixer 420, the output of the mixer 420 generated by long symbol sampling (coarse offset adjustment long symbol sampling) is stored in the memory 520 until the integrator 432 has a Is a vector indicating the estimated angle of the fine offset between the synthesizers of the transceivers 210 'and 220'. When the integrator 432 generates an angle that is an indication of the fine offset, the signal generator 524 calculates the fineness by dividing the angle by the duration of a long sign, or more generally by the integration duration of the integrator 432 Offset estimation. The signal generator 524 then generates a digital sine wave with a frequency equal to the fine offset estimate. The mixer 522 obtains a coarse offset-adjusted long symbol sample from the memory 520 and multiplies it by a digital sine wave generated by the generator 524. Then, the output of the mixer 522 is subjected to fast Fourier transform by the FFT unit 526, and the output of the FFT unit 526 is stored in the memory 527. The mixer 5 2 2 then obtains the coarse offset adjustment long symbol of the second long symbol from the memory 5 2 0 and multiplies it by the digital sine wave generated by the generator 524. Then, the FFT unit 526 performs fast Fourier transform on the output of the mixer 522, and the output of the FFT unit 526 is stored in the memory 527. The averaging circuit 528 obtains each offset-adjusted long symbol, averages the conversions, and stores the average in the memory 440. According to an embodiment, the units 510 and 526 are the same unit. Once the coarse and fine offsets have been calculated, the FFT unit 510 produces, on its output, a Fourier transformed representation of the data symbols and guards. Described below

22 1225335 V. Description of the invention (20) In the embodiment, the output of the unit 510 is used to provide an updated estimate of the offset between the receiver and the transmitter. The description given above including 'other embodiments' in connection with FIG. 4 is also applied to FIG. 5 and need not be repeated here.

In the above description, the frequency offset is estimated by the autocorrelation long or short symbol. The frequency offset can also be updated during the reception of data symbols. During reception of a data symbol, the frequency offset between the transceivers can be re-estimated by estimating the difference between the phase of the pilot carrier in a data symbol and the phase of the pilot carrier during a long symbol. FIG. 6 illustrates a circuit for updating a frequency offset according to an embodiment of the present invention. In circuit 600, a divider circuit 610 receives the output of the FFT unit 605 and the output of the memory 448 in which the channel is stored. The FFT unit 605 generates a frequency domain representation of the received data symbols. The divider circuit 610 divides the output of the FFT unit 605 by the channel estimate.

According to an embodiment, the output of unit 605 is 64 samples of the frequency domain representation of the received data symbol. In another embodiment, the output of unit 605 is 128 samples of the frequency domain representation of the received signal. It should be appreciated that the number of samples is a design consideration and is linked to the number of samples generated by ADC 418 for each long-term symbol. In the embodiment where the unit 605 generates 64 samples, the sample represents a frequency band extending from -10 MHz to +10 MHz. Because only 16.5MHz is used for transmission, there are 52 samples representing data transmission and the other samples only represent guard bands between 20MHz wide channels in an 802.11a standard-compliant system. In the case of 128 samples, the outer 64 are adjacent channels. The 52 samples represent 52 carriers, four of which are pilot carriers, which are used to monitor signal length and carrier phase. According to Yishi 23 1225335 V. Description of the Invention (2l) embodiment, i: 7 and 21 samples are used to indicate the carrier. When the circuit 61 divides the 64 samples of the frequency domain representation of the received data symbol by the channel estimate, the phase of the quotient of the sampling of the reference carrier on which it appears indicates the reference carrier of the data symbol The phase in and the corresponding in the long symbol indicate the difference in the phase difference in the carrier. The average offset circuit 62 selects the quotient of a non-carrier sample on which it appears and determines the average phase difference by adding the phase difference for each pilot carrier and dividing the sum by the number of pilot carriers. The number of carriers shown is four according to an embodiment. According to an embodiment, if the size of the smallest pilot carrier is smaller than one-eighth of the size of the largest pilot carrier, the quotient phase of the smallest pilot carrier is not included in determining the average phase difference. Furthermore, the circuit 62 does not consider the angle of the smallest carrier and uses linear interpolation and the angle of the quotient of the two nearest neighbors of the carrier to obtain the substituted angle. Then, the average phase difference is obtained by adding the phase difference of each of the pilot carriers, including the replacement angle to the minimum quotient, and dividing the sum by the number of pilot carriers. . After determining the average phase difference, the circuit 62 divides the difference by the time elapsed because the fine offset estimation is calculated to determine an update frequency offset, which is even after correction using coarse and fine offset estimates. A measurement of the frequency offset between transceivers. The updated frequency offset is then added to the digital signal generator to generate a digital sine wave to correct the frequency mismatch between the transmitter and receiver. The frequency of the sine wave is the sum of the updated frequency offset and coarse and fine offset estimates. You should understand that you can update the frequency offset by determining the phase difference between the pilot carrier in the channel estimation and the pilot carrier in a data symbol, such as just connected 24 1225335 V. Description of the invention (22) End 6 The figure can also be used in the embodiment linked to Figure 5. In such an embodiment, the divider circuit 601 receives the output of the FFT unit 510 and the channel estimation from the memory 448.

The frequency offset can also be measured by measuring the phase difference of the pilot channel in one of the two data symbols or by measuring the difference between the terminal portion of a data symbol and the ring prefix (or guard interval) of the data symbol. The phase difference is updated. The phase difference in the pilot channel in the two data symbols divided by the time elapsed between the two data symbols is a measure of the frequency offset between the transceivers. Similarly, the phase difference between the terminal portion of a data symbol and its ring prefix divided by the time elapsed between them is a measure of the frequency offset between the transceivers. Figure 7 illustrates a circuit for updating a frequency offset according to other embodiments of the present invention. While the circuit 700 will be described by calculating the frequency offset by estimating the phase difference in the pilot channel of one of the two data symbols, it should be appreciated that the circuit 700 can also be used to estimate the terminal portion and the symbol of a data symbol. Phase difference between guard intervals. In the circuit 700, the output of the receiving unit 705 of the divider circuit 710 is generated by the data symbol at time τ0, and the output is stored in the memory 712. At a certain time To + At, where At is equal to an integer multiple of the duration of a data symbol, the divider circuit 7H) receives the output of unit 705 due to other data symbols and stores the output in the memory 712. Unit 7G5 generates a frequency domain representation of the received signal. The divider circuit 710 divides the first data symbol stored in the memory 712 by the frequency domain representation of the second data symbol. According to an embodiment, the output of the convolutional device 436 is a frequency-domain representation of a data symbol lacking 64 samples. In another embodiment, the output of the unit 705 25 1225335 V. Description of the invention (23)

12 8 samples for the frequency domain representation of the received signal. It should be appreciated that the number of samples is a design consideration and can be linked to the number of samples generated by the ADC 418 for each long time series symbol. In the embodiment where the convolutional device 436 produces 64 samples, the samples represent a frequency band extending from -10 MHz to +10 MHz. Because only 16.5MHz of 20MHz is used for data transmission, there are 52 samples for data transmission, and the remaining samples only indicate a guard band between 20MHz wide channels of an 802.11 & standard-compliant system. 52 samples represent 52 carriers, four of which are pilot carriers, and they are used to monitor the signal length. According to an embodiment, ± 7 and 21 samples are used as pilot carrier samples. When the circuit 710 divides the 64 samples of the frequency domain representation of the first data symbol by the frequency domain representation of the second data symbol stored in the memory 712, the proportion of samples on which a lead carrier appears The phase of indicates the difference between the phase in the pilot carrier of the first data symbol and the phase in the corresponding pilot carrier of the second data symbol. The average offset circuit 72 selects the quotient of the sampling on which a pilot carrier appears and determines the average phase difference by adding the phase difference for each pilot carrier and dividing the sum by the number of pilot carriers. The number of pilot carriers is four according to an embodiment. According to an embodiment, if the magnitude of the smallest quotient of a pilot carrier is less than one-eighth of the magnitude of the largest pilot carrier, the quotient phase of the smallest pilot carrier is not included in determining the average phase difference. Furthermore, the circuit 720 does not consider the angle of the smallest carrier and uses linear interpolation and the angle of the quotient of the two nearest neighbors of the carrier to obtain the substituted angle. Then by adding the phase difference for each pilot carrier, including the replacement angle to the minimum quotient, and dividing the sum by the number of pilot carriers, the average phase difference is obtained. The number of pilot carriers is based on an implementation 26 1225335 5 4. Description of Invention (24) Examples are four. After determining the average phase difference, the circuit 720 divides the difference by the time elapsed between receiving the two data symbols on the antenna 412 to determine the measurement of the frequency offset between the transceivers. This updated frequency offset is then added to the digital signal generator 4 2 2 which adds the updated frequency offset to the coarse and fine offsets and generates a digital sine wave to correct the frequency between the transmitter and receiver Mismatch.

You should realize that the frequency offset is updated by determining the phase difference between the pilot carriers in the two different data symbols. As just illustrated in Figure 7 for illustration, it can also be used in the embodiment described in Figure 5 . In such an embodiment, the divider circuit 710 receives the output of the FFT unit 510.

Figure 8 illustrates a receiver according to an embodiment of the invention. The receiver 800 operates in a similar manner to the receiver 400. Therefore, it is not necessary to repeat the description of the operation of most components. The difference between the receiver 800 and the receiver 400 is the enhancement in the receiver 800, which enables coarse and fine frequency offsets to be determined more correctly. The enhancement of the receiver 800 is a filter 810 for moving the DC offset in the samples from the mixer 420. According to an embodiment, the filter 810 is a low-pass infinite impulse response (HR) filter, but other embodiments may have a different type of filter. The integrator 820 adds the low-pass filtered samples from the filter 810 to the samples from the mixer 420. Because the sampled DC component is removed, the angle from the integrator 432 is more accurate. So 'fine and coarse offset estimation is more correct. Another way to compensate for the occurrence of DC in the signal is to calculate the DC offset that appears in the short and long symbols. Because between the sender and receiver

1225335 V. Description of the invention (25) There is a carrier frequency offset, so the DC offset introduced by the receiving chain is not on the DC of the OFDM signal spectrum transmitted. If this carrier frequency offset is corrected by k before the DC offset correction, the receiver DC offset will move to a frequency with the opposite sign of the carrier frequency offset. For example, per million (ppm) in a carrier with a frequency of 5.25GHz—the uncertainty 40 part corresponds to an offset of 21 OKHz, about 2/3 of the frequency separation between carriers. Figure 9 shows the spectrum of the received 802.11a OFDM symbols, including the carrier slot, and the DC offset of a receiver. As shown in Figure 9, for any non-zero frequency offset, the receiver DC offset will include contributions from nearby data accommodation, as indicated by the OFDM modulation of the 802.11a standard. However, there is always a certain amount of carrier slot from the power amplifier located on the carrier frequency, which translates into DC accommodation after down conversion, and therefore the DC in the transmitted signal spectrum is not exactly zero. According to the 802.1la standard, the power of the carrier slot can be as high as 15dB below the signal power. Assuming that each data carrier has approximately the same amount of power, the power of the carrier slot can actually be higher than the power of each data carrier (-15dB> l / 52), and therefore cannot be ignored. According to an implementation method, the receiver DC offset can be as large as + / _ 100mV. Therefore, according to an embodiment, the full range of ADC 418 is from -500011 ¥ to 500mV. The power of i) C offset can be significantly higher than the power of a data carrier. Most DC offset algorithms use filters. However, because typically only 4x32 = 128 samples are left in the short symbol, the bandwidth of the filter cannot be very narrow. As shown in Figure 9, any filtering operation with a bandwidth greater than the carrier frequency offset will pass the carrier slot and DC offset, and therefore cannot be a correct DC offset estimator. To shift the DC offset from the signal frequency

28 1225335 V. Description of the invention (26) The other parts of the spectrum are separated. We must rely on the fact that the carrier slot system and the data carrier frequency are locked. At the same time, the DC offset is just a signal added to the receiver. Figure 10 illustrates a receiver according to an embodiment of the invention. The receiver 1000 operates in a similar manner to the receiver 400, so it is not necessary to repeat the description of the operation of most components. The difference between the receiver 1000 and the receiver 400 is the enhancement in the receiver 100, which allows the frequency offset to be determined more accurately. The receiver 1000 is enhanced by adding circuitry to determine the DC offset. The receiver 1000 separates the receiver's DC offset from the transmitted spectrum by taking two quick views of the same transmission symbol and calculating the DC offset from the difference between the two quick views. Because the short symbol is a sequence of repeatable identical symbols, two short symbols are used to calculate the DC offset. If AGC 413 completes its operation quickly without occupying too many short symbols, the remaining short symbols can be used for more accurate estimation. According to one embodiment, 2 short symbols are used for coarse DC offset calculation based on coarse symbol timing. It should be appreciated that the number of short symbols used for DC offset calculation is design related, and the present invention includes the use of non-two short symbols. If a coarse frequency offset is known, the phase difference α between 32 samples can be calculated (64 samples if 4 short symbols are available). The sign of A is defined such that if the transmitter carrier frequency is higher than the receiver carrier frequency, then α is positive. This factor will be used to correct the DC offset calculation at the end of the short symbol. If there is a non-zero frequency offset, compared to the DC offset introduced at the receiver, the spectrum of the signal transmitted for every 32 samples will be rotated by this amount of phase. If two short symbols are accumulated separately and are called xl and x2, the DC offset can be calculated as follows: 29 1225335

V. Description of the invention (27) DC offset = (xl-x2) eUa) ~ 32 (l ^ eUa)) (Equation 2)

The receiver 1030 includes an integrator 1010 that subtracts the dc offset from the short symbol used for DC offset measurement by the receiver to the symbol received later. According to an embodiment, because the DC offset cannot be measured until a short symbol has been received and used to determine the coarse frequency offset, the integrator allows the sampling of the short pay number to pass unaffected. In the case where only two short symbols are available for DC offset calculation, the integrator 1020 accumulates the samples of the first short symbol (the first two short symbols are used when D c offset calculation is performed with four short symbols). ) And provide the sum to the DC offset compensator 1030. The integrator 1020 then accumulates the sample of the second button symbol (if four short symbols are used to make the Dc offset meter two, the last two short symbols) and provides the sum to the compensator 1030. When the integrator 432 has generated the coarse offset estimation as described above in connection with FIG. 4, the compensator 10 finds the value of Equation 2 above to determine the DC offset. When the frequency offset is large, the DC offset obtained by using Equation 2 is more correct than the small frequency offset, so that (I-e) will not be a small number in the knife mother. If the frequency offset is actually very small, and the (l-eja) case will be very close to zero, the above equation will introduce too much noise enhancement to be useless. If the frequency offset is actually very small, the related filtering technique explained in connection with Figure 8 will work well, as long as the carrier slot should be encountered as part of the DC offset (which is redefined in the spectrum). Because the coarse offset is available at the end of the short symbol, when the frequency offset is relatively large, the compensator 1030 uses the above equation and the coarse offset to determine the DC offset, or when the frequency offset is small, the compensation The device 1030 uses only (xl + x2) / 64 (Equation 3) to calculate the DC offset. 30 1225335 Fifth, the description of the invention (30) For the month (30), the convolver 43 6 convolves the average of the conversion with a frequency-domain representation of a sine wave to minimize the effect of any residual offset. The operation of the circuit 1100 from the received flood number storage 440 to the memory 448 is as described above and need not be repeated.

After the channel estimation reaches the memory 448, the smoothing circuit obtains the channel estimation from the memory 448 and smoothes it using a finite impulse response (FIR) filter. According to one embodiment, the embodiment has seven Taps, but other numbers of taps are also possible and design related. Smoothing reduces noise effects on the value of the channel estimates. The inverting circuit 1135 then inverts the smoothed channel estimates, stores the inverted and smoothed channel estimates until the frequency domain representation of a data symbol reaches the multiplier 114.

Before the sampling of a data symbol reaches the multiplier 1140, it needs to reach the unit 424 first. The operation of the components between the antenna 412 and the multiplier 420 produces a digital time-domain representation of the data symbols on the fundamental frequency or IF, as explained in the link to Figure 4 above, and need not be repeated here. After its appearance from the multiplier 42, the 'unit 424 performs a Fourier transform of the digital time-domain representation of the offset correction of a data symbol. The gain rising circuit 11110 scales the frequency-domain representation of the data symbols in the same manner as described in connection with FIG. 12 above. The multiplier i 14〇 multiplies the scaled frequency domain representation of the data symbol by the inverted and smoothed channel estimates from the circuit 丨 20 to obtain the frequency domain representation of the data symbol of the equalized channel effect. Figure 13 illustrates a receiver according to an embodiment of the invention. Most of the receiver 1300 operates in a similar manner to the receiver 1100, and most of the operations of its components need not be repeated here. The basic difference is that before the multiplier 1340 does the multiplication, the gain increase in the receiver 1300 is only estimated for the channel ---- 33 1225335

V. Description of the invention (3l)

The fact that it occurred without a sign of information. Therefore, a gain increase after the multiplier 1340 is necessary. The gain increase occurs only for channel estimation, because the frequency domain representation of a data symbol leaves the unit 424 and reaches the multiplier 2440 without any intermediate gain increase. The gain increasing circuit 131o operates in the same manner as the gain increasing circuit 1110, and need not be repeated here. On the other hand, the gain rise repeats the circuit 350, and according to one embodiment, the program 125 is not executed, but in another embodiment, it can execute the program. The repeat circuit 135 obtains the minimum number of left shifts from the gain rise circuit 1310, which is performed on the coefficients of the frequency domain representation of the long symbol. The repeating circuit 135 performs the same number of minimum left shifts on the output of the multiplier 1340. In the embodiment where the repeating circuit 135 repeats the procedure 1200, the circuit 1350 does not receive the minimum left shift number from the circuit 1310, which is performed on the coefficients of the frequency domain representation of the long symbol.

Thus, methods and devices have been described for scaling the representation of an adjustment signal received from a transmitter at a receiver and calculating a frequency offset, and for more accurately determining channel estimates. Although the present invention has been described with reference to specific exemplary embodiments, those skilled in the art can make various modifications and changes to these embodiments without violating the broader spirit and scope of the present invention, as proposed in the scope of patent application of. Accordingly, the descriptions and drawings are to be regarded as illustrative rather than restrictive. 34

Claims (1)

Sixth, the scope of application for patent No. 911217S1 Application for amendment of patent scope May 27, 1993 1. A receiver including an equalization circuit with a scaling circuit to scale the representation of a signal received from a transmitter The receiver includes a frequency domain representation conversion unit that generates a frequency domain representation of at least one adjustment symbol and a frequency domain representation of a data symbol received on the receiver; and a scaling circuit that The frequency domain representation of at least one adjustment symbol is scaled based on the largest coefficient in the frequency domain representation of at least one adjustment symbol to produce a scaled frequency domain representation of at least one adjustment symbol. 2. The receiver of item 1 of the patent application scope, further comprising a divider circuit, which is to divide the scaled frequency domain table of at least one adjustment symbol by the frequency domain of the adjustment symbol not affected by the channel effect Representation to produce a channel estimate. 3 · If the scope of the patent of Zhongli is signed 2 J: Bai Xi Bi, Λ * 35. Mr Domain notation. For example, the receiver of the fourth scope of the patent application, wherein the scaling circuit is t, and the scope of the patent application is to be expressed in the frequency domain of at least-adjustment symbol when the maximum coefficient in the frequency domain representation of at least-adjustment symbol is less than-critical value The maximum coefficient in the method is used to scale the frequency domain representation of at least one adjustment symbol. 6. The receiver according to item 4 of the patent application scope, further comprising an inversion circuit, which is to generate an inverted channel estimation by inverting the channel estimation. 7. The receiver according to item 4 of the patent application scope, further comprising: a smoothing circuit to filter the channel estimation; and an inversion circuit to generate an inversion by inverting the channel estimation Passed channel estimates. 8. The receiver according to item 7 of the patent application scope, further comprising a multiplier, which multiplies the frequency domain representation of the data symbol by the inverted channel estimate. 9. The receiver as claimed in item 7 of the patent application scope, further comprising a multiplier, which is a multiplier that adjusts the frequency domain representation of the data symbol by the inverted channel estimate to produce the channel of the data symbol adjusted. Frequency domain representation; and a second scaling circuit that scales the channel-adjusted frequency domain representation of the data symbol based on the second largest coefficient in the frequency domain representation adjusted by the channel of the data symbol . 10. The receiver according to item 6 of the patent application, wherein the scaling circuit is to scale the frequency domain representation of the data symbol to generate a scaled frequency domain representation of the data symbol, which further includes a multiplier, which is to 1225 The scaled frequency domain representation of the data symbol is multiplied by the inverted channel estimate. U. The receiver according to item 4 of the patent application, wherein the frequency offset compensation circuit includes an offset compensator, which is to generate a periodic signal having a frequency equal to the frequency offset between the receiver and the transmitter. Frequency domain representation, and A convolver, which is to convolve the frequency domain representation of the periodic signal and the scaled frequency domain representation of at least one adjusted symbol to produce an offset-compensated frequency domain representation of at least one adjusted symbol, It substantially removes the frequency offset effect between the receiver and the transmitter. 12 · The receiver of item 4 of the patent application, wherein the scaled frequency domain representation of at least one adjustment symbol includes a scaled frequency domain representation of the first adjustment symbol and a scaled representation of the second adjustment symbol The frequency domain table is not correct 'The circuit further includes: An averaging circuit that generates a scaled frequency domain representation of the first and second adjustment symbols; and wherein the frequency offset compensation circuit includes an offset compensator that generates an A frequency domain representation of a periodic signal with a frequency offset between transmitters; and a convolutional device that is to scale the frequency domain representation of the periodic signal and a scaled frequency domain representation of at least one adjustment symbol Convolution is performed to generate at least one offset-adjusted frequency-domain representation of the adjustment symbol, which substantially removes the frequency offset effect between the 38 #, patent-pending receiver and transmitter. U. The receiver as claimed in claim 12, wherein the divider circuit is to divide the offset-compensated average of the scaled frequency domain representation of the first adjustment symbol and the second adjustment symbol by a substantially unacceptable The frequency domain representation of the adjustment symbols affected by the channel effect to produce a channel estimate. 14. The receiver according to item 13 of the patent application scope, further comprising an inversion circuit for inverting the channel estimate to generate an inverted channel estimate. 15. The receiver as claimed in item 12 of the patent application scope, further comprising:-a smoothing circuit that filters the channel estimation; and an inversion circuit that generates an inversion pass by inverting the channel estimation. Channel estimates. 16. For example, the receiver of claim 15 of the patent application scope further includes a multiplier that multiplies the frequency domain representation of the negative symbol by the inverted channel estimate to generate one of the data symbols. The channel adjusted Frequency domain notation. 17. The receiver according to item 15 of the scope of patent application, further comprising: a multiplier, which multiplies the frequency domain representation of the data symbol by the inverted channel estimation to generate the channel-adjusted frequency of the data symbol Domain representation; and a second scaling circuit that scales the channel-adjusted frequency domain representation of the data symbol based on the next largest coefficient in the frequency domain representation adjusted by the channel of the data symbol. 8. The receiver as claimed in item 16 of the patent application, wherein the zoom circuit is 1225 6. The patent application scope scales the frequency domain representation of the data symbol to generate a scaled frequency domain representation of the data symbol, which further includes a multiplier that multiplies the scaled frequency domain representation of = material by Inverted channel estimates. H-type receiver 'includes an equalization circuit having a scaling circuit for scaling the representation of a signal received from the transmitter. The receiver includes: Λ a frequency domain conversion unit which is to be generated at the receiver Received on A frequency domain representation of at least one adjustment symbol and a frequency domain representation of a data symbol; a scaling circuit that scales at least one adjustment symbol based on a maximum coefficient in the frequency domain representation of at least one adjustment symbol Frequency domain representation to produce a scaled frequency domain representation of at least one adjustment symbol; A divider circuit that divides the scaled representation of at least one adjustment symbol in the frequency domain representation by the frequency domain representation of at least one adjustment symbol based on the time domain representation of the adjustment symbol that is substantially unaffected by the channel effect. Method to generate a channel estimate; an inversion circuit that generates an inverted channel estimate by inverting the channel estimate; and a multiplier that multiplies the frequency domain representation of the data symbol by the inverse Turned channel estimation. 20. The receiver of claim 19, wherein the scaling circuit is to scale based on the maximum coefficient in the frequency domain when the maximum coefficient in the frequency domain representation of at least one adjustment symbol is less than a critical value. 40 VI. Declaring the scope of patents The frequency domain representation of at least one adjustment symbol. 21. The receiver as claimed in claim 20, further comprising a frequency offset compensation circuit, which is a scaled frequency domain representation of at least one adjustment symbol and a frequency domain representation of a periodic signal Convolution is performed to produce an offset-adjusted frequency domain representation of at least one adjustment symbol. 22. The receiver as claimed in claim 21, wherein the scaled frequency domain representation of the at least one adjustment symbol used by the divider to generate the channel estimate is an offset of the at least one adjustment symbol generated by the frequency offset compensation circuit. Adjusted scaled frequency domain representation. 23. A receiver comprising an equalizing circuit having a scaling circuit for scaling the representation of a signal received from a transmitter, the receiver comprising: a frequency domain conversion unit which is to be generated at the receiver The frequency domain representation of the received symbol to the dither symbol and the frequency domain representation of a data symbol are incorrect; a scaling circuit is to scale at least based on the largest coefficient in the frequency domain representation of at least one adjustment symbol A frequency-domain representation of an adjusted symbol to generate a scaled frequency-domain representation of at least one adjusted symbol; a divider circuit that divides a scaled frequency-domain representation of at least one adjusted symbol by The time domain representation of the adjustment symbol affected by the channel effect. At least one frequency domain representation of the adjustment symbol. It is to generate 41 1225¾ ^-1¾ ^ by inverse channel estimation. 6. Inverted channel estimation for patent application scope; a multiplier 'which multiplies the frequency domain representation of the data symbol by the inverted channel estimation to produce the channel-adjusted frequency domain representation of the data symbol; and The first scaling circuit is to scale the channel-adjusted frequency domain representation of the data symbol based on the second largest coefficient in the frequency-domain table adjustment of the channel of the data symbol. 24. The receiver as claimed in claim 23, wherein the scaling circuit is to scale at least one adjustment based on the maximum coefficient in the frequency domain when the maximum coefficient in the frequency domain representation of the adjustment symbol is less than a threshold. Frequency domain representation of symbols. 25. The receiver according to item 23 of the patent application, further comprising a frequency offset compensation circuit, which uses a frequency domain representation of a periodic signal by scaling the frequency domain representation of at least one adjustment symbol Make rolls Product to produce an offset adjustment of at least one adjustment symbol to adjust the scaled frequency domain representation. 26. The receiver as claimed in claim 25, wherein the scaled frequency domain representation of the at least one adjustment symbol used by the divider to generate the channel estimate is an offset of at least one adjustment symbol generated by the frequency offset compensation circuit. Adjust the scaled frequency domain representation. 27. A method of scaling a representation of an adjustment signal received by a transmitter on a receiver, the method comprising: generating a frequency domain representation of at least one adjustment symbol received on the receiver and a Frequency domain representation of data symbols; 42 VI. The scope of the patent application is based on scaling the frequency domain representation of at least one adjustment symbol based on the largest coefficient in the frequency domain representation of at least one adjustment symbol that is less than a critical value to produce a scaled frequency domain of at least one adjustment symbol. Notation; and by generating a frequency domain representation of at least one adjustment symbol by dividing the scaled frequency domain representation of at least one adjustment symbol by the frequency domain representation of at least one adjustment symbol based on the time domain representation of the adjustment symbol that is substantially unaffected by the channel effect estimate. 28. The method of claim 27, further comprising generating at least one adjustment symbol offset by convolving the scaled frequency domain representation of at least one adjustment symbol with the frequency domain representation of a periodic signal. Adjust the scaled frequency domain representation. 29. The method of claim 28, wherein the scaled frequency domain representation of the at least one adjusted symbol used to generate the channel estimate is a scaled frequency domain representation of the offset adjusted at least one adjustment symbol. 30. The method of claim 29, further comprising generating an inverted channel estimate by inverting the channel estimate. 31_ The method of claim 29, further comprising: filtering the channel estimation; and generating an inverted channel estimation by inverting the channel estimation. 32. The method of claim 31, further comprising multiplying the frequency domain representation of the data symbol by the inverted channel estimate. 33. The method according to item 31 of the scope of patent application, further comprising: multiplying the frequency domain representation of the data symbol by the inverted channel estimation. 6. Adjusting the frequency domain representation of the patent application scope to generate the channel of the data symbol. And when the second largest coefficient in the adjusted frequency domain representation of the channel of the data symbol is less than a critical value, the adjusted frequency domain representation of the channel of the data symbol is scaled. 34. The method of claim 29, further comprising: scaling the frequency domain representation of the data symbol to produce a scaled frequency domain representation of the data symbol; and multiplying the scaled frequency domain representation of the data symbol Estimated with inverted channels. 35. The method of claim 29, wherein generating an offset-adjusted frequency domain representation of at least one of the adjustment symbols includes generating a frequency having a frequency equal to the frequency offset between the receiver and the transmitter. Frequency domain representation of the periodic signal, and convolution of the frequency domain representation of the periodic signal and the frequency domain representation of the at least one adjustment symbol to produce an offset-detected frequency domain representation of the at least one adjustment symbol. The effect of frequency offset between the receiver and the receiver is essentially removed. 36. The method of claim 29, wherein the scaled frequency domain representation of at least one adjusted symbol includes at least one scaled frequency domain representation of the adjusted symbol and a scaled frequency of the second adjusted symbol. Domain representation method, the method further comprising: generating a scaled frequency domain representation of the first adjustment symbol and a second adjustment symbol; and a patent-applied frequency domain representation in which offset adjustment scaling of at least one adjustment symbol is generated Including shifting: ί—frequency domain representation of a periodic signal with a frequency equal to the frequency between the receiver and the transmitter is too much; and 1: combining the frequency domain representation of the periodic signal with the first- and second-adjustment symbols The average of the scaled frequency domain representation is convolved with an offset-compensated average of the scaled frequency domain representation of the first dimmed symbol and the second adjusted symbol, which makes the frequency offset between the receiver and the transmitter Removal was removed. For example, the method of claim 36, wherein the channel estimation package is generated by dividing the offset compensation of the scaled symbol and the second adjusted symbol in the frequency domain representation by at least one adjustment that is substantially unaffected by the channel effect. Frequency domain representation of symbols. As in the method of claim 37, the method further includes generating an inverted channel estimate by inverting the channel estimate. For example, the method of claim 36 of the patent scope further includes: filtering the channel estimation; and generating an inverted channel estimation by inverting the channel estimation. For example, the method of claim 39 of the patent scope further includes multiplying the frequency domain representation of the data symbol by the inverted channel estimation to generate the channel-adjusted frequency domain frequency representation of the data symbol. For example, the method in the 39th scope of the patent application further includes: multiplying the frequency domain representation of the material symbol by the inverse channel estimate to produce ten channel-adjusted frequency domain representation of the material symbol. And when the channel range of the data symbol is adjusted in the frequency domain representation, the second range of the patented range is called the coefficient of the second largest coefficient is smaller than a critical value. For example, the method of claim 36 of the patent application scope further includes: scaling the frequency domain representation of the material symbol to generate a scaled frequency domain representation of the data symbol; and scaling the frequency domain representation of the negative material symbol. The representation is multiplied by the inverted channel estimate. The method of claim 36, wherein generating an offset of at least one adjustment symbol adjusts the scaled frequency domain representation including generating a frequency of a periodic signal having a frequency equal to the frequency offset between the receiver and the transmitter Domain representation, and convolution of the frequency domain representation of the periodic signal and the frequency domain representation of at least one adjustment symbol to produce an offset-compensated frequency domain representation of at least one adjustment symbol, which enables the receiver and the transmitter The effect of frequency offset between is substantially removed. A method for scaling a representation of an adjustment signal received from a transmitter on a receiver, the method comprising: generating a frequency domain representation of at least one adjustment symbol received on the receiver and a frequency of a data symbol Domain representation; scaling the frequency domain representation of at least one adjustment symbol to produce a scaled frequency domain representation of at least one adjustment symbol based on the largest coefficient in the frequency domain representation of the at least one adjustment symbol; The scaled frequency domain representation of at least one adjustment symbol divided by-is based on the adjustment symbol that is substantially unaffected by the channel effect f. 6. At least the patent application scope time domain representation-the frequency domain representation of the adjustment symbol;… Inverted channel estimates are generated by inverting the channel estimates, and the frequency domain representation of the data symbols is multiplied by the inverted channel estimates. 45. 46. 47. If the method of applying for the scope of the patent is applied, the step further includes generating at least __adjustment symbols by convolving the _-table reading of the ^ -adjustment symbol and the frequency-domain table of the periodic signal. Of—The frequency domain representation of offset adjustment scaling. For example, the method in the 45th aspect of the patent application, wherein the scaled frequency domain representation of at least _adjusted symbols is at least-the offset of adjusted symbols adjusts the scaled frequency domain representation. A method for scaling a representation of an adjustment signal received from a transmitter on a receiver, the method comprising: generating a frequency domain representation of at least one adjustment symbol received on the receiver and a frequency domain representation of a data symbol Method; scaling the frequency domain representation of at least one adjustment symbol to produce a scaled frequency domain representation of at least one adjustment symbol based on the largest coefficient in the frequency domain representation of at least one adjustment symbol; The scaled frequency domain representation of the adjustment symbol is divided by a frequency domain representation of at least one adjustment symbol based on the time domain representation of the adjustment symbol that is substantially unaffected by the channel effect; an inverse is generated by inverting the channel estimation Passed channel estimates; multiply the frequency domain representation of the data symbols by the inverted channel estimates 6. The scope of the patent application is calculated by adjusting the frequency domain representation of the channel generating the data symbol; and when the second-largest coefficient in the frequency domain representation of the channel adjusting data symbol is less than a critical value, the channel of the scaled data symbol is adjusted In the frequency domain. 48. The method according to item 47 of the patent application scope, further comprising generating at least one adjustment symbol by adjusting the frequency domain representation of at least one adjustment symbol and the frequency domain representation of a periodic signal. In the frequency domain. 49. 10 50. The method of claim 48, wherein the scaled frequency domain representation of at least one adjusted symbol used to generate a channel estimate is an offset of at least one adjusted symbol. The scaled frequency domain representation is adjusted. . A receiver comprising an equalization circuit having a scaling circuit for scaling a representation of a signal received from a transmitter. The receiver includes: a frequency domain conversion unit which is to generate The frequency domain representation of at least one adjustment symbol and the frequency domain representation of a data symbol received on the device; a scaling circuit that is based on the largest coefficient in the frequency domain representation of at least one adjustment symbol Frequency domain representation of at least one adjustment symbol to generate scaled frequency domain representation of at least one adjustment symbol; an offset compensation circuit that scales the frequency domain representation of at least one adjustment symbol and a periodic signal Frequency domain representation to do convolution to generate at least one adjustment symbol offset adjustment scaled frequency domain 48 six 51. 52. 53. 54. 55. Patent application range representation; and ▲-division circuit, which is To divide at least the adjustment symbol's offset adjustment scaled frequency domain representation by the frequency domain of the adjustment symbol based on the time domain table of the adjustment symbol that is essentially unaffected by the channel effect * Notation. For example, the receiver of claim 50, wherein the scaled frequency domain representation of the at least one adjusted symbol used to generate the channel estimate is an offset adjusted scaled frequency domain representation of the evening adjustment symbol. For example, the converter of No. 51 in the scope of patent application further includes an inversion circuit, which is to generate an inverted channel estimation by inverting the channel estimation. For example, the receiver of the scope of patent application No. 51 further includes a smoothing circuit that filters the channel estimation; and an inversion circuit that generates an inverted channel by inverting the channel estimation. estimate. For example, the receiver of the 53rd patent application scope further includes a multiplication method, which multiplies the frequency domain representation of the data symbol by the inverted channel estimate. For example, the receiver of the 53rd scope of the patent application further includes a multiplier, which multiplies the frequency domain representation of the data symbols by the inverted channel estimation to generate the channel-adjusted frequency domain representation of the data symbols. And a second scaling circuit, which is to scale the meal when the second largest coefficient in the frequency domain representation adjusted by the channel of the data symbol is less than a critical value ... Adjusted frequency domain representation of symbolic channels. 56. For example, the receiver of claim 51, wherein the scaling circuit is to scale the frequency domain representation of the data symbol to generate a scaled frequency domain representation of the data symbol, and further includes a multiplier, which is to Multiply the selected frequency domain representation of the symbol by the inverted channel estimate. 57. A receiver comprising an equalizing circuit having a scaling circuit for scaling the representation of a signal received from a transmitter, the receiver comprising: a frequency domain conversion unit which is to be generated at the receiver The frequency domain representation of at least one adjustment symbol and the frequency domain representation of a data symbol received on the above;-a scaling circuit that is based on at least the maximum coefficient in the frequency domain representation of the adjustment symbol to scale at least -Adjust the frequency domain representation of symbols to produce at least _ adjusted symbols in the scaled frequency domain representation;-Offset compensation circuit 'Qian at least-Adjusted symbols in the scaled frequency domain representation and-the frequency domain of the periodic signal The notation is convolved to generate at least-adjusted symbol offset adjustment scaled frequency domain table and a divider circuit, which is gland $, respectfully adjusts the scaled frequency of the offset of at least one adjustment symbol Domain representation divided by 乂 Mo Friends divided by a frequency domain representation of adjustment symbols that are not valid for a single subdomain when adjustment symbols are based on the fact that they are not substantially affected by the channel effect; 夂The scope of the patent application is an inversion circuit, which is to generate an inverted channel estimate by inverting the channel estimation, and a multiplier, which is to multiply the frequency domain representation of the data symbol by the inverted Channel estimation. 58 · A receiver comprising an equalizing circuit having a scaling circuit for scaling a representation of a signal received from a transmitter, the receiver comprising: 频 a frequency domain conversion unit which is to be generated in Frequency domain representation of at least one adjustment symbol and frequency domain representation of a data symbol received on the receiver; a scaling circuit based on the maximum coefficient in the frequency domain representation of at least one adjustment symbol Scaling the frequency-domain representation of at least one adjustment symbol to produce a scaled frequency-domain representation of at least one adjustment symbol; an offset compensation circuit that involves at least-scaling the frequency-domain representation of the adjustment symbol and a period The frequency domain representation of the signal is convolved to generate at least one offset adjustment symbol. The scaled frequency domain representation is not correct, and a division H circuit is expected to adjust the scaled frequency domain representation by at least _adjustment offset. Divide by a frequency-domain representation of the adjustment symbol based on the time-domain representation of the adjustment symbol that is essentially unaffected by the channel effect; an inversion circuit Estimation to generate an inverted channel estimate; a multiplier for force and patent application range, which multiplies the frequency domain representation of the data symbol by the inverted channel estimate; and a second scaling circuit, which When the next largest coefficient in the adjusted frequency domain representation of the channel of the data symbol is less than a critical value, the adjusted frequency domain representation of the channel of the data symbol is scaled. 59. 60. A method of scaling a representation of an adjustment signal received from a transmitter on a receiver, the method comprising: generating a frequency domain representation of at least one adjustment symbol received on the receiver and a Frequency domain representation of data symbols; scaling the frequency domain representation of at least one adjustment symbol to produce a scaled frequency domain representation of at least one adjustment symbol when the maximum coefficient in the frequency domain representation of at least one adjustment symbol is less than a critical value Method; by convolving the scaled frequency domain representation of at least one adjusted symbol and the frequency domain representation of the periodic flood number to generate an offset adjusted scaled frequency domain representation of at least one adjusted symbol; and, A channel estimate is generated by dividing the scaled frequency domain representation of the offset adjustment of the at least one adjustment symbol by the frequency domain representation of the at least one adjustment symbol based on the time domain representation of the adjustment symbol that is substantially unaffected by the channel effect. For example, the method in the 59th scope of the patent application further includes inverting the channel estimation by inverting the channel estimation. The method of claim 59 further includes: filtering the channel estimation; and generating an inverted channel estimation by inverting the channel estimation. 61. VI. Patent Application Range 62. The method of item 61 of the patent application range further includes multiplying the frequency domain representation of the data symbol by the inverted channel estimate. 63. The method of claim 61, further comprising: multiplying the frequency domain representation of the data symbol by an inverted channel estimate to produce a channel-adjusted frequency domain representation of the data symbol; and when the data symbol is The second largest in the channel-adjusted frequency domain representation is the channel-adjusted frequency domain representation of the scaled data symbol when it is less than a critical value. 64_ The method of claim 59, further comprising: scaling the frequency domain representation of the data symbol to produce a scaled frequency domain representation of the data symbol; and multiplying the scaled frequency domain representation of the negative symbol Estimated with inverted channels. 65. The method of claim 59, wherein generating an adjusted adjusted frequency-domain representation of at least one adjustment symbol includes generating a period having a frequency equal to the frequency offset between the receiver and the transmitter The frequency domain representation of the signal and the frequency domain representation of the periodic signal and the frequency domain representation of the at least one adjustment symbol are convolved to produce an offset-compensated frequency domain representation of the at least one adjustment symbol, which enables the receiver The effect of frequency offset between the transmitter and the transmitter is substantially removed. 66. The method of claim 59, wherein the scaled frequency domain representation of at least one adjusted symbol includes a scaled frequency domain representation of a first adjusted symbol and a scaled frequency of a second adjusted symbol The method of applying for a patented domain representation further includes: generating an average of the scaled frequency domain representation of the first adjustment symbol and the second adjustment symbol; and generating an offset adjusted scaled frequency domain representation of at least one adjustment symbol. The method includes: generating a frequency domain representation of a periodic signal having a frequency equal to the frequency offset between the receiver and the transmitter; and scaling the frequency domain representation of the periodic signal with the first adjustment symbol and the first adjustment symbol. The average of the frequency domain representation is convolved to produce an offset compensated average of the scaled frequency domain representation of the first adjustment symbol and the second adjustment symbol, which makes the frequency offset between the receiver and the transmitter The effect is essentially removed. For example, the method of claim 65, wherein generating the channel estimation includes dividing the scaled frequency domain table offset offset average of the first adjustment symbol and the second adjustment symbol by the difference based on the substantial unaffected channel effect. At least one adjustment symbol in the time domain representation. The method according to item 67 of the patent application scope further includes generating an inverted channel estimate by using the channel estimate. The method of claim 64, further comprising: filtering the channel estimation; and generating an inverted channel estimation by using the channel estimation. For example, the method in the 68th scope of the patent application, the further step includes finely reversing the frequency domain representation of the data symbol break to estimate the channel to generate data. 6. The frequency domain representation of the channel adjustment symbol for the patent scope. 71. A method for scaling the representation of the adjustment signal received by the -receiver on the -receiver, the method comprising: generating a frequency domain representation of at least the _adjustment symbol received on the receiver and a Frequency-domain representation of data symbols; scaling at least when the maximum coefficient in at least-adjusted-frequency-domain representation is less than -critical value-adjusted-frequency-domain representation to produce at least one adjusted-frequency-domain representation of the adjusted symbol Adjusting the scaled frequency domain representation by convolving at least the frequency domain representation of the scaling symbol of the adjustment symbol with the frequency domain representation of a periodic signal to generate at least one adjustment symbol offset; and Divide the offset adjustment scaled frequency domain representation of at least one adjustment symbol by the frequency domain representation of at least one adjustment symbol based on the time domain representation of the adjustment symbol that is substantially unaffected by the channel effect to generate a channel estimate; by Inverted channel estimates are generated from the inverted channel estimates; and the frequency domain representation of the data symbols is multiplied by the inverted channel estimates. 72. —A method of scaling a representation of an adjustment signal received by a transmitter on a receiver, the method comprising: generating a frequency domain representation of at least one adjustment symbol received on the receiver and a data Frequency domain representation of symbols; maximum factor in at least one adjusted frequency domain representation of symbols 1225335 6. Scale the frequency domain representation of at least one adjustment symbol to generate a scaled frequency domain representation of at least one adjustment symbol when the scope of the patent application is less than a critical value; by scaling the frequency domain representation of at least one adjustment symbol Convolving with a frequency domain representation of a periodic signal to generate an offset adjustment scaled frequency domain representation of at least one adjustment symbol; Generate a channel estimate by dividing the offset adjusted scaled frequency domain representation of at least one adjustment symbol by the frequency domain representation of at least one adjustment symbol based on the time domain representation of the adjustment symbol that is substantially unaffected by the channel effect. Generating an inverted channel estimate by inverting the channel estimate; and multiplying the frequency domain representation of the data symbol by the inverted channel estimate to produce a channel-adjusted frequency domain representation of the data symbol; and The channel-adjusted frequency domain representation of the data symbol is scaled when the next largest coefficient in the channel-adjusted frequency domain representation of the data symbol is less than a critical value. η 〇 • A method of correcting the effect of a frequency offset between a receiver and a transmitter by giving a value of an adjustment symbol received during a preamble, the method includes: A first vector, the first vector angle of which indicates a fine offset between the receiver and the transmitter; generating a fine offset estimate based on the first vector angle; and Received 56 Positive Replacement Page ^ .93. 5.27 — Den. 6. Multiply the sign of patent application range by-with frequency signal based on fine offset estimation. The method according to item 73 of the scope of patent application further includes: generating on the basis of the vector angle—a frequency domain representation of a second signal having a first frequency substantially equal to the first frequency of the fine and shifted offsets; The frequency domain representation of the number and the frequency domain representation of 'at least-adjusted symbols received on the receiver are convolved to produce an offset-compensated frequency domain representation of at least one adjusted symbol. The channel transfer function should be substantially removed from the transmitter. 75. The method of claim 74 in the scope of patent application, which generates a frequency domain representation of at least one adjustment symbol received on the receiver based on the received J-decay symbol received on the receiver. 76. If the scope of the patent application is 75, the method further includes the method of dividing the offset of at least one adjustment symbol received on the receiver by the frequency domain representation divided by the actual representation. Frequency domain representation of at least one-adjusted symbol transmitted. 77. The method of claim 75, wherein at least the frequency domain representation of the adjustment symbol includes the frequency domain representation of the first adjustment symbol and the frequency domain representation of the second adjustment symbol. The method further includes: By averaging the frequency domain representation of the first adjustment symbol and the frequency domain representation of the second adjustment = to generate an average of the frequency domain representation of at least one adjustment symbol; the scope of the patent application is by averaging the frequency domain of at least one adjustment symbol The average of the representation and the frequency-domain representation of the third signal are convolved to produce an average offset-compensated frequency-domain representation; the average-offset-compensated frequency-domain representation is divided by the actual instruction to send The frequency domain representation of the at least one withered symbol sent by the transmitter to generate a channel estimate. For example, the method of claim 77 of the patent application scope further includes: generating a frequency domain representation of at least one adjustment symbol that substantially represents at least one adjustment symbol sent by the transmitter. The method of claim 76, further comprising: dividing the offset-compensated frequency domain representation of the data symbol by the channel estimate to generate at least one of the pilot carrier and at least one of the adjustment symbols in the data symbol A representation of a phase difference between pilot carriers; and generating an average of the phase difference between at least one pilot carrier in a data symbol and at least one pilot carrier in at least one adjustment symbol based on the representation of the phase difference And based on the average of the phase differences to generate an updated frequency offset for adding to the first signal generator. For example, the method of claim 76 of the patent application scope further includes: generating an offset-compensated frequency domain representation of a first data symbol and an offset-compensated frequency domain representation of a second data symbol; The offset-compensated frequency-domain representation of the symbol is divided by the offset-compensated frequency-domain representation of the second data symbol to generate at least one of the reference carrier in the first data symbol and the claimed patent range in the second data symbol. To J-5, the representation of the phase difference between the carriers is shown; and the expression of the phase difference is based on the generation of the phase difference between at least one carrier in the data symbol and at least-indicator carrier in the adjustment symbol. The average is generated based on the average of the phase difference, and the updated frequency offset is added to the first signal generator. For example, the method in the 73rd aspect of the patent application, further comprising: generating a first time-domain periodic signal based on the first vector angle; and multiplying the time-domain representation of at least one adjustment symbol by the first time-domain periodic signal To produce at least—adjust the symbol's offset-compensated time-domain representation. The method of claim 81, further comprising: converting an offset-compensated time-domain representation of at least one adjusted symbol to produce an offset-compensated frequency-domain representation of at least one adjusted symbol received on a receiver Method; and by dividing the compensated frequency domain representation of at least one adjustment symbol received on the receiver by the frequency domain of at least one adjustment symbol that substantially represents at least one adjustment symbol transmitted on the transmitter Representation to generate channel estimates. For example, the method of claim 82 further includes: providing a frequency domain representation of at least one adjustment symbol that substantially represents at least one adjustment symbol sent by the transmitter. For example, the method of claiming the 82nd patent range further includes: dividing the offset-compensated frequency-domain representation of a symbol with a force, applying patent channel estimates to generate at least one of the reference carrier and the Representation of phase difference between at least one pilot carrier among at least one adjustment symbol received at the receiver; generating at least one pilot carrier among data symbols based on the representation of the phase difference and receiving at the receiver At least one of the at least one adjusted symbol indicates the average of the phase difference between the carriers, and generates an updated frequency offset for adding to the first signal generator based on the average of the phase difference. 85. The method of claim 82, further comprising: providing an offset-compensated frequency domain representation of a first data symbol and an offset-compensated frequency domain representation of a second data symbol; The offset-compensated frequency domain representation of the first data symbol is divided by the offset-compensated frequency domain representation of the second data symbol to generate at least one pilot carrier in the first data symbol and one of the second data symbol. A representation of phase difference between at least one pilot carrier; and based on the representation of phase difference to generate at least one pilot carrier in a first data symbol and at least one pilot carrier in a second data symbol An average of the phase differences is used to generate an updated frequency offset based on the average of the phase differences. 86. The method of claim 73, further comprising: receiving a sample of the short adjustment symbol; generating a filtered sample of the short adjustment symbol based on the received sample of the short adjustment symbol; and The filtered samples of the adjustment symbols are subtracted from 122533 ^ of the short adjustment symbols. Sixth, the patent application scope samples are subtracted to generate the short adjustment symbol samples that make the DC substantially removed. 87. The method of claim 73, further comprising: generating a second vector based on the set sampling of at least one short adjustment symbol that is not compensated for the frequency offset, and that the frequency offset is not compensated for Generate a third vector based on the set sampling of at least one short adjustment symbol; calculate the DC offset (xl-x2) e〇a) —N (l-eUa)) by calculating the value of 15 Where α is the first vector angle, χ1 is the second vector, χ2 is the third direction 篁 ′, and N is the number of samples in at least one short adjustment symbol that is not compensated for the frequency offset; and generated after the short adjustment symbol Difference between samples of received symbols and DC offset. ❿ 88. The method of claim 73, further comprising: generating a second vector based on a set of at least one short adjustment symbol that is not compensated for the frequency offset, and not compensated for the frequency offset In addition, at least-a set of short adjustment symbols is used as a basis to generate a third vector; the value of the following formula is used to calculate the DC offset (xl-x2) ~ 2N ~ where xl is the second vector and χ2 is Third vector, and N is unpaired 61 20 ^ 225335 6. Scope of patent application Number of samples in at least one short adjustment symbol whose frequency offset is compensated; and 10 15 20 Generates the difference between the samples received after the short adjustment symbol and the DC offset. An automatic frequency control circuit that corrects the effect of a frequency offset between a receiver and a transmitter by evaluating short symbols and adjustment symbols received during a preamble. The circuit includes: an autocorrelator that A first vector is generated based on the short adjustment symbol, and the first vector angle indicates a coarse offset between the receiver and the transmitter, and the autocorrelator is based on at least one long adjustment symbol To generate a second vector, where the second vector angle represents the fine offset between the receiver and the transmitter; a frequency offset generator, which generates a coarse offset estimate based on the first vector angle And generating a fine offset estimate based on the second vector angle; a first signal generator to generate a first signal having a first frequency substantially equal to the coarse offset estimate based on the coarse offset estimate; A periodic signal; a first mixer that generates a product of at least one long adjustment symbol received at the receiver and the first periodic signal and adds the product to the autocorrelator, where the autocorrelator is a base product A second vector is generated based on this; where after the frequency offset generator has generated a fine offset estimate, the first signal generator uses a coarse offset estimate and a fine offset estimate. 62 VI. Scope of patent application What is the basis for generating a second period signal with a second frequency; and where * Mix | § is to multiply the symbol received after receiving at least ^ length adjustment symbol by the second period signal. 90. The circuit of item 89 in the scope of patent application, further comprising: Offset compensation frame, which is to generate a third signal with a second vector angle in the frequency domain with a third signal that is substantially specialized in fine offset estimation. Representation; and a convolver that convolves the frequency domain representation of the third signal with the frequency domain representation of at least one long adjustment symbol received on the receiver to produce at least one long adjustment symbol. An offset-compensated frequency-domain representation that allows the frequency offset effect between the receiver and the transmitter to be substantially removed. 91. The circuit according to item 90 of the scope of patent application, further comprising a frequency domain conversion unit, which is to generate at least one received on the receiver based on at least one long adjustment symbol received on the receiver. Frequency domain representation of long adjustment symbols. 92. The circuit of item 91 in the scope of patent application, further comprising a divider circuit, which is divided by the frequency-domain representation of the offset compensated by at least one long adjustment symbol received on the receiver by the substance. A frequency domain representation of at least one long adjustment symbol representing at least one long adjustment symbol sent by the transmitter to generate a channel estimate. 93. If the circuit of claim 91 is applied for, the frequency domain representation of at least one long adjustment symbol includes the frequency domain representation of the first long adjustment symbol and the frequency domain representation of the first long 5-week integer symbol. The circuit further on page 1225335 j 角 a____ /, Shen Jing patent scope includes:, averaging circuit, which is obtained by averaging the frequency domain representation of the first long adjustment symbol and the frequency domain representation of the second long adjustment symbol Generate an average of the frequency domain representation of at least one long adjustment symbol; wherein the convolutional device generates the average by convolving the frequency domain representation of at least one long adjustment symbol and the frequency domain representation of the third signal Offset-compensated frequency-domain representation; and a divider circuit that divides the average offset-compensated frequency-domain representation by at least one of the at least one long adjustment symbol transmitted by the transmitter. The frequency domain representation of a long adjustment symbol produces a channel estimate. 94. The circuit according to item 93 of the scope of patent application, further comprising: a frequency domain representation unit for a long adjustment symbol, which is to generate a frequency domain representation of at least one long adjustment symbol that substantially represents at least one long adjustment symbol sent by the transmitter. law. 95. The circuit of item 92 of the scope of patent application, further comprising: a second divider circuit, which divides the offset-compensated frequency-domain representation of a data symbol by the channel estimate to generate the data symbol. Representation of phase difference between at least one pilot carrier and at least one pilot carrier among at least one long adjustment symbol; and an average phase shift circuit that generates data symbols based on the representation of the phase difference The average of the phase difference between at least one of the pilot carriers and at least one of the pilot carriers in at least one long adjustment symbol is generated on the basis of the average of the phase differences. Update frequency offset on the controller. For example, the circuit of claim 92 of the patent application scope further includes: a first divider circuit, which receives the offset-compensated frequency domain representation of the first and second shell symbols from the convolver and a second data The offset-compensated frequency-domain representation of the 苻 symbol is the offset-compensated frequency-domain representation of the first data symbol divided by the offset-compensated frequency-domain representation of the second data symbol. Generate a representation of the phase difference between at least one pilot carrier in the data symbol and at least one pilot carrier in the at least one long adjustment symbol; and an average phase shift circuit, which uses the representation of the phase difference as Based on the average of the phase difference between at least one pilot carrier in the data symbol and at least one pilot carrier in the at least one long adjustment symbol, and based on the average of the phase difference to generate a Update frequency offset on the signal generator. For example, the circuit in the 89th scope of the patent application further includes: a second signal generator that generates a third period signal based on the second vector angle; and a second mixer that requires at least The time domain representation of a long adjustment symbol cannot be multiplied by the third period signal to produce an offset-compensated time domain representation of at least one long adjustment symbol. For example, the circuit of the 97th patent application scope further includes a first frequency domain conversion unit, which is to convert at least one long adjusted symbol's offset-compensated time domain representation to generate the received signal on the receiver. At least-the offset-compensated frequency domain of the long adjustment symbol 122533 酴 I is replacing the page and cw cr 2 7 VI, "_" notation; and a divider circuit, which is to be used by the receiver The received offset-compensated frequency domain representation of the at least one long adjustment symbol is divided by the frequency domain representation of the at least one long adjustment symbol that substantially represents the at least one long adjustment symbol sent by the transmitter. 99. The circuit according to item 98 of the scope of patent application, further comprising: ❿ a long adjustment symbol frequency domain representation unit, which provides at least one long dimming symbol in the frequency domain representation, which essentially represents at least One long adjustment symbol. 100. The circuit of item 98 of the patent application scope further includes a second divider circuit which divides the frequency-domain representation of the offset of a data symbol by the channel estimate to generate the data symbol. Representation of the phase difference between at least one pilot carrier and at least one pilot carrier in at least one long adjustment symbol received at the receiver; an average phase shift circuit, which is to be expressed in terms of phase difference The method is based on generating an average of the phase difference between at least one pilot carrier in the data symbol and at least one non-carrier carrier in the at least one long adjustment symbol received at the receiver, and the average of the phase differences is The basis is to generate an update frequency offset for adding to the first signal generator. 101. The circuit according to item 98 of the scope of patent application, further comprising: a second divider circuit, which is to receive an offset-compensated frequency-domain representation of a first data symbol from a convolver and a second resource 66 I22533f | is replacing " i " Menopause, 27th "__ VI. The frequency domain representation of the offset compensation material symbol in the patent application range, and the frequency after which the offset of the first data symbol is to be compensated Domain representation divided by offset offset compensation by the second data symbol to generate a phase difference between at least one pilot carrier in the first data symbol and at least one pilot carrier in the second data symbol Representation; an average phase shift circuit based on the representation of the phase difference to generate at least one pilot carrier in the first data symbol and at least one pilot carrier in the second data symbol Based on the average of the phase difference, and based on the average of the phase difference to generate an update frequency offset for adding to the first signal generator. 102. The circuit according to item 99 of the scope of patent application, further comprising: a low-pass filter that receives samples of short adjustment symbols and generates filtered samples of short adjustment symbols, and an adder that The short-adjusted symbol samples are generated by subtracting the filtered samples of the short-adjusted symbols from the short-adjusted samples, which have substantially removed the DC to add to the autocorrelator. The circuit such as the 99th scope of the patent application, further includes a first adder 'to generate a third vector based on a set of at least-short adjustment symbols that are not compensated for the frequency offset, and based on The fourth vector is generated based on the sampling of the set of other at least one short adjustment symbol that has not been compensated for the frequency offset. Theft-DC offset compensator is to receive the first vector and the third port. Calculate the DC bias by calculating the following formula: 122533 6. Range of patent application (xi-x2) eUa) ~ N (l-eUa) Y where ct is the first vector angle, χ1 is the third vector, and χ2 is the first Four vectors, and N is the number of samples in at least one short adjustment symbol that is not compensated for frequency offset; and a second adder is to generate the difference between the samples of the symbol received after the short adjustment symbol and the DC Offset. 104. The circuit of item 99 in the scope of patent application, further comprising: a first adder that generates a third vector based on a set of at least one short adjustment symbol that is not compensated for frequency offset And a fourth vector based on a set of samples of at least one short adjustment symbol that is not compensated for frequency offset; a DC offset compensator that receives the first vector, the third vector, and the fourth The vector is calculated by calculating the DC offset (χΐ — χ2) ~ 2N ~ where xl is the third vector, χ2 is the fourth vector, and Ν is at least one without compensation for the frequency offset. The number of samples in the short adjustment symbol; and a second adder, which is to generate a difference and a DC offset between the samples of the symbol received after the short adjustment symbol. 105 · — An automatic frequency control circuit for correcting a receiver and a transmission by finding the value of an adjustment symbol received during a preamble 68 12253 6. The effect of the frequency offset between the patent-applied rangers. The circuit includes ... an autocorrelator that generates a first vector based on the short adjustment symbol. 'The first vector angle represents the reception The coarse offset between the transmitter and the transmitter, and the autocorrelator is based on at least one long adjustment symbol to generate a second vector. The second vector angle represents the fineness between the receiver and the transmitter. Offset A frequency offset generator that generates a coarse offset estimate based on the first vector angle and a fine offset estimate based on the second vector angle; a first signal generator that uses the The coarse offset estimation is used as a basis to generate a first periodic signal having a first frequency substantially equal to the first frequency of the coarse offset estimation. A first mixer is to generate at least one long adjustment symbol and the first received at the receiver. Product of periodic signals and add the products To the autocorrelator, where the autocorrelator generates a second vector based on the base product; where 'after the frequency offset generator has generated a fine offset estimate', the first signal generator uses a coarse offset estimate and The fine transmission estimation is based on generating a second periodic signal with a second frequency; and wherein the first mixer is to multiply a symbol received after receiving at least one long adjustment symbol by a second periodic signal; an offset compensation It is based on the second vector angle to generate a frequency-domain representation of a third signal with a third frequency that is specialized in fine offset estimation; and 69. Patent Application Scope-Convolution To convolve the frequency domain representation of the third signal with the frequency domain representation of at least one long adjustment symbol received on the receiver to produce an offset-compensated frequency domain representation of at least one long adjustment symbol, It allows the frequency offset effect on the channel transfer function between the receiver and the transmitter to be substantially removed. 106. The circuit of item 105 in the scope of patent application, further including a divider circuit, which is to be expressed in the frequency domain by offset compensation of at least one long adjustment sign received on the receiver. To generate a channel estimate by dividing by the frequency domain representation of at least one long adjustment symbol that is essentially at least one long adjustment symbol sent by the transmitter. ^ 107. The circuit of claim 106 in the scope of patent application, further comprising: a second divider circuit, which divides the offset of a data symbol by 4 members in the frequency domain representation by channel estimation to generate The representation of the phase difference between at least one pilot carrier in the data symbol and the pilot carrier in at least one long adjustment symbol; and an average phase shift circuit, which uses the representation of the phase difference as An average of the phase difference between at least one pilot carrier in the data symbol and at least one pilot carrier in the at least one long adjustment symbol is generated based on the average of the phase difference. Update frequency offset on the number generator. 108. If the circuit of the patent application No. 106, further includes a second divider circuit, which is to receive a first > offset-compensated frequency domain representation of the symbol and a The offset-compensated frequency-domain representation of the second data symbol, and the offset-compensated frequency-domain representation of the material symbol of the first patent application, is divided by the offset-compensated offset of the second data symbol. In the frequency domain to generate a phase difference between at least one pilot carrier in a data symbol and at least one pilot carrier in at least one long adjustment symbol; and an average phase shift circuit, which is based on The phase difference representation is used to generate the average of the phase difference between at least one pilot carrier in the data symbol and the at least one pilot carrier in the at least one long adjustment symbol, and to generate a phase difference based on the average of the phase difference. Use the update frequency offset that should be added to the first signal generator. 109. An automatic frequency control circuit for correcting the effect of a frequency offset between a receiver and a transmitter by obtaining the value of an adjustment symbol received during a preamble, the circuit comprising ... The autocorrelator is based on the short adjustment symbol to generate a first vector. The first vector angle represents the coarse offset between the receiver and the transmitter, and the autocorrelator is based on at least one Generate a second vector based on the long adjustment symbol, where the second vector angle represents the fine offset between the receiver and the transmitter; a frequency offset generator, which is based on the first vector angle A rough offset estimate and a fine offset estimate based on the second vector angle; a first signal generator based on the coarse offset estimate to generate a first A first cycle signal of a frequency; a first mixer, which is required to generate more than 122,533 * positive replacement pages received on the receiver ^ 93. 5 > 27 ί six, the scope of the patent application is less than one long adjustment symbol and the first cycle The product of signals is added to the autocorrelator, where the autocorrelator generates a second vector based on the base product; where 'after the frequency offset generator generates a fine offset estimate, the first signal generator system Generating a second periodic signal with a second frequency based on the coarse offset estimation and the fine offset estimation; The first mixer is to multiply a symbol received after receiving at least one long adjustment symbol by a second period signal; a second signal generator is to generate a third period based on a second vector angle A signal; and a second mixer that multiplies the time-domain representation of at least one long adjustment symbol by the third period signal to produce an offset-compensated time-domain representation of at least one long adjustment symbol. 11 〇. If the circuit of the scope of application for patent No. 109, further includes: a first frequency domain conversion unit, which is to be converted at least one long adjustment Offset-compensated time-domain representation of symbols to produce offset-compensated frequency-domain representation of at least one long adjusted symbol received on a receiver; and a divider circuit, The offset compensated frequency domain representation of the long adjustment symbol received on the transmitter is divided by the frequency domain representation of at least one long adjustment symbol that substantially represents the at least one long adjustment symbol sent by the transmitter. The circuit declaring item 109 of the patent further includes a second divider circuit, which is to receive a first 72 VI. Offset-compensated frequency-domain representation of data symbols in the patent application range and offset-compensated frequency-domain representation of a second data symbol, and the offset-compensated frequency of the first data symbol Domain representation divided by offset-compensated frequency domain representation of the second data symbol to produce a representation of the phase difference between at least one pilot carrier in the data symbol and at least one pilot carrier in the at least one long adjustment symbol Law; and An average phase shift circuit based on the representation of phase difference to generate an average of the phase difference between at least one pilot carrier in a data symbol and at least one pilot carrier in at least one long adjustment symbol, And based on the average of the phase differences, an update frequency offset to be added to the first signal generator is generated. 112 · —An automatic frequency control circuit for correcting the effect of a frequency offset between a receiver and a transmitter by obtaining the value of an adjustment symbol received during a preamble. The circuit includes: An autocorrelator is to generate a first vector based on the short adjustment symbol. The first vector angle represents the coarse offset between the receiver and the transmitter ', and the autocorrelator is based on at least A long adjustment symbol is used as a basis to generate a second vector. The second vector angle represents the fine offset between the receiver and the transmitter. A frequency offset generator is based on the first vector angle. Generating a coarse offset estimate and generating a fine offset estimate based on the second vector angle; a first signal generator which is based on the coarse offset estimate to generate a The 73rd of the first frequency is replacing Page 3 3. Hurried 27 B__ VI. One cycle of the patent application scope; a first mixer, which is to generate the eve received on the receiver; a long adjustment symbol and the first The product of the periodic signal and the product is added to the autocorrelator ', where the autocorrelator generates a second vector based on the base product; where' after the frequency offset generator produces a fine offset estimate 'the first signal is produced The mixer is to generate a second periodic signal with a second frequency based on the coarse offset estimation and the fine offset estimation; and wherein the first mixer is to multiply the symbols received after receiving at least one long adjustment symbol by A second period signal; a low-pass filter that receives samples of short adjustment symbols and generates filtered samples of short adjustment symbols; and an adder that uses filtered samples of short adjustment symbols Subtracting from the short trimming samples to produce short trimming samples, which has substantially removed the DC to add to the autocorrelator. 113 · —An automatic frequency control circuit for correcting the effect of a frequency offset between a receiver and a transmitter by obtaining the value of an adjustment symbol received during a preamble, the circuit comprising: a The autocorrelator is based on the short adjustment symbol to generate a first vector. The first vector angle represents the coarse offset between the receiver and the transmitter, and the autocorrelator is based on at least one Generate a second vector based on the long adjustment symbol. The second vector angle represents the fine offset between the receiver and the transmitter. A frequency offset generator is based on the first vector angle. The scope of the patent application generates a coarse offset estimate and a fine offset estimate based on the second vector angle. A first signal generator is based on the coarse offset estimate to generate a A first periodic signal of a first frequency of an offset estimate; a first mixer that is to produce a product of at least one long adjustment symbol received at the receiver and the first periodic signal and add the product to the autocorrelation Where the autocorrelator generates a second vector based on the base product; where 'after the frequency offset generator generates a fine offset estimate, the first signal generator uses a coarse offset estimate and a fine offset estimate Generating a second periodic signal having a second frequency as a basis; and wherein the first mixer is to multiply a symbol received after receiving at least one long adjustment symbol by a second periodic signal; a first adder, which is A third vector is generated based on a set of samples of at least one short adjustment symbol that is not compensated for frequency offset, and is generated on the basis of a set of samples of other at least one short adjustment symbol that is not compensated for frequency offset. A fourth vector; a DC offset compensator that receives the first vector, the third vector, and the fourth vector, and calculates the DC offset (xl-x2) e (Ja ) ~ N {\-eUa)) where α is the first vector angle, χ1 is the third vector, and χ2 is the fourth positive replacement page # y7 II__ VI. Patent application range vector, and N is the frequency offset Compensation of at least one short The number of samples in the integer symbol; and a second adder to produce a difference and a DC offset between samples of the symbol received after the short adjustment symbol. 114 · An automatic frequency control circuit for correcting the effect of a frequency offset between a receiver and a transmitter by obtaining a value of an adjustment symbol received during a preamble, the circuit includes: The correlator is based on the short adjustment symbol to generate a first vector. The first vector angle represents the coarse offset between the receiver and the transmitter, and the autocorrelator has a length of at least one Adjust the symbol as the basis to generate a second vector, the second vector angle of which represents the fine offset between the receiver and the transmitter; a frequency offset generator, which is to generate a second vector based on the first vector angle A coarse offset estimate and a fine offset estimate based on the second vector angle; a first signal generator based on the coarse offset estimate to generate a first offset A first periodic signal of frequency; a first mixer, which is to generate a product of at least one long adjustment symbol received at the receiver and the first periodic signal and add the product to the autocorrelator, wherein the autocorrelator The second vector is generated based on the base product; where the first signal generator is based on the coarse offset estimation and the fine offset estimation after the frequency offset generator generates a fine offset estimation, the patent application scope is calculated as Basically generates a second periodic signal having a second frequency; and wherein the first mixer is to multiply a symbol received after receiving at least one long adjustment symbol by a second periodic signal; a first adder, which is to A third direction 1 'is generated based on the set sampling of at least one short adjustment symbol that is not compensated for the frequency offset, and is generated based on the set sampling of other at least one short adjustment symbol that is not compensated for the frequency offset. A fourth vector; a DC offset compensator, which receives the first vector, the third vector, and the fourth vector, and calculates the DC offset by calculating the following formula (xl — x2) — IN where xl is a third vector, x2 is a fourth vector, and n is the number of samples in at least one short adjustment symbol that is not compensated for frequency offset; and a second adder, which is to be generated in a short tone And the difference between the current sample of the received symbol after the symbol of displacement. 115. A method of correcting the effect of a frequency offset between a receiver and a transmitter by assigning a value to an adjustment symbol received during a preamble, the method comprising: generating a first based on a short adjustment symbol A vector whose first vector angle represents the coarse offset between the receiver and the transmitter, and based on at least one long adjustment symbol to generate a second vector, the second vector angle of which represents the reception Finely biased I2Wi between transmitter and transmitter 6. Shifting the scope of patent application; generating a coarse offset estimate based on the first vector angle and a fine offset estimate based on the second vector angle; The first period signal of the shift estimation; generating a product of the at least one long adjustment symbol and the first period signal received on the receiver and adding the product to the autocorrelator, wherein the autocorrelator generates the second based on the base product Vector; 10 15 20 After generating the fine offset estimation, the first signal generator is to generate a second periodic signal with a second frequency based on the coarse offset estimation and the fine offset estimation; The received symbol after at least one long adjustment symbol is multiplied by the second period signal. 116. The method of claim 115, further comprising: using a second vector angle to generate a frequency domain representation of a third signal having a third frequency substantially equal to the third frequency of the fine offset estimation; and changing the frequency of the third signal The domain representation and the frequency domain representation of the at least one long adjustment symbol received on the receiver are convolved to produce an offset-compensated frequency domain representation of the long adjustment symbol, which makes the receiver and the transmitter The frequency offset effect between the two is substantially removed. 117. The method of claim 116, further comprising generating a frequency domain table error based on at least one long adjustment symbol received on the receiver to generate at least one long adjustment symbol received on the receiver. 78 VI. Patent Application Range 118 119. 120. The method according to item 117 of the patent application range further includes a frequency domain representation that is compensated by offsetting at least one long adjustment symbol received on the receiver. A channel estimate is generated by dividing by the frequency domain representation of at least one-long adjustment symbol transmitted by at least one long adjustment symbol which is substantially representative of the transmission. For example, the method of claim 117, wherein the frequency domain representation of at least the long adjustment symbol includes a frequency domain representation of the first long adjustment symbol and a frequency domain representation of the first long adjustment symbol. The method further includes: An average of the frequency domain representation of at least one long adjustment symbol is generated by averaging the frequency domain representation of the first long adjustment symbol and the frequency domain representation of the second long adjustment symbol; by averaging the frequency of the at least one long adjustment symbol The average of the domain representation and the frequency domain representation of the second signal are convolved to produce an average offset compensation "quoted in the frequency domain representation; by dividing the averaged offset compensated frequency domain representation A channel estimate is generated by using a frequency domain representation of at least one long adjustment symbol that substantially represents at least one long adjustment symbol sent by the transmitter. For example, the method of claim 119 further includes: generating a frequency domain representation of at least one long adjustment symbol that substantially represents at least one long adjustment symbol sent by the transmitter. If the method of applying for patent scope item 118, further includes: dividing an offset-compensated frequency-domain representation of a data symbol by a channel estimate to generate at least one lead carrier in the data symbol and 121. VI. Patent Application The expression of the phase difference between at least one pilot carrier received in at least one adjustment symbol received on the range receiver; based on the representation of the phase difference, at least one pilot carrier in the data symbol is generated on the receiver At least one of the received at least one adjustment symbol indicates an average of the phase difference between the carriers, and generates an updated frequency offset for adding to the first signal generator based on the average of the phase difference. 122. The method of claim 118, further comprising: generating an offset-compensated frequency-domain representation of a first data symbol and an offset-compensated frequency-domain representation of a second data symbol; Offset-compensated frequency-domain representation of a data symbol divided by offset-compensated frequency-domain representation of a second data symbol to generate at least one of a pilot carrier in the first data symbol and at least one of the second data symbol A representation of the phase difference between the pilot carriers; and based on the representation of the phase difference, generating an average of the phase difference between at least one pilot carrier in the data symbol and at least one pilot carrier in the at least one adjustment symbol, And based on the average of the phase differences, a frequency offset for updating to be added to the first signal generator is generated. 123. The method according to item 115 of the patent application scope, further comprising: generating a third period signal based on the second vector angle; and a second mixer for representing the time domain of at least one long adjustment symbol Multiplying the third period signal to produce an offset-compensated time-domain representation of at least one long adjustment symbol. 9122: w 5 6. The scope of patent application 124. The method of claim 123, further comprising: converting the offset-compensated time domain representation of at least one long adjustment symbol to produce at least one long adjustment received on the receiver. Symbol-shifted frequency-domain representation; and By dividing the offset-compensated frequency domain representation of at least one long adjustment symbol received at the receiver by the frequency domain of at least one long adjustment symbol that substantially represents at least one long adjustment symbol sent by the transmitter Representation to generate a channel estimate. 125. The method of claim 124, further comprising: providing a frequency domain representation of at least one long adjustment symbol that substantially represents at least one long adjustment symbol sent by the transmitter. 126. The method for applying for item 124 of the patent scope further includes: Dividing the offset-compensated frequency-domain representation of the data symbol by the channel estimate to generate at least one pilot carrier in the data symbol and at least one pilot carrier in the at least one adjustment symbol received at the receiver. Phase difference representation; based on the phase difference representation method, an average of the phase difference between at least one pilot carrier in the data symbol and at least one pilot carrier in the at least one adjustment symbol received at the receiver is generated And based on the average of the phase difference to generate an updated frequency offset for adding to the first signal generator. 127. The method of claim 124, further comprising: providing an offset-compensated frequency-domain representation of a first data symbol and an offset-compensated frequency-domain representation of a second data symbol; 81 1225 6. The scope of patent application 10 15 20 Divides the offset-compensated frequency domain representation of the first data symbol by the offset-compensated frequency domain representation of the second data symbol to generate at least one reference in the first data symbol A representation of the phase difference between the carrier and at least one of the second data symbols; and based on the representation of the phase difference, generating at least one of the reference carrier and at least one of the adjustment symbols in the data symbol Shows the average of the phase differences between the carriers, and generates an updated frequency offset for adding to the first signal generator based on the average of the phase differences. 128. The method of claim 115, further comprising: receiving a sample of the short adjustment symbol; generating a filtered sample of the short adjustment symbol based on the received sample of the short adjustment symbol; and The filtered samples are subtracted from the short adjustment symbol samples to produce a short adjustment symbol sample with the DC substantially removed. 129. The method of claim 115, further comprising: generating a second vector based on a set of at least one short adjustment symbol that is not compensated for frequency offset, and that is not compensated for frequency offset Generate a third vector based on the set sampling of at least one short adjustment symbol; calculate the DC offset (xl-x2) eUa by calculating the value of the following formula, where α is the first vector angle, and χ1 is the second Vector, χ2 is the third 82 122 Sacrifice 3, `` Replace 4 :. 93. 27, a 6. The scope of the patent application refers to the number of samples in the at least one short adjustment symbol that does not compensate for the frequency offset; and the difference between the samples and the DC offset that generate the symbols received after the short adjustment symbol . 130. The method for applying for item 115 of the patent scope further includes: Generate a second vector based on a set of samples of at least one short adjustment symbol that is not compensated for frequency offset, and generate a second vector based on a set of samples of at least one short adjustment symbol that is not compensated for frequency offset Third vector; Calculate the DC offset (xl-x2) —2N ~ by calculating the following formula: where xl is the second vector, x2 is the third vector, and N is at least the frequency offset is not compensated The number of samples in a short adjustment symbol; and The difference between the samples of the received symbol after the short adjustment symbol and the DC offset are generated. 13 丨 · A method for correcting the effect of a frequency offset between a receiver and a transmitter by giving a value of an adjustment symbol received during a preamble, the method includes: based on a short adjustment character Generate " the first vector, where the ^ vector angle represents the coarse offset between the receiver and the transmitter, and generate a second vector based on at least one long adjustment symbol, the second of which The vector angle represents the fine offset between the receiver and the transmitter. 6. Shifting the scope of patent application; generating a rough offset estimate based on the first vector angle and generating a fine offset estimate based on the first vector angle; The first period signal of the shift estimation; generating a product of the at least one long adjustment symbol and the first period signal received on the receiver and adding the product to the autocorrelator, wherein the autocorrelator generates the second based on the base product Vector; 10 After generating the fine offset estimation, the first signal generator is to generate a second periodic signal with a second frequency based on the coarse offset estimation and the fine offset estimation; it will receive at least one long The symbol received after adjusting the symbol is multiplied by the second periodic signal; 15 is a frequency domain representation of a third signal having a third frequency substantially equal to the third frequency of the fine offset estimate based on the second vector angle; and The frequency domain representation and the frequency domain representation of the long and long adjustment symbols received on the receiver are convolved to produce offset compensation for at least one long adjustment symbol. Frequency domain notation, which allows the effect of frequency offset between the receiver and the transmitter to be substantially removed. 132. The method of claim 131, further comprising dividing the frequency-domain representation of the offset of at least one long adjustment symbol received on the receiver by substantially representing at least one transmitted by the transmitter. A frequency domain representation of at least one long adjustment symbol of a long adjustment symbol to generate a channel estimate. 20 12253¾ ----- 1¾.¾ pages 6. Patent application scope 133. The method according to item 132 of the patent application scope further includes: dividing an offset-compensated frequency domain representation of a data symbol by a channel estimate to generate at least one of the reference carrier and the data symbol. The representation of the phase difference between at least one pilot carrier in at least one adjustment symbol received at the receiver; 10 15 20 Based on the representation of the phase difference, at least one pilot carrier in the data symbol and the The average of the phase difference between at least one of the pilot carriers received in the at least one adjustment symbol received on the receiver, and based on the average of the phase difference, an updated frequency offset for addition to the first signal generator is generated. . 134. The method of claim 132, further comprising: providing an offset-compensated frequency domain representation of the first data symbol and an offset-compensated frequency domain representation of the second data symbol; Offset-compensated frequency-domain representation of a data symbol divided by offset-compensated frequency-domain representation of a second data symbol to generate at least one pilot carrier in the first data symbol and at least one of the second data symbol A representation of the phase difference between the pilot carriers; and based on the representation of the phase difference to generate at least one pilot carrier in the first data symbol and at least one pilot carrier in the second data symbol. The average of the phase differences is used to generate an updated frequency offset based on the average of the phase differences. 13 5. The method according to item 132 of the scope of patent application, wherein the frequency domain representation of at least one long adjustment symbol includes a frequency domain representation of a first long adjustment symbol and a frequency domain representation of a second long adjustment symbol, This method further extends the scope of patent application to include: averaging the frequency domain representation of the first long adjustment symbol and the frequency domain representation of the second long adjustment symbol to generate an average of the frequency domain representation of at least one long adjustment symbol ; By convolving the average of the frequency domain representation of at least one long adjustment symbol and the frequency domain representation of the second signal to produce an average offset-compensated frequency domain representation; by compensating the average offset The passed frequency domain representation is divided by the frequency domain representation of at least one long adjustment symbol that substantially represents at least one long adjustment symbol sent by the transmitter to generate a channel estimate. A method of correcting the effect of a frequency offset between a receiver and a transmitter by assigning a value to an adjustment symbol received during a preamble 'The method includes: generating a first vector based on a short adjustment symbol The first vector angle represents the coarse offset between the receiver and the transmitter, and a second vector is generated based on at least one long adjustment symbol, and the second vector angle represents the receiver and the Fine offset between transmitters; a coarse offset estimate based on the first vector angle and a fine offset estimate based on the second vector angle; a coarse offset estimate based on the coarse offset estimate The first period signal of the coarse offset estimation; generating a product of at least one long adjustment symbol and the first period signal received on the receiver and adding the product to the autocorrelator, wherein the autocorrelator 6. The patent range correlator generates a second vector based on the base product; after generating the fine offset estimation, the first signal generator generates a first vector based on the coarse offset estimation and the fine offset estimation. A second period signal of two frequencies; multiplying a symbol received after receiving at least one long adjustment symbol by a second period signal; a third period signal based on a second vector angle; and when at least one long adjustment symbol is received The domain representation is multiplied by the third period signal to produce an offset-compensated time domain representation of at least one long adjustment symbol. 137. The method of claim 136, further comprising: converting an offset-compensated time-domain representation of at least one long adjustment symbol to generate offset compensation for at least one long adjustment symbol received on the receiver. Frequency domain representation; and by dividing the offset of at least one long adjustment symbol received at the receiver by the frequency domain representation divided by at least one of the at least one long adjustment symbol transmitted by the transmitter. The frequency domain representation of a long adjustment symbol produces a channel estimate. 13 8. The method of claim 137, further comprising: dividing an offset-compensated frequency-domain representation of a data symbol by a channel estimate to generate at least one pilot carrier in the data symbol and on the receiver At least one of the received at least one adjustment symbol indicates a phase difference between carriers; based on the expression of the phase difference, 87 to 122 of the data symbols are generated. Sixth, the scope of patent application is 5 10 15 20 The average of the phase difference between at least one pilot carrier and at least one pilot carrier in at least one adjustment symbol received on the receiver is generated based on the average of the phase differences Updated frequency offset to be applied to the first signal generator. 139. The method of claim 137, further comprising: providing an offset-compensated frequency domain representation of the first data symbol and an offset-compensated frequency domain representation of the second data symbol; Offset-compensated frequency-domain representation of a data symbol divided by offset-compensated frequency-domain representation of a second data symbol to generate at least one pilot carrier in the first data symbol and at least one of the second data symbol A representation of the phase difference between the pilot carriers; and based on the representation of the phase difference to generate at least one pilot carrier in the first data symbol and at least one pilot carrier in the second data symbol. The average of the phase differences is used to generate an updated frequency offset based on the average of the phase differences. 140 · —An automatic frequency control circuit for correcting the effect of a frequency offset between a receiver and a transmitter by obtaining the value of an adjustment symbol received during a preamble, the circuit comprising: a The autocorrelator is based on the short adjustment symbol to generate a first vector. The first vector angle represents the coarse offset between the receiver and the transmitter, and the autocorrelator is based on at least one Generate a second vector based on the long adjustment symbol, where the second vector angle represents the fine offset between the receiver and the transmitter; a frequency offset generator based on the first vector angle 88 12253 10 15 20 VI. The scope of the patent application generates a coarse offset estimate and a fine offset estimate based on the second vector angle. A first signal generator is to generate a coarse offset estimate based on the coarse offset estimate. A first periodic signal having a first frequency substantially equal to the rough offset estimate; a first mixer that is to produce a product of at least one long adjustment symbol and a first periodic signal received on the receiver and add the product to Autocorrelator, where the autocorrelator generates a second vector based on the base product; where 'after the frequency offset generator generates a fine offset estimate, the first signal generator uses a coarse offset estimate and a fine The offset estimation is based on generating a second period signal having a second frequency; and wherein the first mixer is to multiply a symbol received after receiving at least one long adjustment symbol by a second period signal; and receiving a short adjustment symbol. Sampling; generating filtered samples of the short adjustment symbol based on the received samples of the short adjustment symbol; and removing the filtered samples of the short adjustment symbol from the short tone Sampling the symbols of the short subtracted to produce the sampled DC-adjusted symbol of the substance to be removed. 141. An automatic frequency control circuit for correcting the effect of a frequency offset between a receiver and a transmitter by obtaining the value of an adjustment symbol received during a preamble. The circuit includes: Correlator that generates parameters based on short adjustment symbols 89 Sixth, the scope of the patent application-the -th vector-its first-vector angle indicates the coarse offset between the receiver and the transmitter, and the autocorrelator is to generate a second Vector, the second vector angle of which represents the fine offset between the receiver and the transmitter; a frequency offset generator, which is based on the first vector angle to generate a coarse offset estimate and the second vector Angle-based to generate a fine offset estimate; A first signal generator for generating a first periodic signal having a first frequency substantially equal to the first frequency of the rough offset estimation based on the coarse offset estimation; a first mixer for generating a signal at the receiver Receiving the product of the at least one long adjustment symbol and the first period signal and adding the product to the autocorrelator, wherein the autocorrelator generates a first vector based on the product; Wherein, after the frequency offset generator generates a fine offset estimation, the first signal generator is to generate a second periodic signal having a second frequency based on the coarse offset estimation and the fine offset estimation; and A mixer is to multiply a symbol received after receiving at least one long adjustment symbol by a second period signal; and generate a second vector based on a set of at least one short adjustment symbol that is not compensated for frequency offset. , And generate a third vector based on the set sampling of at least one short adjustment symbol that is not compensated for the frequency offset; calculate the DC offset 90 by calculating the value of the following formula, patent application range (xl- x2) eUa) ~ N (l-eUa) Y where α is the first vector angle, χ1 is the second vector, χ2 is the third vector, and N is the sample in at least one short adjustment symbol that does not compensate for frequency offset. The number of samples; and the difference and the dc offset between the samples of the symbol received after the short adjustment symbol. 142. An automatic frequency control circuit for correcting the effect of a frequency offset between a receiver and a transmitter by obtaining the value of an adjustment symbol received during a preamble, the circuit comprising: A correlator is to generate a vector based on a short adjustment symbol. The first vector angle represents a coarse offset between the receiver and the transmitter, and the autocorrelator needs to adjust the symbol by at least one long. A second vector is generated on the basis of which the second vector angle represents the fine offset between the receiver and the transmitter; a frequency offset generator which generates a coarse offset based on the first vector angle Shift estimation and generating a fine offset estimate based on the second vector angle; a first signal generator based on the coarse offset estimate to generate a first frequency having a first frequency substantially equal to the coarse offset estimate; A first period signal; a first mixer, which is to generate a product of at least one long adjustment symbol received at the receiver and the first period signal and add the product to the autocorrelator, wherein the autocorrelator is Based on the product, a second vector of the patented range is generated. Among them, after the frequency offset generator generates a fine offset estimate, the first signal generator generates a first offset based on the coarse offset estimate and the fine offset estimate. A second periodic signal having a second frequency; and wherein the first mixer is to multiply a symbol received after receiving at least one long adjustment symbol by a second periodic signal; 10 at least one without compensation for frequency offset A second vector is generated based on the set sampling of the short adjustment symbols, and a third vector is generated based on the set sampling of at least one short adjustment symbol that is not compensated for the frequency offset; Calculate the DC offset (xl-x2) — IN where xl is the second vector, X2 is the third vector, and n is the number of samples in at least one short adjustment symbol that is not compensated for the frequency offset; and generated in The difference between the samples of the received symbol after the short adjustment symbol and the DC offset. 15