KR100685999B1 - Apparatus and Method of rake combining using noise estimator - Google Patents

Abstract

The present invention relates to rake synthesis using a noise predictor. The present invention provides a CDM receiver using antenna diversity, a Walsh despreader for extracting symbols of a CDM channel by multiplying Walsh codes, a symbol generator for generating pilot and data symbols of the extracted symbols, and the generated symbols. A channel compensator for channel compensation by predicting path size and phase characteristic information of a corresponding finger, a noise predictor for extracting noise symbol power using a Walsh code not used in the Walsh despreader, and a symbol generated by the symbol generator A noise compensator for outputting a noise power value of a finger corresponding to a reliable signal to a noise compensator of each finger, and a noise compensator receiving the noise power value and performing noise compensation on the channel compensated signal Provided is a rake synthesis apparatus using a predictor. Therefore, according to the present invention, it is effective to implement an optimal rake synthesizer having a high SNR in consideration of noise components during rake synthesis.

CDM, Rake synthesis, noise predictor, satellite DMB

Description

Apparatus and Method of rake combining using noise estimator

1 is a block diagram showing the configuration of a rake synthesizer according to the prior art;

Figure 2 is a block diagram showing the structure of a rake synthesizer according to the present invention

3 is a block diagram showing the configuration of a noise predictor according to the present invention.

4 is a block diagram showing the configuration of a pilot power predictor according to the present invention.

5 is a block diagram showing the configuration of a noise compensator according to the present invention.

6 is a block diagram conceptually showing the overall configuration of a satellite DMB receiver to which a rake synthesizer is applied according to the present invention;

-Explanation of symbols for the main parts of the drawings-

110 Walsh despreader 120 Symbol generator

130: channel compensator 140: noise compensator

150: Deskewer 160: Noise Predictor

170: pilot power predictor 200: noise compensation controller

300: finger controller

The present invention relates to rake synthesis, and more particularly, to a rake synthesis method and apparatus using a noise predictor applied to satellite DMB of the CDM (Code Division Modulation) method.

Digital Multimedia Broadcasting (DMB) can be divided into terrestrial DMB and satellite DMB.

The terrestrial DMB provides audio and video services during the movement based on Orthogonal Frequency Division Multiplexing (OFDM), and the satellite DMB is based on CDM (Code Division Modulation) to compensate for satellites and the ground gap filler (Gap Filter). To enable audio and video services on the go.

Currently, the technical standard of satellite DMB adopted in Korea and Japan is the system E method defined by the International Telecommunication Union (ITU), which basically adopts the CDM transmission method, and uses CD quality sound and weather, traffic and video information using various channels. It is a new concept service of representative communication and broadcasting convergence.

The satellite DMB uses frequencies in the uplink 13.824 to 13.883 GHz band and the downlink 2.630 to 2.655 GHz and 12.21 to 12.23 GHz bands, and supports up to 64 channels (32 channels in Korea and Japan). There is a wide range of features.

The transmission channel of the satellite DMB is a wireless mobile reception channel, and the amplitude of the received signal is not only time-varying, but also the Doppler shift of the received signal spectrum occurs due to the mobile reception.

In consideration of the transmission and reception in such a channel environment, the satellite DMB transmission scheme adopts the CDM scheme, and interleaves the time domain signals to correct errors occurring in the transmission channel.

The CDM method multiplies the data to be transmitted by a pseudo noise signal having a data rate much faster than that of the data to spread and transmit the signal by spreading the frequency, so that the signal exists over a wide band and thus narrow-band interference ), It is possible to reduce the deterioration of reception performance due to the multipath through the receiver of the RAKE structure.

Diversity is the most widely used method for overcoming the degradation of reception performance due to such a multipath input signal. The purpose of the diversity is to reduce the depth and length of fading caused by the multipath input signal by supplying two or more transmission signals over different independent fading channels to the receiver.

There are various ways to implement this, and 'Multipath Diversity' which can be used in the dual CDM receiver is described as follows.

In a frequency selective fading channel, a received signal is composed of transmission signals that pass through independent multipaths. At this time, the signals passing through each path have different path delays, magnitudes, and phases. Multipath diversity is a method of decomposing and synthesizing these multipath signals.

The CDM receiver can decompose signals along different paths using the characteristics of the PN code. Accordingly, the multipath diversity is realized by synthesizing the multipath signals decomposed according to the PN code characteristics.

There are various methods of synthesizing, and the most widely used method is Maximum Ratio Combining (MRC). The principle of the maximum ratio synthesis method (MRC) is as follows.

Assuming that the signal of the transmitting end is transmitted in a PSK (Phase Shift Keying) method, the transmission signal S (t) = U (t). (U (t) = +1, -1)

The transmission signal is changed through the air path as follows.

는 시간 t에서 k번째 전송 경로의 채널 특성을 나타내는 복소수이며,

는 시간 t에서 k번째 전송 경로에 첨가되는 잡음이다.Where above

Is a complex number representing the channel characteristics of the kth transmission path at time t,

Is the noise added to the k th transmission path at time t.

At this time, the most common linear method (Linear Combining Rule) is as follows.

는 합성을 위한 합성 계수이며, 최적으로 합성된 결과는 신호대 잡음비(SNR : Signal to Noise Ratio)가 최대가 되어야 할 것이다.Where above

Is a synthesis coefficient for synthesis, and the optimally synthesized result should have a maximum signal-to-noise ratio (SNR).

Therefore, the signal component in the above formula is

It is expressed as follows, and therefore, the SNR is as follows.

가 시간의 변화에 둔감하다 가정할 경우, 시간 t동안에는 일정한 값을 갖는다고 볼 수 있으므로 시간 간격 t가 사라져

으로 바뀌게 되고, 송신 신호 성분 u(t)는 +1, -1 값만 가지므로 power term에서는 1로써 표현되게 된다. 마지막으로 Nk는 k번재 다중 경로에 인가된 잡음의 변화를 의미한다.In Equation 4

Is insensitive to changes in time, the time interval t disappears because it can be considered to have a constant value during time t.

Since the transmission signal component u (t) has only +1 and -1 values, it is represented as 1 in the power term. Finally, Nk is the change in noise applied to the kth multipath.

는 다음과 같다.Applying Cauchy-Schwarz Inequality to find the maximum value in the above equation is the maximum

Is as follows.

This means that in order to achieve maximum non-synthesis (MRC), each multipath input signal must be multiplied by the conjugate of the channel characteristics passed by each signal, and divided and compensated by dividing the noise characteristics before synthesis. .

In this case, when the noise applied to each multipath channel is white noise, the Nk terms are all the same regardless of k. Therefore, the Nk term is ignored and the maximum non-synthesis method can be implemented by compensating only the multipath channel characteristics.

Therefore, the Rake Combiner used in the CDM receiver compensates for the channel characteristics of the multipath input signal applied to each finger, and compensates each channel after the channel delay is compensated. It will be implemented in an additive manner.

1, which will be described with reference to FIG. 1 as follows.

1 is a block diagram showing the configuration of a rake synthesizer according to the prior art.

As shown in FIG. 1, the rake synthesizer is implemented in such a manner that channel characteristics and channel delays are compensated and added at each finger 10-1, 10-2, ..., 10-n.

That is, in the symbol generator of each finger, the PN despreader and the Walsh despread signal are integrated and dumped into a symbol, and a channel compensator and a channel delay compensation for the symbol are performed. Deskewer) to print, and add them all together.

Likewise, in order to implement Maximum Ratio Combining, the rake synthesizer should compensate not only the channel characteristic of each multipath input signal but also the noise characteristic of the channel.

However, in general, since the noise characteristic is assumed to be the same in all multipath channels in consideration of general white noise, the noise characteristic does not compensate.

However, in the case of satellite DMB using antenna diversity technology using two antennas, two RF stages exist, and thus multipath signals having different white noises exist.

Therefore, there is a need for estimating different white noises generated from different RF stages and reflecting them in synthesis.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and an object of the present invention is to obtain a noise predictor that maximizes SNR by synthesizing in consideration of the amount of different white noise between antennas in a satellite DMB using antenna diversity. It is to propose a rake synthesis apparatus and method used.

In order to achieve the above object, the present invention, in the CDM receiver using the antenna diversity, a Walsh despreader to extract the symbols of the CDM channel by multiplying the Walsh code, and a symbol for generating a pilot and data symbols of the extracted symbols With generator. A channel compensator for predicting the channel size and phase characteristic information of the corresponding finger of the generated symbol by channel compensation, a noise predictor extracting noise symbol power using a Walsh code not used by the Walsh despreader, and the symbol generator A noise compensator for outputting a noise power value of a finger corresponding to a reliable signal among the symbols generated by the signal to the noise compensator of each finger, and a noise compensator for noise compensation from the channel compensated signal by receiving the noise power value. Provided is a rake synthesis apparatus using a noise predictor configured to include.

The noise compensation controller is provided with antenna information of each finger of the CDM receiver, and provides a noise power value of a finger corresponding to a reliable signal for each antenna, and a finger for providing antenna information of each finger. A controller is included.

A pilot power predictor is used to predict pilot power to find the reliable signal.

The noise predictor includes a Walsh despreader that performs Walsh despreading using the unused Walsh code, an Integrate & Dump unit that receives the despreaded signal and generates a noise symbol, and a power value of the generated noise symbol. And a multiplier for multiplying a proportional constant value for using the calculated noise symbol power for noise compensation.

The noise compensator divides a signal whose channel path characteristic is compensated by the noise power according to a maximum ratio synthesis method (MRC) formula.

In the Rake synthesis method of a CDM receiver using antenna diversity, the method comprises: extracting a symbol of a CDM channel by multiplying a Walsh code, generating a pilot and a data symbol of the extracted symbol; Predicting channel size and phase characteristic information of a corresponding finger of a symbol to compensate for the channel; extracting a noise symbol power using a Walsh code not used in the symbol extraction step of the CDM channel; The present invention provides a rake synthesis method using a noise predictor including noise compensation of each finger by using noise power values of a finger corresponding to a reliable signal.

Therefore, according to the present invention, it is effective to implement an optimal rake synthesizer having a high SNR in consideration of noise components during rake synthesis.

Hereinafter, with reference to the accompanying drawings, preferred embodiments of the present invention that can specifically realize the above object will be described.

In addition, the term used in the present invention was selected a general term that is widely used at present, but according to the emergence of a new technology, the term that the applicant deemed most appropriate in the present invention was arbitrarily used, and the meaning of the term in the corresponding description. It will be explained clearly. Therefore, in the understanding of the present invention, it is intended that the present invention should be understood as the meaning of terms rather than simple names of terms.

2 is a block diagram showing the structure of a rake synthesizer according to the present invention.

As shown in FIG. 2, the rake synthesizer according to the present invention has a Walsh despreader 110, a noise estimator 160, a noise compensator 140, and pilot power in a conventional rake synthesizer structure. It includes a predictor (Pilot Power Estimator) 170, a noise compensation controller (200) and a finger controller (300).

Looking at the operation relationship of the rake synthesizer according to the present invention configured as described above are as follows.

First, in case of CDM signal received through multipath, it has the ability to resolve the multipath due to the characteristics of the PN code used in transmission, but cannot separate the added noise in each independent multipath.

In more detail, the process of decomposing each multipath in the multipath input signal is covered with a PN code using the characteristics of the PN code, and the PN code despreading process is performed on the transmitted signal. This is because the signal of the displaced path is made close to 0 because the noise added in the multipath is not a PN coded signal, so that the receiving end does not despread the PN code of any start position.

Therefore, since noise symbols obtained at each finger remain noise added in all received multipaths, noise component decomposition of each multipath is impossible.

However, in the case of a CDM receiving system using antenna diversity, it is possible to know which antenna is the signal coming from. Therefore, the white noise added by an RF system such as a tuner at the receiving end can be distinguished and the aforementioned Maximum Ratio Combining formula can be distinguished. It can be reflected according to.

In other words, the different RF stages exist to implement antenna diversity, so that different white noise is generated according to the maximum ratio synthesis (MRC) formula in the RAKE synthesizer.

To this end, the noise predictor 160 extracts only a noise component of a received signal component using a Walsh code that is not used.

That is, since Walsh codes having 64 code lengths are used for channel multiplexing according to the satellite DMB standard, 64 broadcast channels can theoretically be transmitted simultaneously, but 32 broadcast channels are actually used due to inter-code interference problems. . Therefore, the noise predictor 160 extracts noise components using Walsh codes that are not used for broadcast multiplexing.

The mathematical development process of this principle is as follows.

는 송신 정보를 나타내는 함수,

는 왈시 코드 함수, p(k)는 PN 코드 함수, n(k)는 송신 신호가 거쳐온 경로에서 첨가된 잡음 함수를 의미한다.In Equation 6, r (k) is chip data received through one path, k is a time index, and i is a channel index (i = 0: pilot channel, i = 1 to 32). : data channel), β is a coefficient representing the characteristic of the path that the transmission signal

Is a function representing the send information,

Is the Walsh code function, p (k) is the PN code function, and n (k) is the noise function added in the path through which the transmission signal is passed.

At this time, if the received r (k) PN despreading is as shown in equation (7).

If the PN despread r (k) dr (k) is despread with Walsh using a Walsh code not used in the above-described broadcast multiplexing, Equation 8 is obtained.

In Equation 8, integrate & dump in 64 chip units to remove data channel components including pilots and to leave only noise components can extract noise symbols.

Using this principle, the noise predictor 160 generates a noise power using a Walsh code that is not used for broadcast multiplexing, and generates noise power.

Subsequently, the symbol generator 120 of FIG. 2 generates pilot and data channel symbols using Integration & Dump. The channel compensator 130 receiving the signal predicts the magnitude and phase characteristic information of the path through which the signal applied to the finger passes and conjugate compensation according to the MRC method. Compensation).

At this time, the pilot power predictor 170 predicts the power of the received signal through the path currently fingered to the noise compensation controller 200 by using the pilot symbol.

The noise compensation controller 200 receives the antenna information of the fingers currently hanging from the finger controller 300, the noise power value of each finger, and the pilot power value, and then outputs the largest pilot power among the fingers input from the antenna 1. The noise power value of the finger is sent to the noise compensator 140 of the fingers hanging on the antenna 1.

Similarly, the noise power value of the finger having the largest pilot power among the fingers input from the antenna 2 is transmitted to the noise compensator 140 of the fingers of the antenna 2.

Therefore, the noise compensator 140 receives a signal whose path characteristic is compensated from the channel compensator 130 and divides it into the noise power described above. For hardware implementation, an approximation using a ROM table is actually performed. Division is performed. Details thereof will be described later.

Afterwards, the deskewer 150 at the rear stage compensates for the delay distortion occurring while passing the path of the current finger.

As described above, the reason why the rake synthesizer operates according to the present invention is that, as described above, the noise of each independent path through which the signals received through the multipath cannot be extracted separately, but the two are generated due to antenna diversity. This is because the white noise difference of the RF stage can be compensated for.

Therefore, the most reliable signal received from antenna 1 is assumed to be a signal with a large pilot power, and the noise power value predicted from the finger on which the signal is applied is used for noise compensation of all signals received by antenna 1. . Of course, the same applies to antenna 2.

In this case, it is assumed that a signal having high reliability is a signal having high pilot power, but this is an embodiment for implementation, and therefore, a signal having high reliability may be selected and used by various methods besides pilot power. In light of this, it will be self explanatory.

On the other hand, with reference to the accompanying drawings, the operation relationship of the above-described block according to the present invention in more detail as follows.

3 is a block diagram showing the configuration of a noise predictor according to the present invention.

As shown in FIG. 3, the noise predictor 160 according to the present invention includes an unused Walsh code despreader 161, an integrate & dump unit 163, and a power calculator 165. And a constant multiplier 167.

The despreader 161 performs Walsh despreading using a Walsh code that is not used for broadcast multiplexing.

The Integrate & Dump unit 163 receives the despread signal and generates a noise symbol, and the power calculator 165 calculates a power value of the generated noise symbol.

At this time, the power calculator 165 extracts the noise symbol power by the I ^ 2 + Q ^ 2 method by using the I (Inphase) noise symbol and the Q (Quadratur) noise symbol.

Thereafter, the constant multiplier 167 of the rear stage multiplies and outputs a proportional constant value for using the noise power for noise compensation.

4 is a block diagram showing the configuration of a pilot power predictor according to the present invention.

As illustrated in FIG. 4, the pilot power predictor 170 includes a power calculator 171 and a low pass filter 173.

The pilot power predictor 170 having the above configuration calculates the power by receiving the I and Q symbol data of the pilot channel in the power calculator 171, and then calculates a high frequency component by the baseband filter 173 at a later stage. To remove it.

In this case, the baseband filter 173 is used because a change in the pilot power value that is too sensitive may deteriorate reliability in selecting a finger having the largest power.

5 is a block diagram showing the configuration of a noise compensator according to the present invention.

As shown in FIG. 5, the noise compensator 140 according to the present invention includes a time division mux 141, a value to address 143, and a ROM table. 145 and a time division demux 147.

The noise compensator 140 receives and divides a noise component according to a Maximum Ratio Combining formula. However, the design of the divider on the actual hardware has a problem that the volume is very large.

Therefore, according to the present invention, after receiving the dividend and the divider as inputs, normalize the value appropriately, allocate the value as the address value of the ROM table, and divide the result value contained in the address. Use the method of using.

Of course, it will be apparent that a divider of various methods may be used to implement the noise compensator 140.

The divider using the ROM table 145 according to the present invention uses time division multiplexing-demultiplexing of parallel input signals using a fast clock to maximize volume efficiency. The serialization method is used.

Accordingly, the time division mux 141 and the time division demux 147 are blocks for time division multiplexing-demultiplexing the I, Q symbol data of the pilot channel and the I and Q symbol data of the data M channel in the data 1.

Accordingly, the signal passing through the time division mux is divided by the number, and the noise power value input from the noise compensation controller 200 is received as the divider, and the address switcher 143 performs the normalization of the value through the ROM. It is assigned to the address of the table 145.

The ROM table 145 is a table composed of a two-dimensional array and outputs a value corresponding to the address.

For example, when an address is (x, y), x corresponds to a dividend and y corresponds to a dividend. Therefore, if the address is (6,3), the value of the ROM table at the address is 6 2, the result of / 3, is stored.

The value is output by parallelizing the noise compensated symbol data serialized through the time division demux 147 into pilot channels I, Q, and data 1 to data M channels I and Q.

Meanwhile, FIG. 6 is a block diagram conceptually showing the overall configuration of the satellite DMB receiver to which the rake synthesizer according to the present invention is applied.

As shown in FIG. 6, the received signal input to the antenna is converted into a baseband signal through a tuner 1, and the magnitude of the signal input through the AGC (Automatic Gain Controller) 3 at a later stage is kept constant. do.

A signal having a relatively constant magnitude passing through the AGC 3 is sampled through an analog to digital converter (A / D) 5 and converted into a digital signal.

In the CDM transmission method, in order to demodulate a signal, the acquisition of a pseudo-noise sequence used for spreading the signal has to be prioritized. This process is divided into two stages, acquisition and tracking.

The division unit of the pseudo noise sequence is called a chip, and the acquisition is a process of securing signal synchronization within ± 1/2 chips at a receiver, which is performed by a searcher 7. The tracking of the signal means fine tuning of the found signal and is performed in the trackers 9-1, ..., 9-n.

The signal synchronized through this process is despread by multiplying the pseudonoise sequence generated by the receiver in the PN despreaders 11-1, ..., 11-n, and the RAKE synthesizer 15 according to the present invention. Is entered.

The RAKE synthesizer 15 extracts the symbols of the CDM channel by multiplying the Walsh codes used to distinguish the CDM channels, and in consideration of the white noise between the antenna 1 and the antenna 2 using the noise predictor according to the present invention. The extracted symbol is synthesized.

This process is performed in all the multiple paths that the searcher 7 finds, each referred to as a finger.

Meanwhile, the frequency offset estimator 13 estimates the frequency offset for each finger, synthesizes it, and feeds it back to the tuner 1 to correct the frequency offset.

The rake synthesis is performed for all CDM channels which are desired to be demodulated. In particular, the pilot channel, which is a control channel, contains information on interleaved size and convolutional coding rate and thus must be demodulated.

Thereafter, the timing is extracted from the frame and super frame timing extraction circuit 17 using the pilot channel, and Viterbi decoding, deinterleaving, RS decoding, and the like are performed in the corresponding blocks 19, 21, and 23, respectively. Through the decoded signal, the data mode detector 25 obtains information on the interleave size and the convolutional coding rate to use for decoding the data channel.

On the other hand, the data channel is displayed on the screen by A / V decoding through a process such as time deinterleaving, Viterbi decoding, byte deinterleaving, RS decoding, etc. through the information obtained from the pilot signal.

On the other hand, the present invention can be applied to any CDM receiver using antenna diversity.

The present invention is not limited to the above-described embodiments, and as can be seen in the appended claims, modifications can be made by those skilled in the art to which the invention pertains, and such modifications are within the scope of the present invention.

The rake synthesis method and apparatus using the noise predictor according to the present invention described above have the following effects.

First, there is an effect of implementing an optimal rake synthesizer having a high SNR in consideration of the noise component during rake synthesis.

Second, the output of the rake synthesizer with high SNR has the effect of increasing the reception rate in the case of the actual urban shadow area or low reception signal power.

Claims (11)

In a CDM receiver using antenna diversity, A Walsh despreader that multiplies the Walsh codes to extract the symbols of the CDM channel; A symbol generator for generating pilot and data symbols of the extracted symbols; A channel compensator for predicting the channel size and the phase characteristic information of the corresponding finger of the generated symbol by channel compensation; A noise predictor for extracting noise symbol power using Walsh codes not used in the Walsh despreader; A noise compensation controller for outputting a noise power value of a finger corresponding to a reliable signal among symbols generated by the symbol generator to a noise compensator of each finger; And, And a noise compensator configured to receive the noise power value and compensate for noise in the channel compensated signal. The method of claim 1, wherein the noise compensation controller, And receiving the antenna information of each finger of the CDM receiver and providing a noise power value of a finger corresponding to a reliable signal for each antenna. The method of claim 2, Rake synthesis apparatus using a noise predictor, characterized in that it comprises a finger controller for providing antenna information of each finger. The method of claim 1, And a pilot power predictor for predicting pilot power and outputting the pilot power to the noise compensation controller in order to find a reliable signal from the symbol generator. The noise predictor of claim 1, A Walsh despreader that performs Walsh despreading using the unused Walsh code; An Integrate & Dump unit for receiving the despread signal and generating a noise symbol; A power calculator for calculating a power value of the generated noise symbol; And a multiplier for multiplying the calculated noise symbol power by a proportional constant value for use in noise compensation. The method of claim 1, wherein the noise compensator, And dividing the channel compensated signal by the noise power. The method of claim 6, And the noise compensator uses a divider using a ROM table. The method of claim 1, And an inverse squash for compensating for the delay distortion of the received signal from the channel compensator to each finger. In the rake synthesis method of the CDM receiver using the antenna diversity, Multiplying the Walsh codes to extract the symbols of the CDM channel; Generating pilot and data symbols of the extracted symbols; Channel compensation by predicting path size and phase characteristic information of a corresponding finger of the generated symbol; Extracting noise symbol power using a Walsh code not used in the symbol extraction step of the CDM channel; And, And compensating for noise at each finger by using a noise power value of a finger corresponding to a signal with high reliability among the generated symbols. The method of claim 9, wherein the noise compensation step, And receiving the antenna information of each finger of the CDM receiver, and providing a noise power value of a finger corresponding to a reliable signal for each antenna. The method of claim 10, The high reliability signal is a rake synthesis method using a noise predictor, characterized in that the largest pilot power signal.

Images