CN101127753A - A Channel Estimation Method Applicable to Multi-carrier System - Google Patents

Abstract

The utility model discloses a channel estimation method for multi-carrier multi-antenna systems, which comprises the following steps: extracting the received signals of pilot frequency locations at the receiving end and conducting a preliminary estimate of the channel to attain the initial channel time domain impulse responses of each antenna; inputting the initial channel time domain impulse responses into the energy compensation module to select the effective path, and conduct energy compensation for each effective path and eliminate the impacts on the channel estimation of the energy spreading due to virtual carrier waves, so as to achieve a more accurate time domain impulse response; implementing the Fourier transformation to achieve the response estimate of channel frequency domain.

Description

A Channel Estimation Method Applicable to Multi-carrier System

technical field

The invention relates to the field of wireless communication, especially the future mobile communication system of high-speed communication, and provides a high-performance channel estimation method suitable for multi-carrier systems, especially for multi-carrier multi-antenna systems.

Background technique

The future development trend of mobile communication services is high-quality multimedia services, and the throughput requirements of the cell will be increased to 100Mbps ~ 1Gbps. However, the spectrum resources available for future mobile communications are limited. In order to achieve high data rate information transmission on limited spectrum resources, a feasible way is to develop new air interface technologies with higher spectrum utilization.

Multiple-Input Multiple-Output (MIMO) technology, also known as multi-antenna technology, can obtain high spectrum utilization and diversity gain. The combination of MIMO technology and multi-carrier technology, especially Orthogonal Frequency Division Multiplexing (OFDM), has a processing complexity that is only linearly related to the system bandwidth. In recent years, it has received more and more attention in the world, and will become the core technology of the physical layer of the future wireless communication system. In a multi-antenna multi-carrier communication system, the receiver must complete the relevant detection work based on the estimated channel frequency domain characteristics between each transmitting and receiving antenna pair. Channel estimation accuracy is critical to the performance of multi-antenna multi-carrier systems. In addition, since all transmitting antennas transmit signals continuously at the same time, what is received on the receiving antenna is aliasing of signals from all transmitting antennas, which brings greater difficulty to channel estimation.

In an actual system, in order to avoid distortion caused by the roll-off region of the frequency domain response of the filter, part of the carrier of the transmission bandwidth is usually vacant as a guard frequency band. Due to the existence of the guard frequency band, the orthogonality of the pilot is destroyed, the channel estimation performance is reduced, and it is difficult for some high-performance channel estimation methods to adopt simplified algorithms. Therefore, it is necessary to design a more reasonable pilot sequence and a higher-precision receiver terminal algorithm. At present, the channel estimation method for the multi-antenna multi-carrier system under the presence of virtual carriers is mainly based on the least squares criterion (LS) channel estimation method, which is based on the fact that the multipath delay is known, and requires matrix inversion to Guaranteed accuracy. The disadvantage of the above method is that the calculation complexity is high, which cannot meet the needs of practical applications.

Contents of the invention

Aiming at the above-mentioned problems in channel estimation, the present invention proposes a channel estimation method suitable for multi-carrier systems, especially for multi-carrier multi-antenna systems. The invention can well compensate the energy leakage caused by the virtual carrier, effectively suppress the additive Gaussian white noise, and thus obtain better channel estimation performance.

The invention discloses a channel estimation method for a multi-carrier multi-antenna system, comprising steps:

(1) Extract the received signal at the pilot position at the receiving end, perform preliminary channel estimation, and obtain the initial channel time-domain impulse response of each antenna;

(2) Input the obtained initial channel time domain impulse response into the energy compensation module, select the effective path, and perform energy compensation for each effective path, eliminate the influence of the energy spread caused by the virtual carrier on the channel estimation, and obtain more Accurate time domain impulse response;

(3) Perform Fourier transform to obtain frequency domain response estimation of the channel.

Preferably, for a system in which the multipath delay of the channel is unknown, the method includes the steps of:

(1) For the nth symbol, first extract the pilot to calculate the initial value h ⁰ (n) of the time-domain channel estimation on each antenna;

(2) Enter the energy compensation cycle module with unknown time delay, detect the current remaining channel impulse response h ^l (n) in the first cycle, to find the strongest path (l=0 is the initial value), and judge the path Whether it is an effective path, if the path is an effective path, the energy diffusion of the effective path is restored, that is, the complex amplitude of the path is multiplied by the normalized energy diffusion compensation function to obtain the energy diffusion g of the path ^l (n), and remove the strongest path and its energy spread from the remaining channel impulse response to obtain the current remaining channel impulse response h ^l (n)=h ^l-1 (n)-g ^l (n) , while updating the constructed channel impulse response, ${\tilde{h}}^l\left(\mathrm{no}\right)={\tilde{h}}^{l-1}\left(\mathrm{no}\right)+g^l\left(\mathrm{no}\right),$ Otherwise end the loop;

(3) After the cycle ends, analyze the initial value h ⁰ (n) of the channel impulse response, replace the sampling points smaller than the threshold value with the corresponding constructed channel estimation impulse response, and obtain the final channel impulse response h(n ), and get the final frequency domain response estimate H(n) after Fourier transform.

Preferably, for a system with known channel multipath delay, the method includes the steps of:

(1) For the nth symbol, extract the pilot and calculate the initial value h ⁰ (n) of the time-domain channel estimation on each antenna;

并令 ${\hat{h}}_{\mathrm{delay}}\left(n\right)={\hat{h}}_{\mathrm{delay}}^0\left(n\right),$ 接着进入时延已知的能量补偿循环模块，直到满足条件结束循环跳出循环，在每次循环中将每个有效径的复幅度乘以归一化的能量扩散补偿函数，构造由于能量扩散带来的各径之间相互的干扰信号，从有效径的初始信道估计中消除干扰信号，并用此结果更新

得到更为准确的有效径信息；(2) Take out the effective path according to the known multipath time delay, and obtain the initial channel estimation of the effective path

and order ${\hat{h}}_{\mathrm{delay}}\left(\mathrm{no}\right)={\hat{h}}_{\mathrm{delay}}^0\left(\mathrm{no}\right),$ Then enter the energy compensation loop module with known time delay until the condition is met and the loop jumps out. In each loop, the complex amplitude of each effective path is multiplied by the normalized energy diffusion compensation function to construct the The mutual interference signals between the paths of , from the initial channel estimation of the effective path Remove the interfering signal in , and use this result to update

Obtain more accurate effective path information;

乘以归一化系数，构造最终时域信道估计冲激响应

经过傅立叶变换后得到最终的频域响应估计值H(n)。(3) Will

Multiply by the normalization coefficient to construct the final time-domain channel estimation impulse response

After Fourier transform, the final frequency domain response estimation value H(n) is obtained.

Preferably, a normalized energy spread compensation function is constructed according to the effective subcarrier position, the effective subcarrier position is filled into the same energy signal, and the remaining positions are set to zero, and the function obtained after energy normalization by inverse Fourier transform is normalized Normalized energy spread compensation function.

Preferably, the method is applicable to single-send-single-receive, single-send-multiple-receive or multiple-send-single-receive communication systems.

Preferably, the set of all current strongest paths is the effective path experienced by each pair of transmitting antennas and receiving antennas, and its position is the channel multipath delay.

Preferably, whether the energy or absolute maximum value of the current strongest path is greater than a set threshold value, if it is greater than the set threshold value, it is judged as an effective path, otherwise it is judged as an ineffective path.

Preferably, the threshold value is set according to the noise floor.

Preferably, the normalization coefficient is a ratio of the total number of subcarriers to the number of available subcarriers.

Preferably, eliminating interference signals between paths is based on a serial interference cancellation method, or based on a parallel interference cancellation method.

与上一次循环的

的差小于设定门限。Preferably, the loop end condition is the current

with the previous cycle of

The difference is less than the set threshold.

Preferably, the cycle end condition is reaching a set number of cycles.

For a system with unknown channel time delay, accurate multipath time delay information can be obtained by using the method of the invention. Moreover, the present invention can well compensate the energy leakage caused by the virtual carrier, effectively suppress the additive Gaussian white noise, and thus obtain better channel estimation performance.

Description of drawings

Fig. 1 is the block diagram of the channel estimation method for the scheme of unknown channel time delay according to the present invention;

Fig. 2 is a block diagram of a channel estimation method for a scheme with known channel delay according to the present invention.

Detailed ways

According to the present invention, a channel estimation method for a multi-carrier multi-antenna system is provided, including specific steps:

At the sending end, preferably, a pilot with a long number of available subcarriers is selected, and then subjected to inverse Fourier transform to obtain a time-domain pilot signal, and then different cyclic shift processing is performed on the time-domain pilot signal on each antenna; Finally, a cyclic prefix is added to form a transmission frame with data symbols and sent out through the transmission antenna.

Preferably, constant amplitude-zero autocorrelation (CAZAC) sequences with different offset phases are used as pilots, and the maximum possible number of bits is selected for the number of cyclic shifts, that is, the ratio of the total number of system subcarriers to the number of transmitting antennas, and the cyclic shift is guaranteed The number of bits is greater than the maximum multipath delay.

At the receiving end, first extract the received signal in the frequency domain of the pilot position, perform preliminary channel estimation in the frequency domain, set the channel response of the virtual carrier part to zero, and then perform an inverse Fourier transform on the initially estimated channel to obtain the channel time domain of each antenna Impulse response, next, select the effective path, and perform energy compensation for each effective path to obtain a more accurate time domain impulse response, and finally perform Fourier transform to obtain the final channel frequency domain response estimate. The specific steps of the receiving end are as follows:

For systems with unknown channel multipath delay:

(1) First extract the pilot signal, remove the virtual carrier after Fourier transform, and obtain the pilot data on the effective subcarrier position; then divide the received frequency domain pilot signal and the local pilot sequence; then divide the virtual carrier After the carrier position is filled with zeros, inverse Fourier transform is performed to obtain the channel impulse responses of different antennas; finally, according to the information of different cyclic shift positions of different antennas at the transmitting end, the effective data is intercepted, and the remaining positions are zero-filled to obtain the time of each transmitting antenna. The initial value h ⁰ (n) of the domain channel impulse response.

(2) Detect the current remaining channel impulse response h ^l (n) to find the current strongest path, where l is the number of cycles, and l=0 is the initial value.

(3) Judging whether the path is an effective path, if it is an effective path, go to step (4); otherwise end the loop and go to step (7).

Preferably, it can be judged whether the energy or the absolute maximum value of the current strongest path is greater than a set threshold value. If it is greater than the set value, it is judged as an effective path, otherwise it is judged as an ineffective path.

(4) To recover the energy diffusion of the effective path, multiply the normalized energy compensation function by the complex amplitude of the path. Preferably, the following sinc function can be used:

${\mathrm{<span\; class="notranslate">\; Sinc</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\frac{\mathrm{<span\; class="notranslate">\; sin</span>}<span\; class="notranslate">\; (</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&tau;</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; \&pi;</span>}<span\; class="notranslate">\; \&times;</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&tau;</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; \&pi;</span>}}& \mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; \&NotEqual;</span>{\mathrm{<span\; class="notranslate">\; \&tau;</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}\\ \mathrm{<span\; class="notranslate">\; 1</span>}& \mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&tau;</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}\end{array}\right.$

Among them, τ _l represents the position of the lth effective path, M is the number of effective subcarriers, and N is the OFDM symbol length. Obtain the energy expansion g ^l (n) of this path.

(5) The energy expansion of the strongest path is removed from the remaining channel impulse response to obtain the current remaining channel impulse response h ^l (n)=h ^l-1 (n)-g ^l (n).

(6) Update and construct the channel impulse response, ${\tilde{h}}^l\left(\mathrm{no}\right)={\tilde{h}}^{l-1}\left(\mathrm{no}\right)+g^l\left(\mathrm{no}\right),$ in In order to construct the initial value of the channel impulse response, the value of each sampling point is zero, and return to step (3).

(7) Analyze the initial value h ⁰ (n) of the channel impulse response, and replace the sampling points smaller than the threshold value with the corresponding constructed channel estimation impulse response to obtain the final time-domain channel impulse response h(n), After Fourier transform, the final frequency domain response estimation value H(n) is obtained.

For a system with known channel multipath delay:

(1) First extract the pilot frequency, remove the virtual carrier after Fourier transform, and obtain the pilot data on the effective subcarrier position; then divide the received frequency domain pilot signal and the local pilot sequence; then divide the virtual carrier After the position is filled with zeros, inverse Fourier transform is performed to obtain the channel impulse response of different antennas; finally, according to the information of different cyclic shift positions of different antennas at the transmitting end, the effective data is intercepted, and the remaining positions are zero-filled to obtain the time domain of each transmitting antenna Initial value h ⁰ (n) of the channel impulse response.

(2) Take out the effective path according to the known multipath time delay, and obtain the initial channel estimation of the effective path and order ${\hat{h}}_{\mathrm{delay}}\left(\mathrm{no}\right)={\hat{h}}_{\mathrm{delay}}^0\left(\mathrm{no}\right).$

中消除干扰信号，用此结果更新

得到更为准确的有效径的信道估计值。(3) Multiply the complex amplitude of each effective path by the normalized energy dispersal compensation function to construct the mutual interference signal between the paths due to energy dispersal, and estimate from the initial channel of the effective path

Eliminate interfering signals in , and use this result to update

A more accurate channel estimation value of the effective path is obtained.

Eliminate interfering signals between the paths,

Preferably, based on the serial interference cancellation method;

Preferably, it can also be based on a parallel interference cancellation method.

(4) If the cycle end condition is met, go to step (5); otherwise, go back to step (3).

与上一次循环的

的差小于设定门限；Preferably, the loop end condition is the current

with the previous cycle of

The difference is less than the set threshold;

Preferably, the cycle end condition is reaching a set number of cycles.

乘以归一化系数，构造最终时域信道估计冲激响应

经过傅立叶变换后得到最终的频域响应估计值H(n)。其中，归一化系数为总子载波数与可用子载波数的比值。(5) will

Multiply by the normalization coefficient to construct the final time-domain channel estimation impulse response

After Fourier transform, the final frequency domain response estimation value H(n) is obtained. Wherein, the normalization coefficient is a ratio of the total number of subcarriers to the number of available subcarriers.

The present invention is also applicable to single send single receive (SISO), single send multiple receive (SIMO) and multiple send single receive (MISO) communication systems.

The channel estimation method for a multi-carrier multi-antenna system of the present invention will be described below with reference to the accompanying drawings and specific embodiments.

A typical multi-antenna multi-carrier system is taken as an example to illustrate the effectiveness of the channel estimation method based on energy compensation proposed by the present invention. It is assumed that the antenna configuration of the system is 4 transmitting antennas and 4 receiving antennas, that is, N _T =4, _NR =4. The entire bandwidth is divided into 512 subcarriers by using OFDM modulation, ie N=512, 300 of which are effective subcarriers for transmitting pilot and data, ie M=300, and the remaining 212 subcarriers are virtual carriers.

At the sending end, first generate M=300 long pilot sequences, and fill the virtual carrier position with zeros; then perform inverse Fourier transform to obtain time-domain pilot signals; then perform N/ N _T =128 points of cyclic shift; finally add a cyclic prefix, form a transmission frame with data symbols and send it out through the transmission antenna.

At the receiving end, the channel estimation method is the same for all receiving antennas. For each receiving antenna, first extract the pilot signal, remove the virtual carrier after Fourier transform, and obtain the pilot data of M=300 points on the effective subcarrier position; then combine the received frequency domain pilot signal with the local pilot sequence Phase division; Next, perform inverse Fourier transform after padding the virtual carrier position with zeros to obtain the channel impulse response of different antennas; finally, according to the information of different cyclic shift positions of different antennas at the transmitting end, intercept the effective data of each transmitting antenna respectively, and the rest The position is filled with zeros to obtain the initial value h _tx ⁰ (n) of the time-domain channel impulse response of each transmitting antenna, t=1, 2, 3, 4. Since the processing of the channel impulse response of each transmitting antenna is the same, the following loop part is only processed for one transmitting antenna.

FIG. 1 is a block diagram of a channel estimation method for a scheme with unknown channel delay according to the present invention. Specifically, for a system with unknown channel multipath delay, the following loop processing is performed:

In the lth loop:

Step 1: Detect the current remaining channel impulse response h ^l (n) to find the current strongest path, where l is the number of cycles, and l=0 is the initial value. Determine whether the path is an effective path, if it is an effective path, enter the second step, otherwise end the loop.

The second step: restore the energy diffusion of the effective path, and multiply the complex amplitude β _l of the path by the normalized energy compensation function to obtain the energy expansion of the path g ^l (n) = β _l × Sinc _l (τ ).

${\mathrm{<span\; class="notranslate">\; Sinc</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\frac{\mathrm{<span\; class="notranslate">\; sin</span>}<span\; class="notranslate">\; (</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&tau;</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; \&pi;</span>}<span\; class="notranslate">\; \&times;</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&tau;</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; \&pi;</span>}}& \mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; \&NotEqual;</span>{\mathrm{<span\; class="notranslate">\; \&tau;</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}\\ \mathrm{<span\; class="notranslate">\; 1</span>}& \mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&tau;</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}\end{array}\right.$

Among them, τ _l represents the position of the lth effective path.

为构造信道冲激响应的初始值，其各采样点值均为零。然后转入第一步继续循环。Step 3: remove the energy extension of the strongest path from the remaining channel impulse response to obtain the current remaining channel impulse response h ^l (n)=h ^l-1 (n)-g ^l (n). Update construct channel impulse response, ${\tilde{h}}^l\left(\mathrm{no}\right)={\tilde{h}}^{l-1}\left(\mathrm{no}\right)+g^l\left(\mathrm{no}\right),$ in

In order to construct the initial value of the channel impulse response, the value of each sampling point is zero. Then turn to the first step to continue the cycle.

After the cycle ends, analyze the initial value h ⁰ (n) of the channel impulse response, and replace the sampling points smaller than the threshold value with the corresponding constructed channel estimation impulse response to obtain the final channel impulse response h(n). After Fourier transform, the final frequency domain response estimate H(n) is obtained.

Fig. 2 is a block diagram of a channel estimation method for a scheme with known channel delay according to the present invention. Specifically, for a system with unknown channel multipath delay, take the scheme based on the principle of parallel interference cancellation as an example, and the specific steps are as follows:

并令 ${\hat{h}}_{\mathrm{delay}}\left(n\right)={\hat{h}}_{\mathrm{delay}}^0\left(n\right).$ The first step: take out the effective path according to the known multipath delay, and obtain the initial channel estimation of the effective path

and order ${\hat{h}}_{\mathrm{delay}}\left(\mathrm{no}\right)={\hat{h}}_{\mathrm{delay}}^0\left(\mathrm{no}\right).$

乘以归一化的能量扩散补偿函数，并将各径的能量扩散相加，得到各径之间相互的干扰信号。The second step: the complex amplitude of each effective path

Multiply the normalized energy spread compensation function, and add the energy spread of each path to get the mutual interference signal between the paths.

得到更为准确的有效径信息。Step 3: Initial channel estimation from the effective path Subtract the interfering signal from , and use this result to update

Get more accurate effective path information.

Step 4: If the number of cycles reaches the set number, go to step 5; otherwise, go back to step 2.

经过傅立叶变换后得到最终的频域响应估计值H(n)。Step Five: Put Multiply by the normalization coefficient N/M=1.7067 to construct the final time-domain channel estimation impulse response

After Fourier transform, the final frequency domain response estimation value H(n) is obtained.

The simulation adopts a typical urban channel, and the number of paths is L=6. The carrier frequency is 2GHz, the system bandwidth is 5MHz, and the subcarrier spacing is 15kHz. When the signal-to-noise ratio (SNR) is 28dB, the mean square error (MSE) of the channel estimation method with the unknown channel delay scheme can reach 10 ^-3 , and the MSE of the channel estimation method with the known channel delay scheme can reach 4×10 ^{- 4} , while the MSE of the method without energy compensation is only 7×10 ^-2 .

The simulation results show that this method can well compensate the energy leakage caused by the virtual carrier and improve the accuracy of channel estimation.

Claims (10)

1. A channel estimation method for a multi-carrier system, comprising steps: (1) Extract the received signal at the pilot position at the receiving end, perform preliminary channel estimation, and obtain the initial channel time-domain impulse response of each antenna; (2) Input the obtained initial channel time-domain impulse response into the energy compensation module, select the effective path, and perform energy compensation for each effective path, eliminate the influence of the energy spread caused by the virtual carrier on the channel estimation, and obtain more Accurate time domain impulse response; (3) Perform Fourier transform to obtain frequency domain response estimation of the channel. 2. The method according to claim 1, wherein for a system with unknown multipath time delay of the channel, the method comprises the steps of: (1) For the nth symbol, first extract the pilot to calculate the initial value h ⁰ (n) of the time-domain channel estimation on each antenna; (2) Enter the energy compensation cycle module with unknown time delay, detect the current remaining channel impulse response h ^l (n) in the first cycle, to find the strongest path (l=0 is the initial value), and judge the path Whether it is an effective path, if the path is an effective path, the energy diffusion of the effective path is restored, that is, the complex amplitude of the path is multiplied by the normalized energy diffusion compensation function to obtain the energy diffusion g of the path ^l (n), and remove the strongest path and its energy spread from the remaining channel impulse response to obtain the current remaining channel impulse response h ^l (n)=h ^l-1 (n)-g ^l (n) , while updating the constructed channel impulse response, ${\tilde{h}}^l\left(\mathrm{no}\right)={\tilde{h}}^{l-1}\left(\mathrm{no}\right)+g^l\left(\mathrm{no}\right),$ Otherwise end the loop; (3) After the cycle ends, analyze the initial value h ⁰ (n) of the channel impulse response, replace the sampling points smaller than the threshold value with the corresponding constructed channel estimation impulse response, and obtain the final channel impulse response h(n ), and get the final frequency domain response estimate H(n) after Fourier transform. 3. The method according to claim 1, wherein for the system of known channel multipath time delay, the method comprises the steps of: calculating the initial value h ⁰ (n) of time domain channel estimation on each antenna; (2) Take out the effective path according to the known multipath time delay, and obtain the initial channel estimation of the effective path

and order ${\hat{h}}_{\mathrm{delay}}\left(\mathrm{no}\right)={\hat{h}}_{\mathrm{delay}}^0\left(\mathrm{no}\right),$ Then enter the energy compensation loop module with known time delay until the condition is met and the loop jumps out. In each loop, the complex amplitude of each effective path is multiplied by the normalized energy diffusion compensation function to construct the The mutual interference signals between the paths of , from the initial channel estimation of the effective path Remove the interfering signal in , and use this result to update Obtain more accurate effective path information; (3) Will

Multiply by the normalization coefficient to construct the final time-domain channel estimation impulse response After Fourier transform, the final frequency domain response estimation value H(n) is obtained. 4. The method according to claim 1, wherein the normalized energy spread compensation function is constructed according to the effective subcarrier position, the effective subcarrier position is filled into the same energy signal, and the remaining positions are set to zero, and the energy is carried out by inverse Fourier transform. The function obtained after normalization is the normalized energy spread compensation function. 5. The method according to claim 1, wherein the method is applicable to single-send-single-receive, single-send-multiple-receive or multiple-send-single-receive communication systems. 6. The method according to claim 2, wherein the set of all current strongest paths is the effective path experienced by each pair of transmitting antennas and receiving antennas, and its position is the channel multipath delay. 7. The method according to claim 2, wherein whether the energy or absolute amplitude maximum value of the current strongest path is greater than a set threshold value, if it is greater than the set threshold value, it is judged as an effective path, otherwise it is judged as an ineffective path . 8. The method of claim 3, wherein the normalization coefficient is determined based on the diffusion energy to energy ratio. . 9. The method according to claim 3, wherein eliminating interference signals between paths is based on a serial interference cancellation method, or based on a parallel interference cancellation method. 10. The method of claim 3, wherein the loop end condition is the current

with the previous cycle of

The difference is less than the set threshold or reaches the set number of cycles.

Images