WO2011009355A1

WO2011009355A1 - Adaptive equalizer, circuit and method for generating coefficients of equalization filter

Info

Publication number: WO2011009355A1
Application number: PCT/CN2010/074102
Authority: WO
Inventors: 周园
Original assignee: 中兴通讯股份有限公司
Priority date: 2009-07-23
Filing date: 2010-06-18
Publication date: 2011-01-27
Also published as: CN101616107A; CN101616107B

Abstract

An adaptive equalizer, a circuit and a method for generating the coefficients of the equalization filter, which are used in the multiple-input multiple-output (MIMO) system, are provided by the present invention. The method includes: using the sliding window filter to perform the sliding window filtering for the combined data to obtain the estimated value of the data; comparing the estimated value and a preset desired value so as to generate an error signal; using the interpolation filter to perform the interpolation on the data processed by the pulse shaping filter so as to generate an interpolation signal, wherein the length of the interpolation filter is the same as that of the sliding window filter, and calculating the error signal and the interpolation signal to obtain the coefficients of the equalization filter, wherein the length of the sliding window filter can be changed. The invention achieves the technical effects of simple implementation, low complexity for calculating the coefficients of the equalization filter, and adjustable adaptive convergence rate.

Description

Adaptive equalizer and equalization filter

TECHNICAL FIELD The present invention relates to the field of communications, and in particular, to a coefficient generation circuit and method for an adaptive equalizer and an equalization filter for a multiple-input multiple-output (MIMO) system. BACKGROUND OF THE INVENTION In order to meet the needs of future wireless communication, break through the bottleneck of system capacity and improve the reliability of communication, MIMO technology is gradually being introduced into wireless communication systems such as WLAN and 3G mobile communication systems. Existing receivers including adaptive equalizers for MIMO systems typically include: equalization receivers, RAKE receivers, G-RAKE receivers, SIC-GRAKE receivers, SIC-equalization receivers, and the like. As shown in FIG. 1, an adaptive equalizer for a MIMO system in the related art is generally obtained by processing pulse shaped reception data and pulse shaping, equalization filtering, rate conversion, and combined multiplexed data. The coefficient of the filter. In the related art, the coefficient generation circuit of the equalization filter for the MIMO system has a fixed frequency of the coefficients, so that the coefficient update frequency of the equalization filter is fixed and cannot be changed with the change of the channel environment, so that when the channel environment changes rapidly, The coefficient update frequency of the equalization filter changes slowly with respect to the channel environment, and the adaptive equalizer output data error is large. SUMMARY OF THE INVENTION An object of the present invention is to provide a coefficient generation circuit and method for an adaptive equalizer and an equalization filter for a MIMO system, which can solve a channel environment change caused by a fixed coefficient update frequency of an equalization filter in the related art. When it is fast, the output data has a large error and other technical problems. According to an aspect of the present invention, a coefficient generation circuit for an equalization filter of a multiple input multiple output system is provided, comprising: a sliding window filter for performing sliding window filtering on the combined data Obtaining an estimated value of the data; a data comparison module, configured to compare the estimated value with a predetermined expected value to generate an error signal; and an interpolation filter for interpolating data processed by the pulse shaping filter of the adaptive equalizer To generate an interpolation signal, the length of the interpolation filter is equal to the length of the sliding window filter; an adaptive operation module for calculating the error signal and the interpolation signal to obtain coefficients of the equalization filter, wherein the length of the sliding window filter variable. Preferably, the sliding window filter and the interpolation filter are both finite impulse response filters. Preferably, the adaptive operation module calculates the error signal and the interpolation signal by using a least squares method or a least mean square error algorithm to obtain coefficients of the equalization filter. Preferably, the length of the sliding window filter is one of: 16, 32, 64, 128 or 256. According to another aspect of the present invention, an adaptive equalizer for a multiple input multiple output system is provided, comprising: an equalization filter for equalizing data processed by a pulse shaping filter; a coefficient generation circuit And a coefficient generating circuit, comprising: a sliding window filter, configured to perform sliding window filtering on the combined data of the multiplexed signal combining module to obtain an estimated value of the data; and a data comparison module, configured to: Comparing the estimated value with a predetermined expected value to generate an error signal; an interpolation filter for interpolating the data processed by the pulse shaping filter to generate an interpolation signal, the length of the interpolation filter being equal to the length of the sliding window filter And an adaptive operation module, configured to calculate the error signal and the interpolation signal to obtain coefficients of the equalization filter, wherein the length of the sliding window filter is variable. Preferably, the adaptive equalizer further comprises: a pulse shaping filter for performing pulse shaping filtering on data received by the antenna of the MIMO system; and a rate conversion module for equalizing the pair of adaptive equalizers The data after the equalization of the filter is subjected to rate conversion; the multi-channel signal combining module is configured to combine the data converted by the rate conversion module. Preferably, the equalization filter, the sliding window filter, the pulse shaping filter, and the interpolation filter are all finite impulse response filters. Preferably, the adaptive operation module calculates the error signal and the interpolation signal by using a least squares method or a least mean square error algorithm to obtain coefficients of the equalization filter. According to still another aspect of the present invention, a coefficient generation method for an equalization filter of a multiple input multiple output system is provided, including: performing sliding window filtering on the combined data by using a sliding window filter to obtain an estimation of data a value; comparing the estimated value with a predetermined expected value to generate an error signal; Interpolating the data processed by the pulse shaping filter of the adaptive equalizer by using an interpolation filter to generate an interpolation signal, the length of the interpolation filter being equal to the length of the sliding window filter; calculating the error signal and the interpolation signal to obtain The coefficients of the equalization filter, wherein the length of the sliding window filter is variable. Preferably, calculating the error signal and the interpolation signal to obtain the coefficients of the equalization filter specifically comprises: calculating the error signal and the interpolation signal by using a least squares method or a least mean square error algorithm to obtain coefficients of the equalization filter. Preferably, the length of the sliding window filter is one of: 16, 32, 64, 128 or 256. By means of the above at least one technical solution of the present invention, the coefficients of the equalization filter are obtained by calculating the error signal and the interpolation signal respectively generated by the sliding window filter and the interpolation filter of variable length and equal length, so that the equalization filtering is performed. The coefficient update frequency of the device is variable, thereby avoiding the technical problem that the error of the adaptive equalizer output data is large due to the change of the coefficient update frequency of the equalization filter relative to the channel environment when the channel environment changes rapidly, and the reduction is achieved. The technical effect of adaptive equalizer output error and improved adaptive equalization convergence speed. The drawings are intended to provide a further understanding of the invention, and are intended to be a part of the description of the invention. In the drawings: FIG. 1 is a schematic structural diagram of an adaptive equalizer for a MIMO system in the related art; FIG. 2 is a coefficient generation circuit of an equalization filter for a multiple input multiple output system according to a first embodiment of the present invention; Figure 3 is a block diagram of an adaptive equalizer for a multiple input multiple output system in accordance with a second embodiment of the present invention; Figure 4 is an equalization filter for a multiple input multiple output system in accordance with a third embodiment of the present invention; FIG. 5 is a schematic diagram showing the structure of an adaptive equalizer for a WCDMA communication system with two inputs and two outputs according to a fourth embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The preferred embodiments of the present invention are described with reference to the accompanying drawings. In the following description, numerous specific details are set forth However, it is apparent that the present invention may be practiced without these specific details. Further, in the case of no conflict, that is, without departing from the spirit and scope of the appended claims, the following embodiments And various details in the embodiments can be variously combined. First Embodiment FIG. 2 is a block diagram of a coefficient generation circuit for an equalization filter of a multiple input multiple output system according to a first embodiment of the present invention. As shown in FIG. 2, the coefficient generation circuit 200 for the equalization filter of the multiple input multiple output system according to the first embodiment of the present invention includes: a sliding window filter 202 for performing sliding window filtering on the combined data. Obtaining an estimated value of the data; the data comparison module 204 is configured to compare the estimated value with a predetermined expected value to generate an error signal; and the interpolation filter 206 is configured to perform data processed by the pulse shaping filter of the adaptive equalizer Interpolating to generate an interpolation signal, the length of the interpolation filter is equal to the length of the sliding window filter; an adaptive operation module 208, configured to calculate the error signal and the interpolation signal to obtain coefficients of the equalization filter, wherein the sliding window filter The length is variable. The coefficient generation circuit for the equalization filter of the multiple input multiple output system according to the first embodiment of the present invention performs the error signal and the interpolation signal respectively generated by the sliding window filter and the interpolation filter which are variable in length and equal in length The operation is performed to obtain the coefficients of the equalization filter, so that the coefficient update frequency of the equalization filter is variable, thereby avoiding the adaptive equalization caused by the slow change of the coefficient update frequency of the equalization filter relative to the channel environment when the channel environment changes rapidly. The technical problem of large output error of the device achieves the technical effect of reducing the output error of the adaptive equalizer and improving the convergence speed of the adaptive equalization. For example, when the channel environment changes better than the 'I only or the channel environment is better, the coefficient generation circuit of the equalization filter for the MIMO system of the embodiment of the present invention can increase the sliding window filter and the interpolation filter. The length of the coefficient update frequency of the equalization filter is slowed down, thereby reducing the computational complexity of the entire system without affecting the reception performance; when the channel environment changes rapidly or the channel environment is poor, according to an embodiment of the present invention The coefficient generation circuit for the equalization filter of the multiple input multiple output system can make the equalization filter by reducing the length of the sliding window filter and the interpolation filter The coefficient update frequency becomes faster, so that the adaptive equalizer can better track the channel change, thereby reducing the output data error of the adaptive equalizer; in short, by changing the length of the sliding window filter and the interpolation filter, according to the present The coefficient generation circuit of the equalization filter for a multiple input multiple output system of the inventive embodiment is capable of achieving an optimum balance between performance and computational load. Preferably, the sliding window filter and the interpolation filter are all finite impulse responses (Finite Impulse

Response, FIR) Filter. The sliding window filter, the pulse shaping filter and the interpolation filter are all FIR filters, so that the data path structures of the respective antennas in the MIMO system are identical, and the control thereof is basically similar, thereby achieving the purpose of simplifying the system structure. Preferably, the adaptive operation module calculates the error signal and the interpolation signal by using a least squares method or a least mean square error algorithm to obtain coefficients of the equalization filter.最小Using the least squares method or the least mean square error algorithm can achieve good reception performance with low computational complexity. If the least squares adaptive algorithm is used, the system mainly includes multiplication and addition, avoiding such as G-RAKE receiver. Complex operations of matrix inversion in related techniques reduce computational complexity. Alternatively, the error signal and the interpolated signal may be calculated using a constant envelope algorithm to obtain coefficients of the equalization filter. Preferably, the length of the sliding window filter is one of: 16, 32, 64, 128 or 256. The length of the sliding window filter can vary with the channel environment. For example, when the channel environment changes rapidly or the channel environment is poor, 16 chips can be taken as the length of the sliding window filter, due to the length of the interpolation filter. The length of the sliding window filter is equal, so the length of the interpolation filter is also 16 chips. When the channel environment changes slowly or the channel environment is good, 256 chips can be taken as the length of the sliding window filter. Since the length of the interpolation filter is equal to the length of the sliding window filter, the length of the interpolation filter is also 256 chips. In addition, since the processing unit of the coefficient generation circuit for the equalization filter of the multiple input multiple output system according to the embodiment of the present invention is a chip, compared with the coefficient generation circuit of the equalization filter in which the symbol is a processing unit in the related art , with higher processing precision, resulting in lower block error rate. The coefficient generation circuit for the equalization filter of the multiple input multiple output system according to the first embodiment of the present invention performs the error signal and the interpolation signal respectively generated by the sliding window filter and the interpolation filter which are variable in length and equal in length Calculate the coefficients of the equalization filter to achieve a reduction Adapt to the equalizer output error and improve the technical performance of the system. Second Embodiment FIG. 3 is a block diagram of an adaptive equalizer for a multiple input multiple output system according to a second embodiment of the present invention. As shown in FIG. 3, an adaptive equalizer for a multiple input multiple output system according to a second embodiment of the present invention includes: an equalization filter 302 for equalizing data processed by a pulse shaping filter; coefficient generation The circuit 200 is configured to provide a coefficient to the equalization filter, and the coefficient generation circuit 200 includes: a sliding window filter 202, configured to perform sliding window filtering on the combined data of the multiplexed signal combining module to obtain an estimated value of the data; The module 204 is configured to compare the estimated value with a predetermined expected value to generate an error signal, and the interpolation filter 206 is configured to perform interpolation on the data processed by the pulse shaping filter to generate an interpolation signal, and the length and the sliding of the interpolation filter The length of the window filter is equal; the adaptive operation module 208 is configured to calculate the error signal and the interpolation signal to obtain coefficients of the equalization filter, wherein the length of the sliding window filter is variable. An adaptive equalizer for a multiple input multiple output system according to a second embodiment of the present invention is obtained by computing an error signal and an interpolation signal respectively generated by a sliding window filter and an interpolation filter of variable length and equal length. The coefficients of the equalization filter are such that the coefficient update frequency of the equalization filter is variable, thereby avoiding the adaptive equalizer output data due to the slow change of the coefficient update frequency of the equalization filter relative to the channel environment when the channel environment changes rapidly. The technical problem of large error is to achieve the technical effect of reducing the output error of the adaptive equalizer and improving the convergence speed of the adaptive equalization. For example, when the channel environment changes better than the 'I only or the channel environment is better, the coefficient generation circuit of the equalization filter for the MIMO system of the embodiment of the present invention can increase the sliding window filter and the interpolation filter. The length of the coefficient update frequency of the equalization filter is slowed down, thereby reducing the computational complexity of the entire system without affecting the reception performance; when the channel environment changes rapidly or the channel environment is poor, according to an embodiment of the present invention The coefficient generation circuit for the equalization filter of the multiple input multiple output system can make the coefficient update frequency of the equalization filter faster by reducing the length of the sliding window filter and the interpolation filter, thereby making the adaptive equalizer better Tracking channel changes, thereby reducing the output data error of the adaptive equalizer; in summary, by varying the length of the sliding window filter and the interpolation filter, the equalization filter for a multiple input multiple output system according to an embodiment of the present invention The coefficient generation circuit provides the best balance between performance and computational load. Preferably, the adaptive equalizer further comprises: a pulse shaping filter 304 for performing pulse shaping filtering on data received by the antenna of the multiple input multiple output system; and a rate conversion module 306 for equalizing the adaptive equalizer The data after the equalization of the filter is subjected to rate conversion; the multi-path signal combining module 308 is configured to combine the data converted by the rate conversion module. In the embodiment of the present invention, the number of pulse shaping filters is equal to the number of receiving antennas in the MIMO system; the number of equalization filters is equal to the number N of receiving antennas in the MIMO system multiplied by the number M of transmitting antennas, that is, each receiving The antenna is connected to M equalization filters. Preferably, the equalization filter, the sliding window filter, the pulse shaping filter, and the interpolation filter are all FIR filters. The equalization filter, the sliding window filter, the pulse shaping filter, and the interpolation filter are all FIR filters, so that the data path structures of the respective antennas in the MIMO system are identical, and the control thereof is basically similar, thereby simplifying the system structure. the goal of. In addition, since the filters of the respective sections are FIR filters, the existing IP CORE system can be implemented, so that the system according to the embodiment of the present invention can be applied to implementations based on different manufacturers, different types of FPGAs or ASICs. Alternatively, the equalization filter can also use an infinite impulse response filter. Preferably, the adaptive operation module calculates the error signal and the interpolation signal by using a least squares method or a least mean square error algorithm to obtain coefficients of the equalization filter.最小Using the least squares method or the least mean square error algorithm can achieve good reception performance with low computational complexity. If the least squares adaptive algorithm is used, the system mainly includes multiplication and addition, avoiding such as G-RAKE receiver. Complex operations of matrix inversion in related techniques reduce computational complexity. Alternatively, the error signal and the interpolated signal may be calculated using a constant envelope algorithm to obtain coefficients of the equalization filter. An adaptive equalizer for a multiple input multiple output system according to a second embodiment of the present invention is obtained by computing an error signal and an interpolation signal respectively generated by a sliding window filter and an interpolation filter of variable length and equal length. The coefficients of the equalization filter are used to achieve the technical effect of reducing the output error of the adaptive equalizer, improving the convergence speed of the adaptive equalization, and improving the performance of the system. Third Embodiment FIG. 4 is a flowchart of a coefficient generation method of an equalization filter for a multiple input multiple output system according to a third embodiment of the present invention. As shown in FIG. 4, a coefficient generating method for an equalization filter of a multiple input multiple output system according to a third embodiment of the present invention includes the following steps: Step S402, using a sliding window filter to perform sliding window on the combined data. Filtering to obtain an estimated value of the data; Step S404, comparing the estimated value with a predetermined expected value to generate an error signal; Step S406, interpolating the data processed by the pulse shaping filter of the MIMO system by using an interpolation filter To generate an interpolation signal, the length of the interpolation filter is equal to the length of the sliding window filter; in step S408, the error signal and the interpolation signal are calculated to obtain coefficients of the equalization filter, wherein the length of the sliding window filter is variable. A coefficient generation method for an equalization filter of a multiple input multiple output system according to a third embodiment of the present invention performs an error signal and an interpolation signal respectively generated by a sliding window filter and an interpolation filter having variable lengths and equal lengths The operation is performed to obtain the coefficients of the equalization filter, so that the coefficient update frequency of the equalization filter is variable, thereby avoiding the adaptive equalization caused by the slow change of the coefficient update frequency of the equalization filter relative to the channel environment when the channel environment changes rapidly. The technical problem of large output error of the device achieves the technical effect of reducing the output error of the adaptive equalizer. Preferably, calculating the error signal and the interpolation signal to obtain the coefficients of the equalization filter specifically comprises: calculating the error signal and the interpolation signal by using a least squares method or a least mean square error algorithm to obtain coefficients of the equalization filter.最小Using the least squares method or the least mean square error algorithm can achieve good reception performance with low computational complexity. If the least squares adaptive algorithm is used, the system mainly includes multiplication and addition, avoiding such as G-RAKE receiver. Complex operations of matrix inversion in related techniques reduce computational complexity. Alternatively, the error signal and the interpolated signal may be calculated using a constant envelope algorithm to obtain coefficients of the equalization filter. Preferably, the length of the sliding window filter is one of: 16, 32, 64, 128 or 256. The length of the sliding window filter can vary with the channel environment. For example, when the channel environment changes rapidly or the channel environment is poor, 16 chips can be taken as the length of the sliding window filter, due to the length of the interpolation filter. It is equal to the length of the sliding window filter, so the length of the interpolation filter is also 16 chips at this time; when the channel environment changes slowly or the channel environment is good, 256 chips can be taken. As the length of the sliding window filter, since the length of the interpolation filter is equal to the length of the sliding window filter, the length of the interpolation filter is also 256 chips at this time. A coefficient generation method for an equalization filter of a multiple input multiple output system according to a third embodiment of the present invention performs an error signal and an interpolation signal respectively generated by a sliding window filter and an interpolation filter having variable lengths and equal lengths The operation is performed to obtain the coefficients of the equalization filter, thereby achieving the technical effect of reducing the output error of the adaptive equalizer, improving the convergence speed of the adaptive equalization, and improving the performance of the system. Fourth Embodiment Hereinafter, a multi-input and multi-output according to a fourth embodiment of the present invention will be described with reference to a structural diagram of an adaptive equalizer for a WCDM A communication system in which two inputs and two outputs are shown in FIG. The coefficient generation method of the system's equalization filter is described in detail. A coefficient generating method for an equalizing filter for a multiple input multiple output system according to a fourth embodiment of the present invention includes the following steps: Step 1: Receive antennas 1 and 2 of a receiver of a WCDMA communication system receive data and respectively transmit Into the FIR filters S 1 and S2 for pulse shaping filtering; Step 2: After filtering the data received by the antenna 1, it is respectively sent to two equalization filters F11 and F21 connected to the pulse shaping filter, and the antenna 2 After the received data is filtered, it is respectively sent to two equalization filters F12 and F22 connected to the pulse shaping filter, and the equalization filter coefficients may be arbitrarily set initial values; Step 3: F11 and F12 equalization The subsequent data is merged after data rate conversion. The data after F21 and F22 equalization are combined after the rate conversion unit is downsampled to the normal chip rate of 3.84 Mcps. Step 4: The combined two channels of data are respectively in the sliding window filter. Sliding window filtering is performed in C1 to generate an estimated value of the data; the length of the sliding window filter may be 16, 32, 64, 128 or 256 according to the channel environment, and the sliding window filter may be pre-generated. The number is read out in sequence when used; Step 5: The estimated value of the generated data is compared with the expected signal dl to generate two error signals el and e2; Step 6: Antenna data filtered by the pulse shaping filter The equalization filter is also sent to the interpolation filter II for interpolation to generate four-way interpolation signals ul l, ul2, u21, u22, and the interpolation filter also uses an FIR filter, the filter length is variable, and The sliding window filter has the same length, and the coefficients of the sliding window filter can be generated in advance, and are sequentially read out when used; Step 7: The error signal el and the corresponding interpolation signals u11 and ul2 are sent to the corresponding adaptive operation unit W1, and the corresponding equalizers F11 and F12 coefficients are calculated using the LMS or RLS algorithm; the error signal e2 and the corresponding interpolation signal u21 are obtained. And u22 is sent to the corresponding adaptive operation unit, and the corresponding equalizer F21 and F22 coefficients are calculated by using the LMS or RLS algorithm; Step 8: updating the corresponding equalization filter coefficients according to the operation result; Step 9: repeating step 4 to gather 1-8 To process the next data. A coefficient generation method for an equalization filter for a multiple input multiple output system according to a fourth embodiment of the present invention performs an error signal and an interpolation signal respectively generated by a sliding window filter and an interpolation filter having variable lengths and equal lengths The operation is performed to obtain the coefficients of the equalization filter, thereby achieving the technical effect of reducing the output error of the adaptive equalizer and improving the performance of the system. With the above at least one aspect of the present invention, the coefficients of the equalization filter are obtained by calculating the error signal and the interpolation signal respectively generated by the sliding window filter and the interpolation filter of variable length and equal length, thereby achieving reduction Technical effect of adaptive equalizer output error, improved adaptive equalization convergence speed, and improved system performance; in summary, by changing the length of the sliding window filter and the interpolation filter, the multi-input multi-output system according to an embodiment of the present invention The equalization filter's coefficient generation circuit provides the best balance between performance and computational load. Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general-purpose computing device, which can be concentrated on a single computing device or distributed over a network composed of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device, such that they may be stored in the storage device by the computing device, or they may be separately fabricated into individual integrated circuit modules, or they may be Multiple modules or steps are made into a single integrated circuit module. Thus, the invention is not limited to any specific combination of hardware and software. The above is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the scope of the present invention are intended to be included within the scope of the present invention.

Claims

A coefficient generation circuit for an equalization filter of a multiple input multiple output system, comprising:

a sliding window filter, configured to perform sliding window filtering on the combined data to obtain an estimated value of the data;

a data comparison module, configured to compare the estimated value with a predetermined expected value to generate an error signal;

An interpolation filter, configured to interpolate data processed by the pulse shaping filter of the adaptive equalizer to generate an interpolation signal, the length of the interpolation filter being equal to the length of the sliding window filter;

An adaptive operation module, configured to calculate the error signal and the interpolation signal to obtain coefficients of the equalization filter,

Wherein, the length of the sliding window filter is variable. The coefficient generation circuit according to claim 1, wherein the sliding window filter and the interpolation filter are both finite impulse response filters. The coefficient generation circuit according to claim 1, wherein the adaptive operation module calculates the error signal and the interpolation signal by using a least squares method or a least mean square error algorithm to obtain the equalization filter. Coefficient. The coefficient generation circuit according to claim 1, wherein the length of the sliding window filter is one of: 16, 32, 64, 128 or 256. An adaptive equalizer for a multiple input multiple output system, comprising:

An equalization filter for equalizing the data processed by the pulse shaping filter; a coefficient generation circuit, configured to provide a coefficient to the equalization filter, where the coefficient generation circuit includes:

a sliding window filter, configured to perform sliding window filtering on the combined data of the multiplexed signal combining module to obtain an estimated value of the data; a data comparison module, configured to compare the estimated value with a predetermined expected value to generate an error signal;

An interpolation filter, configured to interpolate data processed by the pulse shaping filter to generate an interpolation signal, the length of the interpolation filter being equal to a length of the sliding window filter;

Wherein, the length of the sliding window filter is variable.

The adaptive equalizer according to claim 5, further comprising:

The pulse shaping filter is configured to perform pulse shaping filtering on data received by an antenna of the MIMO system;

a rate conversion module, configured to rate convert the data that is equalized by the equalization filter of the adaptive equalizer;

The multiplex signal combining module is configured to merge data converted by the rate conversion module.

The adaptive equalizer according to claim 6, wherein the equalization filter, the sliding window filter, the pulse shaping filter, and the interpolation filter are all finite impulse response filters. .

8. The adaptive equalizer according to claim 5, wherein the adaptive operation module calculates the error signal and the interpolation signal by using a least squares method or a least mean square error algorithm to obtain the The coefficient of the equalization filter.

A coefficient generation method for an equalization filter of a multiple input multiple output system, comprising:

Sliding window filtering is performed on the combined data by using a sliding window filter to obtain an estimated value of the data;

Comparing the estimated value with a predetermined expected value to generate an error signal; interpolating the data processed by the pulse shaping filter of the adaptive equalizer with an interpolation filter to generate an interpolation signal, the length of the interpolation filter being The sliding window filters are of equal length; Calculating the error signal and the interpolation signal to obtain coefficients of the equalization filter,

Wherein, the length of the sliding window filter is variable.

The method for generating a coefficient according to claim 9, wherein the calculating the error signal and the interpolation signal to obtain the coefficient of the equalization filter comprises:

The error signal and the interpolated signal are calculated using a least squares method or a least mean square error algorithm to obtain coefficients of the equalization filter.

The coefficient generation method according to claim 9, wherein the sliding window filter has a length of one of: 16, 32, 64, 128 or 256.