CN101931593B - Channel equalization method and device - Google Patents

Channel equalization method and device Download PDF

Info

Publication number
CN101931593B
CN101931593B CN 200910108237 CN200910108237A CN101931593B CN 101931593 B CN101931593 B CN 101931593B CN 200910108237 CN200910108237 CN 200910108237 CN 200910108237 A CN200910108237 A CN 200910108237A CN 101931593 B CN101931593 B CN 101931593B
Authority
CN
Grant status
Grant
Patent type
Prior art keywords
channel
equalization
method
device
channel equalization
Prior art date
Application number
CN 200910108237
Other languages
Chinese (zh)
Other versions
CN101931593A (en )
Inventor
周化雨
汪开龙
汪赟
Original Assignee
Tcl集团股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Grant date

Links

Abstract

本发明公开了一种信道均衡方法及其装置,其包括以下步骤:对接收信号进行信道匹配,获得待处理信号;对所述待处理信号进行最小均方误差-块判决反馈均衡处理,获得系列检测结果;抽取所述系列检测结果中的最后一个输出作为已检测符号;将所述已检测符号作为干扰,从接收信号中消除,生成新的接收信号。 The present invention discloses a method and apparatus for channel equalization comprising the steps of: receiving a channel signal matching signal to be processed is obtained; the signal to be processed minimum mean square error - block decision feedback equalization processing to obtain series a detection result; extract output of the last in the series as a detection result detected symbol; the symbol detected as the interference eliminated from the received signal, generate a new received signal. 本发明能够对MMSE-IC信道均衡方法进行优化处理,获得额外增益。 The present invention can be optimized for the MMSE-IC channel equalization method, to obtain additional gain.

Description

一种信道均衡方法及其装置 A channel equalizing method and apparatus

技术领域 FIELD

[0001] 本发明属于通信系统联合检测技术领域,尤其涉及的是丽SE-1C (Minimum MeanSquare Error-1nterference Cancellation,最小均方误差-干扰消除方法)信道均衡方法的一种优化实现方法。 [0001] The present invention belongs to the field of communication systems joint detection technology, and particularly to the Korea SE-1C (Minimum MeanSquare Error-1nterference Cancellation, minimum mean square error - interference elimination method) for optimizing channel equalization method of implementation.

背景技术 Background technique

[0002] 现有技术中,突发模式(GSM,TDD-CDMA等)的通信系统也称为块传输系统。 [0002] In the prior art, burst-mode (GSM, TDD-CDMA etc.) communication system is also called block transmission system. 当信号带宽大于信道的相干带宽,就会产生符号间干扰(ISI, Inter Symbol Interference),这样的信道称为ISI信道或频率选择性衰落信道。 When the signal bandwidth is greater than the coherence bandwidth of the channel, is generated inter-symbol interference (ISI, Inter Symbol Interference), ISI Such a channel is called channel or frequency selective fading channel. 当宽带通信越来越普遍,通信中的ISI问题越来越突出,是制约接收机性能的重要因素。 As more and more common broadband communications, ISI communication problems more prominent, it is an important factor restricting receiver performance. 一般对抗ISI问题的方法是均衡。 General approach is balanced against the ISI problem. 块传输系统有其特有的矩阵形式均衡方法,或称为块均衡方法。 Block transmission system has its unique matrix equalization method, or a method referred to as block equalization.

[0003] 常用的块均衡方法有最小均方误差-块线性均衡(MMSE-BLE,Minimum MeanSquare Error-Block Linear Equalization)和最小均方误差-块判决反馈均衡(MMSE-BDFE, Minimum Mean Square Error-Block Decision Feedback Equalization)。 [0003] Frequently block equalization MMSE methods - block linear equalizer (MMSE-BLE, Minimum MeanSquare Error-Block Linear Equalization) and minimum mean square error - block decision feedback equalizer (MMSE-BDFE, Minimum Mean Square Error- Block Decision Feedback Equalization). 常用于V-BLAST下的MMSE-1C均衡方法也可应用于块传输系统的均衡,它是一种MMSE滤波和串行干扰消除相结合的方法。 Commonly used in the V-BLAST equalization MMSE-1C equalization method may be applied to the transmission system block, which is a method of combining the MMSE filter and interference cancellation.

[0004] 由于时分双工-码分多址(Time Division Duplex-Code Division MultipleAccess, TDD-CDMA)也是一种块传输系统,因此也可采用块均衡方法。 [0004] Since a time division duplex - code division multiple access (Time Division Duplex-Code Division MultipleAccess, TDD-CDMA) transmission system is also a block, so the block equalization method may be employed. 对于TDD-CDMA系统将矩阵形式的均衡方法和数据检测结合起来,这就是所谓的联合检测算法。 For the TDD-CDMA system and a matrix form equalization data detection method of combining which is called joint detection algorithm. 联合检测技术既能克服远近效应,又能减轻由多径信道产生的符号间干扰和多址干扰(MAI,Multiple-access interference)。 Both joint detection technique to overcome near-far effect, but also to reduce inter-symbol generated by the multipath channel interference and multiple access interference (MAI, Multiple-access interference). CDMA 系统可以分为异步CDMA 和同步CDMA。 A CDMA system may be divided into synchronous and asynchronous CDMA CDMA. 异步CDMA一般不能保证扩频码之间在不对齐情况下的正交性,因此本质上就会有MAI。 Asynchronous CDMA aligned generally can not be guaranteed without a case where the orthogonality between spreading codes, so there will be essentially MAI. 同步CDMA系统中的扩频码一般是正交的,在非频率选择选择性衰落信道中,并无ISI,也无MAI。 Synchronous CDMA system spreading codes are typically orthogonal to choose selective fading channel, the ISI is not a non-frequency, no MAI. 但在频率选择选择性衰落信道中,多个路径传播的信号产生了ISI,并且破坏了扩频码之间的正交性,引起了MAI。 However, in a frequency selective fading channel selection, a plurality of signal propagation produces the ISI, and the destruction of orthogonality between spreading codes, causing the MAI. 宽带系统实际中经历的信道是频率选择性衰落的,因此ISI和MAI是影响宽带CDMA的性能的主要因素。 Broadband system actually experienced in the channel is frequency selective fading, ISI and MAI so are the main factors affecting the performance of the wideband CDMA.

[0005] 实际上,MMSE-1C中的MMSE滤波就是一个MMSE-BLE过程。 [0005] Indeed, MMSE filtering MMSE-1C in MMSE-BLE is a process. 通常并没有对MMSE-1C中的MMSE-BLE过程进行优化,所以输出结果不能获得额外的增益,抗噪声能力较差,需要进一步的改进。 And usually not optimized for MMSE-BLE process of MMSE-1C, the output result can not be obtained additional gain, poor noise immunity, further improvement is required.

发明内容 SUMMARY

[0006] 本发明的目的在于提供一种信道均衡方法及其装置,其能够对丽SE-1C信道均衡方法进行优化处理,获得额外增益,且提高抗噪声能力。 [0006] The object of the present invention is to provide a method and apparatus for channel equalization, which is capable of Korea SE-1C method of optimizing channel equalization process to obtain additional gain and improve noise immunity.

[0007] 为了实现上述目的,本发明采用如下技术方案: [0007] To achieve the above object, the present invention adopts the following technical solution:

[0008] 本发明提供了一种信道均衡方法,其包括以下步骤: [0008] The present invention provides a channel equalization method, comprising the steps of:

[0009] A、对接收信号进行信道匹配,获得待处理信号;[0010] B、对所述待处理信号进行最小均方误差-块判决反馈均衡,获得系列检测结果; [0009] A, the received signal channel matched to give signal to be processed; [0010] B, the signal to be processed to the minimum mean square error - block decision feedback equalizer, the detection result obtained series;

[0011] C、抽取所述系列检测结果中的最后一个输出作为已检测符号; [0011] C, to extract the last output of the series as a detection result detected symbols;

[0012] D、将所述已检测符号作为干扰,从接收信号中消除,生成新的接收信号。 [0012] D, as the detected symbol interference, eliminating from the received signal, generate a new received signal.

[0013] 所述的方法,其中,还包括一判断步骤:判断是否检测完所述接收信号中的所有符号,若没有,则返回步骤A重复执行均衡过程,直到检测出所有符号;否则,输出结果。 [0013] The method, wherein the determining further comprises a step of: determining whether all tests are completed, the symbols of the received signal, if not, returns to step A balancing process is repeatedly performed until all the symbols detected; otherwise, the output result.

[0014] 所述的方法,其中,所述步骤D包括: [0014] The method, wherein the step D comprises:

[0015] 对所述已检测符号进行判决,选择该已检测符号最可能对应的发射符号; [0015] The detected symbols for the decision to select the detected transmitted symbol corresponding to the most likely symbol;

[0016]把判决得到的符号作为干扰,将该干扰从所述接收信号中去除,获得所述新的接收信号。 [0016] The decision symbols obtained as the interference, the interference is removed from the received signal, said received signal newly obtained.

[0017] 所述的方法,其中,从所述接收信号中除去所述干扰的过程按照下式进行: [0017] The method, wherein the received signal from the interference removing process is carried out according to the following formula:

[0018] [0018]

Figure CN101931593BD00051

[0019] 其中,A表示接收信号;ri"表示所述新的接收信号;(H)i表示H的第i列,这里的H表示信道冲激响应矩阵;1;表示所述干扰。 [0019] where, A denotes a received signal; ri "represents the new received signal; (H) i represents the i-th column of H, where H represents a channel impulse response matrix; 1; represents the interference.

[0020] 所述的方法,其中,所述最小均方误差-块判决反馈均衡方法中,对待处理信号依次进行前向替换、后向替换处理;并在所述后向替换处理中添加所述块判决反馈处理。 [0020] The method, wherein the minimum mean square error - block decision feedback equalization process, treated sequentially processed signal prior to the replacement process to the post-replacement; and adding to the replacement process in the later block decision feedback process.

[0021] 所述的方法,其中,在所述后向替换处理中添加所述块判决反馈处理的过程按照以下公式进行: [0021] The method, wherein the block is added to the decision feedback process after the replacement processing procedure in accordance with the following formula:

Figure CN101931593BD00052

[0024] 其中L表示所述系列检测结果;zn表示进行所述前向替换后的处理结果; < 是 [0024] wherein L represents a series of the detection result; Zn represents a result of the forward processing after the replacement; <is

第η个元素为I的单位向量;L为对角元素为I的下三角矩阵;1表示单位矩阵;□ {.}是判决算子#是一个块中的符号数。 The first element η is a unit vector I; L is the diagonal elements of the triangular matrix I; 1 denotes a unit matrix; {.} □ # operator is the decision is the number of symbols in a block.

[0025] 本发明还提供了一种信道均衡装置,包括:信道匹配模块,用于对接收信号进行信道匹配、并获得待处理信号;中间处理模块,用于对所述待处理信号进行最小均方误差-块判决反馈均衡,并获得系列检测结果;选择模块,用于对所述系列检测结果进行选择,并抽取最后一个输出作为已检测符号;干扰消除处理模块,用于将已检测符号作为干扰,从接收信号中消除,并生成新的接收信号;及判断模块,用于判断是否检测完所述接收信号中的所有符号,若没有则重新启动所述信道匹配模块,否则输出结果; [0025] The present invention also provides a channel equalization apparatus, comprising: a channel matching module configured to match the received signal channel, and the signal to be processed is obtained; intermediate processing module, the signal to be processed for the minimum average square error - block decision feedback equalizer, and a detection result obtained series; selection module configured to select the series of the detection result, and extracts the last output as detected symbol; interference elimination processing module, for the detected symbols as interference, to eliminate from the received signal, and generates a new received signal; and a determining module, for determining whether the reception completion detecting signal for all symbols, if not then restart the channel matched module, or output;

[0026] 所述的装置,其中,所述干扰消除处理模块包括:判决单元,用于对所述已检测符号进行判决,选择该已检测符号最可能对应的发射符号;及剔除单元,用于把判决得到的符号作为干扰,将该干扰从所述接收信号中去除,获得所述新的接收信号; Means [0026], wherein the interference elimination processing module comprising: a decision unit for the decision detected symbols, selecting the detected symbol transmitted symbol corresponding to the most likely; and expulsion unit, for the decision symbols obtained as the interference, the interference is removed from the received signal, said received signal newly obtained;

[0027] 其中,所述干扰消除处理模块接收信号中除去所述干扰的过程按照下式进行: [0027] wherein the interference cancellation process in the interference removal processing module receives a signal using the following formula:

Figure CN101931593BD00053

其中,ri表示输入的符号;ri"表示所述新的接收信号;(H)i表示H的第i列,这里的H表示信道冲激响应矩阵;J;表示所述干扰; Where, ri represents the input symbols; ri "represents the new received signal; (H) i represents the i th column of H, where H represents a channel impulse response matrix; J; represents the interference;

[0028] 所述的装置,其中,所述中间处理模块包括:前向处理单元,用于对所述待处理信号进行前向替换处理;后向处理单元,用于对所述前向替换处理结果进行后向替换处理,并添加判决反馈处理。 Means [0028], wherein the intermediate processing module comprises: a forward processing unit for the signal to be processed prior to the replacement process carried out; backward processing unit for the replacement process forward results are processed to replace, and add the decision feedback process.

[0029] 发明效果:本发明将MMSE-1C中的MMSE-BLE过程改成MMSE-BDFE过程对MMSE-1C信道均衡方法进行优化,获得了一种最小均方误差-干扰消除及块判决反馈相结合的均衡方法。 [0029] Effect of the Invention: The present invention MMSE-BLE process MMSE-1C MMSE-BDFE into the process MMSE-1C optimizing channel equalization method, to obtain a minimum mean square error - block decision feedback and interference cancellation phase balanced incorporated methods. MMSE-BDFE过程中解出的最后一个输出具有额外的判决增益,因此改进后的MMSE-1C有较好的性能。 MMSE-BDFE process the output of the last solution has the additional decision gain, improved MMSE-1C have better performance.

附图说明 BRIEF DESCRIPTION

[0030] 图1为本发明的方法流程图; [0030] FIG. 1 is a flowchart of a method of the present invention;

[0031 ] 图2为本发明实施例的方法流程图; [0031] FIG. 2 is a flowchart of a method embodiment of the present invention;

[0032] 图3为K=8时,MMSE-BDFE与MMSE-1C-BDFE的抗噪声能力效果图; [0032] FIG. 3 is 8, MMSE-BDFE and MMSE-1C-BDFE noise immunity renderings K =;

[0033] 图4为Κ=16时,MMSE-BDFE与MMSE-1C-BDFE的抗噪声能力效果图。 [0033] FIG. 4 is a Κ = 16 when, MMSE-BDFE and MMSE-1C-BDFE effect of noise immunity FIG.

具体实施方式 detailed description

[0034] 如图1所示,本发明提供的信道均衡方法,包括以下步骤: [0034] As shown in FIG 1, the channel equalization method of the present invention is provided comprising the steps of:

[0035] 步骤100,对接收信号进行信道匹配,获得待处理信号; [0035] Step 100, the received signal channel matched to give signal to be processed;

[0036] 步骤200,对所述待处理信号进行最小均方误差-块判决反馈均衡,获得系列检测结果; [0036] Step 200, the signal to be processed to the minimum mean square error - block decision feedback equalizer, the detection result obtained series;

[0037] 步骤300,抽取所述系列检测结果中的最后一个输出作为已检测符号; [0037] Step 300, extracting the last output as a series of detection results detected symbols;

[0038] 步骤400,将所述已检测符号作为干扰,从接收信号中消除,生成新的接收信号。 [0038] Step 400, the detected symbols as the interference eliminated from the received signal, generate a new received signal.

[0039] 本发明通过将MMSE-1C中的MMSE-BLE过程改成MMSE-BDFE过程对MMSE-1C信道均衡方法进行优化,获得了一种最小均方误差、干扰消除及块判决反馈相结合的均衡方法。 [0039] The present invention will MMSE-BLE process MMSE-1C into MMSE-BDFE process MMSE-1C optimizing channel equalization method, to obtain a minimum mean square error, and the block decision feedback interference cancellation combining equalization method. 由于丽SE-BDFE过程中解出的最后一个输出具有额外的判决增益,因此改进后的丽SE-1C有较好的性能,抗干扰性强,以下结合图2的具体实施例详细说明上述四个步骤的具体实现方法以及本发明所带来的技术效果。 Since the final output of Korea SE-BDFE solving process has the additional decision gain, Korea SE-1C improved better performance, interference is strong, specific embodiments in conjunction with the following detailed description of Figure 2 above four specific method steps of the present invention and the technical effects brought about.

[0040] 假设接收信号为r,对应的发射符号包含N个符号,对应的信道冲激响应矩阵H也就对应有N列相对应的数组,图2给出具体的方法流程图。 [0040] assuming that the received signal is r, corresponding transmission symbol containing N symbols, the corresponding channel impulse response matrix H will have a corresponding array of N columns corresponding to, Figure 2 shows a flowchart of a specific method.

[0041 ] 如图2所示,上述步骤100中对接收信号r进行信道匹配,获得待处理信号M,其采用如下公式(I)来实现, [0041] As shown, in step 100 the received signal r, 2 channel matched obtain M signal to be processed, which uses the following formula (I) is achieved,

[0042] M {I ) [0042] M {I)

[0043] 其中,ri表示第i次循环的输入信号讯表示包含H的第i到N列组成的矩阵。 [0043] where, ri represents the input information signal represents the i-th cycle comprising matrices H i-th to N-th column.

[0044] 上述步骤200对所述待处理信号进行最小均方误差-块判决反馈均衡算法,获得系列检测结果。 [0044] The step 200 the signal to be processed minimum mean square error - block decision feedback equalization algorithm, obtaining series of detection results. 此处的最小均方误差-块判决反馈均衡算法在具体实现时,是在常规Cholesky算法(包括前向替换和后向替换过程)中的后向替换过程内,添加块判决反馈处理过程,单独的Cholesky算法属于现有技术已经公开的算法,在此不作详细说明。 Here the minimum mean square error - block decision feedback equalization algorithm specific implementation, the decision feedback process is conventional in the Cholesky algorithm (alternatively forward and backward comprises replacing process) into the replacement process, adding block alone Cholesky algorithm is the algorithm has been disclosed prior art, not described in detail herein. 具体过程参见图2所示,如下: Referring to the specific process shown in Figure 2, as follows:

[0045] 首先,对步骤100处理获得的结果进行前向替换处理,即对M进行前向替换,得到z,如下式(2)所示: [0045] First, the result obtained in the processing of step 100 is processed prior to the replacement, i.e. M is to replace the former, to give Z, the following formula (2):

Figure CN101931593BD00071

[0047] 其中,L为对角元素为I的下三角矩阵;D为对角矩阵#是一个块中的符号数,i为已检测符号。 [0047] where, L is the diagonal elements of the triangular matrix I; D is a diagonal matrix # is the number of symbols in a block, i is detected symbols.

[0048] 然后,在z的后向替换处理中,按照以下公式(3)和⑷添加块判决反馈处理: [0048] Then, after the replacement process of z, the following equation (3) and the block decision feedback process ⑷ added:

Figure CN101931593BD00072

[0051] 其中,£„表示所述系列检测结果;zn表示进行所述前向替换后的处理结果;<是 [0051] wherein, £ "denotes a detection result of said series; Zn represents a result of the forward processing after the replacement; <is

第η个元素为I的单位向量;L为对角元素为I的下三角矩阵;1表示单位矩阵;□ {.}是判决算子#是一个块中的符号数。 The first element η is a unit vector I; L is the diagonal elements of the triangular matrix I; 1 denotes a unit matrix; {.} □ # operator is the decision is the number of symbols in a block.

[0052] 如图2所示,上述步骤200中将获得一系列的检测结果,所以在步骤300中,抽取系列检测结果4中的最后一个输出作为已检测符号鳥,即4 =4。 [0052] As shown in FIG. 2, 200 in the above step to obtain a series of detection results, so at step 300, extracting the last series of 4 outputs the detection result as detected symbol birds, i.e. 4 = 4. 通过此步骤可以逐次把 This step can be successively put

美提取出来,这样就将块判决反馈应用于干扰消除中。 US extracted, so that will be applied to block decision feedback interference cancellation.

[0053] 如图2所示,上述步骤300的干扰消除处理过程包括以下两步: [0053] As shown in FIG. 2, step 300 of the above-described interference cancellation process comprising the following two steps:

[0054] 首先,对已检测符号為进行判决,选择该已检测符号最可能对应的发射符号,SPf,'=0 |.v}; [0054] First, the detected symbols for the decision, selecting the detected symbol corresponding to the most likely transmitted symbols, SPf, '= 0 | .v};

[0055] 然后,把判决得到的符号作为干扰,将该干扰从输入的符号中去除,获得所述用于输出的符号,如下式: [0055] Then, the decision symbols obtained as the interference, the interference is removed from the input symbol, the symbol is obtained for output, the following formula:

[0056] [0056]

Figure CN101931593BD00073

[0057] 其中,表不输入的符号;y/ 1表不所述用于输出的符号;(H) i表不H的第i列,这里的H表示信道冲激响应矩阵;巧表示所述干扰。 [0057] wherein the symbol table is not inputted; y / 1 are not the symbol table for output; i-th column (H) i of the table is not H, where H represents a channel impulse response matrix; represents a clever interference.

[0058] 如图2所示,在具体实现时需要逐一对一定数量的符号进行信号均衡处理,而处理得结果是需要对所有的符号都进行检测,所以,在图1中,本发明还包括一判断步骤500:判断是否检测完所述接收信号中的所有符号,若没有,则返回步骤A重复执行均衡过程,直到检测出所有符号;否则,输出结果。 [0058] As shown, one by one requires a certain number of symbols is 2 when embodied equalization signal processing, and the processing result is the need to obtain all the symbols are detected, so that, in FIG. 1, the present invention further comprises a judgment step 500: determining whether all tests are completed, the symbols of the received signal, if not, returns to step a balancing process is repeatedly performed until all the symbols detected; otherwise, output.

[0059] 总之,如图2所示,本发明通过添加块判决处理将丽SE-1C的每一步中的MMSE-BLE过程都进行了优化,改变为MMSE-BDFE过程,得到的方法能获得额外的增益。 [0059] In summary, as shown, the present invention is the MMSE-BLE decision processing procedure for each step in Korea SE-1C are performed by the addition of 2 blocks optimized for the MMSE-BDFE changing process, additional methods can be obtained gain. 每次MMSE-BDFE算法得到的符号是MMSE-BDFE过程中解出的最后一个输出。 Per symbol MMSE-BDFE algorithm of MMSE-BDFE is solved during the last output. 改用MMSE-BDFE后,解出的最后一个元素有额外的判决增益,因此改进后的MMSE-1C有较好的性能。 After use MMSE-BDFE, solving the last element of the additional decision gain, improved MMSE-1C have better performance. 改进后的丽SE-1C称为最小均方误差-干扰消除-块判决反馈均衡,即丽SE-1C-BDFE方法,它在传统的MMSE-1C上进行优化获得了额外的增益,但又不同于传统的MMSE-BDFE方法,它结合了干扰消除方法,其抗噪声性能更强,具体可参见图3和图4。 Korea SE-1C improved called minimum mean square error - Interference Cancellation - block decision feedback equalization, i.e., Korea SE-1C-BDFE method, which is optimized in the conventional MMSE-1C to obtain additional gain but different MMSE-BDFE the traditional method, which combines the interference cancellation method, which is more anti-noise properties, see particularly Figures 3 and 4.

[0060] 上述图2所示的本发明的方法写为伪码,如下所示: [0060] The method of the present invention shown in FIG. 2 is written in pseudo-code as follows:

[0061] i=lto N [0061] i = lto N

[0062] % 滤波[0063] Ci为对ri进行MMSE-BDFE算法得到的向量 [0062] Filtering% [0063] Ci performs vector ri for the algorithm of MMSE-BDFE

[0064] i,为A的第一个兀素(最后一个输出) [0064] i, A is the first pixel Wu (last output)

[0065] %判决 [0065] verdict%

[0066] [0066]

Figure CN101931593BD00081

[0067] %干扰消除 [0067]% interference cancellation

[0068] [0068]

Figure CN101931593BD00082

[0069] i — i+1 [0069] i - i + 1

[0070] 将上述方法应用于TD-SCDMA系统(一种TDD-CDMA冲的联合检测算法可以获得较好的性能。使用在TD-SCDMA标准中推荐的多径衰落信道参数,如下表I (多径衰落信道参数)所示: [0070] The above-described method is applied to TD-SCDMA system (joint detection algorithm TDD-CDMA punch better performance can be obtained using the recommended standard, TD-SCDMA in multipath fading channel parameters, the following table I (multiple channel fading parameters) as shown:

[0071] [0071]

Figure CN101931593BD00083

[0073] 采用上述case3的参数,假设信道在一个突发周期内是接近非时变的,即在一个突发周期内信道抽头是固定的。 [0073] With the above parameters the case3, assume that the channel is in a burst period close time-invariant, i.e., within a burst cycle is fixed channel tap. 最大的时间延迟为16个码片周期。 The maximum time delay of 16 chip periods. 所有用户使用相同的扩频因子Q,并且Q=16,N=32 ;载波频率为2GHz。 All users use the same spreading factor Q, and Q = 16, N = 32; the carrier frequency of 2GHz. 对于接收机,信道冲激响应分别通过传统的MMSE-BDFE与本发明MMSE-1C-BDFE方法进行均衡处理,对应的抗噪声能力图分别如图3和图4所示。 For the receiver, the channel impulse response are processed for equalization by a conventional MMSE-BDFE and MMSE-1C-BDFE method of the present invention, the corresponding noise immunity are shown in FIG. 3 and FIG. 4.

[0074] 如图3所示,活动码道数K=8时的MMSE-BDFE与MMSE-1C-BDFE的性能比较;图3中横坐标Ed/NQ表示信噪比,纵坐标BER表示比特错误率,可见,MMSE-1C-BDFE比MMSE-BDFE具有更好的抗噪声能力。 [0074] As shown, the number of active code channels MMSE-BDFE Comparative 3 K = 8, and when the performance of MMSE-1C-BDFE; FIG. 3 abscissa Ed / NQ represents SNR, and the ordinate represents the bit error BER rate, visible, MMSE-1C-BDFE has better noise immunity than the MMSE-BDFE.

[0075] 如图4所示,活动码道数K=16时的MMSE-BDFE与MMSE-1C-BDFE的性能比较;图4中横坐标Ed/NQ表示信噪比,纵坐标BER表示比特错误率,可见,MMSE-1C-BDFE比MMSE-BDFE具有更好的抗噪声能力。 [0075] As shown in FIG. 4, the number of active code channels K = MMSE-BDFE compared with the performance of MMSE-1C-BDFE at 16; FIG. 4, the abscissa Ed / NQ represents SNR, and the ordinate represents the bit error BER rate, visible, MMSE-1C-BDFE has better noise immunity than the MMSE-BDFE.

[0076] 基于上述方法,本发明可以通过软件编程的形式对计算机内部性能进行改进,和/或通过添加硬件功能模块、或者修改硬件结果来在通用设备上实现上述方法所带来的功能。 [0076] Based on the above method, the present invention may be modified by the internal form of the performance of computer programming software, and / or by adding a hardware module, or to modify the hardware to implement the foregoing method results in a general equipment brought function. 比如,本发明还提供了一种信道均衡装置,所述装置包括:信道匹配模块,用于对接收信号进行信道匹配、并获得待处理信号;中间处理模块,用于对所述待处理信号进行最小均方误差-块判决反馈均衡,并获得系列检测结果;选择模块,用于对所述系列检测结果进行选择,并抽取最后一个输出作为已检测符号;干扰消除处理模块,用于将所述已检测符号作为干扰,从接收信号中消除,并生成新的接收信号;及判断模块,用于判断是否检测完所述接收信号中的所有符号,若没有则重新启动所述信道匹配模块,否则输出结果。 For example, the present invention also provides a channel equalization apparatus, said apparatus comprising: a channel matching module configured to match the received signal channel, and the signal to be processed is obtained; intermediate processing module, for the signal to be processed minimum mean square error - block decision feedback equalizer, and a detection result obtained series; selection module configured to select the series of the detection result, and extracts the last output as detected symbol; interference elimination processing module, for the as detected symbol interference, eliminating from the received signal, and generates a new received signal; and a determining module, for determining whether all tests are completed, the symbols of the received signal, if not restarting the channel matched module, or output.

[0077] 为了有效的添加块判决反馈处理,在中间处理模块中还设置有:前向处理单元,用于对待处理信号进行前向替换处理;后向处理单元,用于对前向替换处理结果进行后向替换处理,并添加块判决反馈处理。 [0077] In order to effectively add the block decision feedback process, the intermediate processing module is further provided with: a forward processing unit for processing signals for processing before treatment to replace; backward processing unit for processing results of the forward replacement after treatment to replace and add a block decision feedback process. 另外,上述干扰消除处理模块还可以设置:判决单元,用于对所述已检测符号进行判决,选择该已检测符号最可能对应的发射符号;及剔除单元,用于把判决得到的符号作为干扰,将该干扰从所述接收信号中去除,获得所述新的接收符号。 Further, the interference canceller processing module may be further provided: a decision unit for the decision detected symbols, selecting the detected symbol transmitted symbol corresponding to the most likely; and culling unit for decision symbols obtained as the interference , removing the interference from the received signal, the received symbols to obtain new. 上述各个功能模块或单元的实现方法具体参见上述关于方法的相关说明,在此不作重复说明。 Implementation method above respective functional modules or units refer to specific instructions regarding the above method, is not repeated here.

[0078] 综上所述,本发明的均衡方法获得额外增益,提高抗干扰性能,也可以用于联合检测。 [0078] In summary, the method of the present invention obtained by equalizing additional gain, noise immunity and the detection may also be used in combination. 联合检测一般用于TDD-CDMA系统。 Joint detection is generally used in TDD-CDMA system. 经历ISI信道的TDD-CDMA系统与单用户系统有类似的信道模型,只不过信道卷积矩阵为多个用户组合信道冲激响应(包含了用户扩频码和信道冲激响应)组成的卷积矩阵。 Subjected to ISI channel TDD-CDMA system and the single-user system with a similar channel model, the channel convolution matrix but is a combination of a plurality of user channel impulse response (including the spreading code and the user channel impulse response) composed of convolution matrix. 由于有类似的信道模型,检测时可以使用类似的算法,只不过检测出的符号包括各个符号周期内各个用户的符号。 Because similar channel model, using a similar algorithm can be detected, but the detected symbols comprises a symbol of each user in each symbol period.

[0079] 应当理解的是,上述针对本发明较佳实施例的描述较为详细,并不能因此而认为是对本发明专利保护范围的限制,本发明的专利保护范围应以所附权利要求为准。 [0079] It should be understood that the above description is directed to preferred embodiments of the present invention in some detail, and therefore can not be considered as limiting the scope of protection of the present invention, the scope of protection of the present invention shall be defined in the appended claims.

Claims (2)

  1. 1.一种信道均衡方法,其包括以下步骤: A、对接收信号进行信道匹配,获得待处理信号; B、对所述待处理信号进行最小均方误差-块判决反馈均衡,获得系列检测结果; C、抽取所述系列检测结果中的最后一个输出作为已检测符号; D、将所述已检测符号作为干扰,从接收信号中消除,生成新的接收信号; 其中,所述步骤D包括: 对所述已检测符号进行判决,选择该已检测符号最可能对应的发射符号; 把判决得到的符号作为干扰,将该干扰从所述接收信号中去除,获得所述新的接收信号; 还包括一判断步骤:判断是否检测完所述接收信号中的所有符号,若没有,则返回步骤A重复执行均衡过程,直到检测出所有符号;否则,输出结果; 从所述接收信号中除去所述干扰的过程按照下式进行: A channel equalization method, comprising the steps: A, a received signal channel matched to give signal to be processed; B, the signal to be processed to the minimum mean square error - block decision feedback equalizer, the detection result obtained Series ; C, and finally a series of decimating the output of the detection result as a detected symbol; D, as the detected symbol interference, eliminating from the received signal, generating a new received signal; wherein said step D comprises: the detected symbols for the decision, selecting the detected symbol corresponding to the most likely transmitted symbol; the symbol decision obtained as the interference, the interference is removed from the received signal, said received signal newly obtained; further comprises a determination step of: determining whether all tests are completed, the symbols of the received signal, if not, returns to step a balancing process is repeatedly performed until all the symbols detected; otherwise, output; removing the interference from the received signal process using the following formula:
    Figure CN101931593BC00021
    其中, 表示输入的符号;ri"表示所述新的接收信号;(H)i表示H的第i列,这里的H表示信道冲激响应矩阵;ί表示所述干扰; 所述最小均方误差-块判决反馈均衡处理方法中,对待处理信号依次进行前向替换、后向替换处理;并在所述后向替换处理中添加所述块判决反馈处理; 在所述后向替换处理中添加所述块判决反馈处理的过程按照以下公式进行: Wherein the symbols representing the input; ri "represents the new received signal; (H) i represents the i-th column of H, where H represents a channel impulse response matrix; ί represents the interference; the minimum mean square error - block decision feedback equalization processing method, the signal is sequentially treated prior to the replacement process, the replacement process to the post; and adding the block to a decision feedback process after the replacement process; adding to the replacement process in the rear process of the block decision feedback processing in accordance with the following formula:
    Figure CN101931593BC00022
    其中,匕表示所述系列检测结果;zn表示进行所述前向替换后的处理结果;<是第η个元素为I的单位向量;L为对角元素为I的下三角矩阵;1表示单位矩阵;0是判决算子;N是一个块中的符号数。 Wherein said series dagger represents a detection result; Zn represents a result of the forward processing after the replacement; <η is the first element I is a unit vector; L is the diagonal elements of a lower triangular matrix I; represents a unit 1 matrix; 0 is the decision operator; N is the number of symbols in a block.
  2. 2.一种信道均衡装置,其特征在于,所述装置包括: 信道匹配模块,用于对接收信号进行信道匹配、并获得待处理信号; 中间处理模块,用于对所述待处理信号进行最小均方误差-块判决反馈处理,并获得系列检测结果; 选择模块,用于对所述系列检测结果进行选择,并抽取最后一个输出作为已检测符号; 干扰消除处理模块,用于将所述已检测符号作为干扰,从接收信号中消除,并生成新的接收信号;及判断模块,用于判断是否检测完所述接收信号中的所有符号,若没有则重新启动所述信道匹配模块,否则输出结果; 其中,所述干扰消除处理模块包括:判决单元,用于对所述已检测符号进行判决,选择该已检测符号最可能对应的发射符号;及剔除单元,用于把判决得到的符号作为干扰,将该干扰从所述接收信号中去除,获得所述新的接收信号;其中 A channel equalization means, characterized in that said apparatus comprises: a channel matching module configured to match the received signal channel, and the signal to be processed is obtained; intermediate processing module, the signal to be processed to the minimum mean square error - block decision feedback process, and a detection result obtained series; selection module configured to select the series of the detection result, and extracts the last output as detected symbol; interference elimination processing module, configured to have the as detected symbol interference, eliminating from the received signal, and generates a new received signal; and a determining module, for determining whether all tests are completed, the symbols of the received signal, if not restarting the channel matching module, otherwise, outputs result; wherein the interference elimination processing module comprising: a decision unit for the decision detected symbols, selecting the detected symbol corresponding to the most likely transmitted symbol; and culling unit for decision symbols obtained as interference, the interference is removed from the received signal, said received signal newly obtained; wherein 所述干扰消除处理模块接收信号中除去所述干扰的过程按照下式进行: — (//)..XV,.:其中,ri表示输入的符号& "表示所述新的接收信号;⑶i表示H的第i列,这里的H表示信道冲激响应矩阵;5;表示所述干扰; 所述中间处理模块包括: 前向处理单元,用于对所述待处理信号进行前向替换处理; 后向处理单元,用于对所述前向替换处理结果进行后向替换处理,并添加块判决反馈处理; 在所述后向替换处理中添加所述块判决反馈处理的过程按照以下公式进行: The interference canceling signal processing module receives the interference removing process according to the following formula: - (//) .. XV,: where, ri represents the input symbols & "indicates that the new received signal; ⑶i FIG. i-th column of H, where H represents a channel impulse response matrix; 5; represents the interference; said intermediate processing module comprises: a forward processing unit for the signal to be processed prior to the replacement process performed; after the processing unit configured to perform the processing result of the replacement process to the replacement of the front, and add processing block decision feedback; during the addition of the block decision feedback process to the replacement process carried out after the following formula:
    Figure CN101931593BC00031
    其中,I表示所述系列检测结果;zn表示进行所述前向替换后的处理结果; 是第n个元素为I的单位向量;L为对角元素为I的下三角矩阵;1表示单位矩阵;:«是判决算子;N是一个块中的符号数。 Wherein, I represents the detection result of said series; Zn represents a result of the forward processing after the replacement; n-th element is a unit vector I; L is a lower triangular diagonal elements of the matrix I; 1 represents a unit matrix ;: «decision is operator; N is the number of symbols in a block.
CN 200910108237 2009-06-18 2009-06-18 Channel equalization method and device CN101931593B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 200910108237 CN101931593B (en) 2009-06-18 2009-06-18 Channel equalization method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 200910108237 CN101931593B (en) 2009-06-18 2009-06-18 Channel equalization method and device

Publications (2)

Publication Number Publication Date
CN101931593A true CN101931593A (en) 2010-12-29
CN101931593B true CN101931593B (en) 2013-12-11

Family

ID=43370524

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 200910108237 CN101931593B (en) 2009-06-18 2009-06-18 Channel equalization method and device

Country Status (1)

Country Link
CN (1) CN101931593B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2354676A (en) 1999-09-20 2001-03-28 Lucent Technologies Inc Iterative (turbo) equalisation
CN1706133A (en) 2002-10-16 2005-12-07 高通股份有限公司 Transmission scheme for multi-carrier MIMO systems
CN1809994A (en) 2003-06-25 2006-07-26 皇家飞利浦电子股份有限公司 Frame format decoder and training sequence generator for wireless lan networks
CN101123596A (en) 2006-08-12 2008-02-13 三星电子株式会社 Channel estimation method and system using linear correlation based interference cancellation (LCIC) combined with decision-feedback-equalization (DFE)

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2354676A (en) 1999-09-20 2001-03-28 Lucent Technologies Inc Iterative (turbo) equalisation
CN1706133A (en) 2002-10-16 2005-12-07 高通股份有限公司 Transmission scheme for multi-carrier MIMO systems
CN1809994A (en) 2003-06-25 2006-07-26 皇家飞利浦电子股份有限公司 Frame format decoder and training sequence generator for wireless lan networks
CN101123596A (en) 2006-08-12 2008-02-13 三星电子株式会社 Channel estimation method and system using linear correlation based interference cancellation (LCIC) combined with decision-feedback-equalization (DFE)

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
Angel Lozano等.《Layered Space-Time Receivers for Frequency-Selective Wireless Channels》.《IEEE TRANSACTIONS ON COMMUNICATIONS》.2002,第50卷(第1期),65-73.
Gong Yan-fei等.《A Method of Time-Varying Rayleigh Channel Tracking in MIMO Radio System》.《Journal of Electronic Science and Technology of China》.2005,第3卷(第4期),321-324.
张博.《MIMO系统信号检测算法的研究》.《科协论坛》.2008,92-93.

Also Published As

Publication number Publication date Type
CN101931593A (en) 2010-12-29 application

Similar Documents

Publication Publication Date Title
US6618433B1 (en) Family of linear multi-user detectors (MUDs)
Klein et al. Zero forcing and minimum mean-square-error equalization for multiuser detection in code-division multiple-access channels
US20060034352A1 (en) Methods and apparatus to facilitate improved code division multiple access receivers
US6956893B2 (en) Linear minimum mean square error equalization with interference cancellation for mobile communication forward links utilizing orthogonal codes covered by long pseudorandom spreading codes
US20050094713A1 (en) Unified MMSE equalization and multi-user detection approach for use in a CDMA system
Tidestav et al. Reuse within a cell-interference rejection or multiuser detection?
US20040228392A1 (en) Fourier-transform based linear equalization for MIMO CDMA downlink
US20090213907A1 (en) Method and Apparatus for Block-Based Signal Demodulation
EP1035681A2 (en) Method for interference cancellation
US7602838B2 (en) Linear turbo equalization using despread values
WO2003084097A1 (en) Transmit processing using receiver functions
CN1604512A (en) United detecting method and apparatus
Tugnait et al. Blind asynchronous multiuser CDMA receivers for ISI channels using code-aided CMA
CN1874189A (en) Method and device for concurrent eliminating same frequency interference in TDS-CDMA
Madhow et al. Minimum mean squared error interference suppression for direct-sequence spread-spectrum code-division multiple-access
US6873596B2 (en) Fourier-transform based linear equalization for CDMA downlink
US7974997B2 (en) Hardware matrix computation for wireless receivers
US20070171960A1 (en) Apparatus and methods for implementing a split equalizer filter for sparse channels
JPH06141021A (en) Transmitter/receiver for code division multiplex communication
US20090316759A1 (en) Multi-user detection using equalization and successive interference cancellation
US20070025299A1 (en) Interference cancellation within wireless transceivers
US20100020854A1 (en) Method and Apparatus for Communication Signal Processing Based on Mixed Parametric and Non-Parametric Estimation of Impairment Correlations
EP1093235A2 (en) Method and apparatus for detecting multiple signals in a CDMA network
CN102025392A (en) Interference elimination method and device
Chen et al. A robust adaptive DFE receiver for DS-CDMA systems under multipath fading channels

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
C14 Grant of patent or utility model