CN100544327C - Low complexity detector for serial interference deletion in minimum mean square error - Google Patents

Low complexity detector for serial interference deletion in minimum mean square error Download PDF

Info

Publication number
CN100544327C
CN100544327C CN 200510080117 CN200510080117A CN100544327C CN 100544327 C CN100544327 C CN 100544327C CN 200510080117 CN200510080117 CN 200510080117 CN 200510080117 A CN200510080117 A CN 200510080117A CN 100544327 C CN100544327 C CN 100544327C
Authority
CN
China
Prior art keywords
detector
matrix
lt
overscore
gt
Prior art date
Application number
CN 200510080117
Other languages
Chinese (zh)
Other versions
CN1697430A (en
Inventor
刘元安
刘思杨
罗振东
明 赵
Original Assignee
北京邮电大学
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
Application filed by 北京邮电大学 filed Critical 北京邮电大学
Priority to CN 200510080117 priority Critical patent/CN100544327C/en
Publication of CN1697430A publication Critical patent/CN1697430A/en
Application granted granted Critical
Publication of CN100544327C publication Critical patent/CN100544327C/en

Links

Abstract

本发明提供一种适用于多入多出(MIMO)系统的最小均方误差串行干扰删除(MMSE-SIC)检测器,其主要特点为:用于每一次检测的最小均方误差(MMSE)加权向量可以通过简单的递推方法得到,在递推计算的同时直接提供基于MMSE排序准则的排序方案。 The present invention provides a method suitable for multiple input multiple output (MIMO) system is minimum mean square error serial interference cancellation (MMSE-SIC) detector, which is mainly characterized by: means for detecting each of the minimum mean square error (MMSE) weight vector can be obtained by a simple recursive method, direct ordering scheme based on MMSE criteria while sorting recursive calculation. 与传统的MMSE-SIC检测器相比,本发明提供的检测器大幅度地降低了计算复杂度,而且没有带来任何性能损失。 Compared with the traditional MMSE-SIC detector, the detector of the present invention provides greatly reduces the computational complexity, and did not cause any performance loss. 另外,本发明也适用于其它可建模成MIMO系统的通信系统,例如码分多址(CDMA)系统。 Further, the present invention is also applicable to other systems may be modeled as a MIMO communication system, such as code division multiple access (CDMA) system.

Description

一种低复杂度的最小均方误差串行干扰删除检测器 A low-complexity MMSE detector serial interference canceller

技术领域 FIELD

本发明属于无线通信技术领域,涉及一种应用于多入多出(MIMO)系统的检测器,同时这种检测器也适用于其它可建模成MIMO系统的通信系统,例如码分多址(CDMA)系统。 The present invention belongs to the technical field of wireless communication, to a multiple input multiple output applied to the detector (MIMO) system, while such a detector is also applicable to other systems may be modeled as a MIMO communication system, such as code division multiple access ( CDMA) system.

背景技术 Background technique

随着蜂窝移动通信、多媒体业务的发展,世界范围内无线通信的容量需求在迅速增长,而可利用的无线频率资源却十分有限,如何在有限的频带内提供更大的信道容量成为未来高速无线通信系统发展的主要挑战。 With the development of cellular mobile communications, multimedia services, and capacity requirements for wireless communications is growing rapidly worldwide, and the available radio frequency resources are very limited, how to provide greater channel capacity to become the future high-speed wireless in a limited frequency band the main challenge to the development of communication systems. MIMO系统是一种利用多根发射天线和多根接收天线进行数据传输的无线通信系统,可以提供非常大的信道容量,在理想传播条件下其频谱利用率与天线数目呈线性关系。 A MIMO system using a plurality of transmit antennas and multiple receive antennas for data transmission in a wireless communication system that can provide a very large channel capacity, spectrum efficiency with which the number of antennas in a linear propagation conditions over. 由于MIMO系统具有极高的频镨利用率,因此被认为是未来高速无线通信系统的主要物理层技术之一。 Since the MIMO system has high frequency utilization praseodymium, it is considered one of the major physical layer technologies for future high-speed wireless communication system.

最小均方误差串行干扰删除(MMSE-SIC)检测器是适用于MIMO系统的一种有效的检测器,它才艮据最小均方误差(MMSE)准则从多个接收数据流中依次检测出发送的数据符号,当检测出某一发送符号后,将此符号所引起的干扰从接收信号中删除掉,然后再检测下一个发送符号。 Minimum mean square error serial interference cancellation (MMSE-SIC) detector is applied to the system an effective MIMO detector, according to which only Gen minimum mean square error (MMSE) criterion from the plurality of received data streams are sequentially detected transmitting data symbols, when a transmission symbol is detected, the interference caused by this symbol removed from the received signal, and then detects a next transmission symbol. 其检测顺序可以根据一定的排序准则(如:最小均方误差排序准则、最大信千噪比排序准则等)来确定。 Detecting order may be sorted according to certain criteria (such as: minimum mean square error ordering criterion, the maximum signal-to-noise ratio of one thousand sort criteria, etc.) is determined. 在传统的MMSE-SIC检测器中,由于计算MMSE矩阵时要多次进行复杂的矩阵求逆和排序运算,当收发天线数目较多时,其计算复杂度非常高。 In the conventional MMSE-SIC detector, due to complex matrix inversion calculation and sorting multiple computing MMSE matrix, when a large number of transceiver antennas, the computational complexity is very high.

如何在保证性能不下降的前提下,降低检测器的运算复杂度是此检测器进行实际应用的关键。 How to ensure that performance does not drop under the premise of reducing the computation complexity of the detector is the key to this practical application the detector.

发明内容 SUMMARY

本发明的目的在于提供一种MMSE-SIC检测器,它在保证检测性能不变的前提下,大幅度降低运算复杂度。 Object of the present invention is to provide an MMSE-SIC detector which ensure detection performance at the same premise, a significant reduction in computational complexity.

本发明提供的MMSE-SIC检测器的技术方案为:在同一发送时隙内该检测器包含连续iV次检测(7V表示发射天线数目),在第/次检测时,由扩展加权矩阵获得用于本次检测的加权向量co,.和其对应的发射天线序号& ,然后对发射符号A进行检测得到其判决值^ ,并从接收信号中删除^对其它未检测信号的干扰。 Technical Solution MMSE-SIC detector of the present invention provides: transmitting in the same slot of the detector comprises a continuous-time detection iV (7V denotes the number of transmit antennas) in the first / determinations, obtained by the weighting matrix for extended the weight vector co ,. detected and its corresponding transmitting antenna number &, and then detecting transmitted symbols a decision value which is obtained ^, and ^ remove interference from the received signal to other non-detection signal. 其关键点在于扩展加权矩阵,^采用递推计算的方法得到,加权向量co,. The key point is to extend the weighting matrix ^ recursion calculation method obtained weight vector co ,.

和其对应的发射天线序号A可从中直接获取。 And its corresponding transmission antenna number from which A direct access.

,^的递推算法如下: ^ Recursive algorithm is as follows:

1、计算扩展加权矩阵^1=1^, Rw可由R,,R2,…,R^逐步递推得到: 当^1时,计算R,(iihj2+cr2)—![hf7 o"]。这里,h,示信道矩阵H的 1, calculates the extended weight matrix ^ 1 = 1 ^, Rw by R ,, R2, ..., R ^ gradually recursive obtained: ^ 1 when calculating R, (iihj2 + cr2) - [hf7 o "] here!. , h, the channel matrix H shown

这里,〜=R;—^., R;—,表示由 Here, ~ = R; -. ^, R; -, represented by the

第/列;o"表示噪声的标准差;(*广表示共轭转置;H,ll表示Frobenius范数 The first / column; o "represents the noise standard differential; (* denotes a conjugate transpose wide; H, ll represents the Frobenius norm

当2Sy《iV时,R, When 2Sy "iV time, R,

的前m列构成的矩阵;m表示接收天线数目; 〃,=era,, A 二一2 + H^ ||2 +£72 IId; i卩)一1, f乂=h;. —Hyd, H"表示信道矩阵H的前乂—1 M column matrix composed of the front; m represents the number of receiving antennas; 〃, = era ,, A twenty-one 2 + H ^ || 2 + £ 72 IId; i Jie) a 1, f qe = h ;. -Hyd, H "represents a channel matrix H -1 front qe

列构成的矩阵;g7. =[a/;" —Ad)了, «表示转置;G;=d^,。 2、计算扩展加权矩阵^^ (/ = 2,3,...,AO。 Column matrix configuration;. G7 = [a /; "-Ad), and« denotes a transpose; G; = d ^ ,. 2, extended calculated weighting matrix ^^ (/ = 2,3, ..., AO .

^V,:Hlp,一J-2A,—!p二q,.—i 这里,A.,表示删除W.,的第l ,行后得到的矩阵;V.,表示删除A.,的第 !.. ^ V,: Hlp, a J-2A, - p two q, .- i where, A, represents deleted W., of l, matrix obtained after the row; V, represents deleted A., of

l一丄Z — l Z — ii—1 I —I a Shang l Z - l Z - ii-1 I -I

m + ,列后得到的矩阵;表示^一的第f M行;t表示删除p,_,的第M +《m个元素后得到的向量。 m +, obtained after column matrix; ^ represents a line of f M; T represents delete p, _, of M + "after the vector of m elements obtained.

获取o,.和A的方法如下: And A o ,. acquisition method is as follows:

1、抽取扩展加权矩阵W的前m列得到mmse加权矩阵W.,抽取扩展加权矩阵#;的后iV - 十1列得到排序矩阵Z,。 1, the weighting matrix W is extracted before expansion column to give mmse m weighting matrix W., extraction extended # weighting matrix; after iV - ten give a sort matrix Z ,.

2、 取出Z,.最小对角线元素对应的行序号^ ,从W,.抽取第A行得到lxM 维的加权向量0); , (0;所对应的发射天线序号即为& 。 2, Z ,. extracted row number corresponding to the smallest diagonal element ^, W ,. extracted from row A-dimensional weight vector obtained lxM 0);, (0; corresponding to transmit antenna number is the &.

实施本发明的有益效果在于:与传统的MMSE-SIC检测器相比,本发明提供的MMSE-SIC检测器采用了一种简单、高效的递推算法计算每一次检测所需的MMSE加权向量,在递推计算的同时直接提供基于MMSE排序准则的排序方案,这种处理方法极大地降低了检测器的'运算复杂度,同时保证了检测器的性能没有受到任何损失。 Embodiment of the present invention is that the beneficial effects: Compared with a conventional MMSE-SIC detector, the present invention provides MMSE-SIC detector employs a simple, computationally efficient recursive algorithm for each assay desired MMSE weighting vector, while providing the recursive calculation based on MMSE scheme directly ordering the ranking criteria, this approach greatly reduces the 'computational complexity of the detector, while ensuring the performance of the detector does not suffer any loss.

附图说明 BRIEF DESCRIPTION

图1是MIMO系统的基本原理框图。 FIG 1 is a basic block diagram of a MIMO system.

图2是MMSE-SIC检测器第z'次检测的流程图。 FIG 2 is a flowchart of the MMSE-SIC detector z 'detected time. 这里,/ = 1,2,…,〃 。 Here, / = 1,2, ..., 〃. 图3是计算加权向量o,.和其对应发射天线序号&,的流程图。 3 is a flowchart calculated weight vector and its corresponding transmit o ,. & antenna number, the. 这里, Here,

,'二1,2,…,7V。 , 'Two 1,2, ..., 7V.

图4是计算扩展加权矩阵^^的流程图。 FIG 4 is a flowchart of the calculation of extended ^^ weighting matrix.

图5是计算扩展加权矩阵,^.的流程图。 5 is a flowchart EXPANDED weighting matrix ^. A. 这里,z、2,3,…,7V。 Here, z, 2,3, ..., 7V.

图6是本发明提供的MMSE-SIC检测器与传统MMSE-SIC检测器的性能比 FIG 6 is the performance of MMSE-SIC detector and conventional MMSE-SIC detector than that provided by the present invention

较图(QPSK)。 FIG compared (QPSK).

图7是本发明提供的MMSE-SIC检测器与传统MMSE-SIC检测器的性能比较图(16QAM)。 FIG 7 is a performance comparison of FIG MMSE-SIC detector and conventional MMSE-SIC detector of the present invention provides (16QAM).

具体实施方式 Detailed ways

下面通过附图和实施例对本发明进行详细阐述。 The following elaboration of the present invention in detail by figures and examples.

本发明提供的检测器适用于MIMO系统,或是能够建4莫为MIMO系统的其它通信系统。 The present invention provides a detector suitable for the MIMO system, or can be a MIMO system built 4 mo other communication systems. 例如,本发明可直接用作CDMA系统的多用户检测器。 For example, the present invention can be used directly as a multi-user detector CDMA system. 下面以MIMO系统为例进行描述。 Below an example MIMO system will be described.

图1是MIMO系统的基本原理框图。 FIG 1 is a basic block diagram of a MIMO system. 在发射端,数据比特首先被映射成为信号星座中的信号,经过串并变换后形成多个并行的基带信号,然后经过调制 At the transmitting end, the data bits are mapped into a first signal constellation, after serial-parallel conversion of a plurality of parallel baseband signals are formed, and then the modulated

后分别从多个不同的天线同时发射出去;经过无线信道衰落后,来自不同发射天线的信号与噪声叠加后被多个天线同时接收,经过解调后生成多个并行基带信号,MIMCH企测器利用信道估计器产生的信道状态信息从基带信号中恢复出原始凄t据。 Respectively, from the plurality of different antennas transmitted simultaneously; after channel fading, the signal after the noise is superimposed from different transmit antennas of a plurality of receiving antennas at the same time, after the demodulated baseband signals to generate a plurality of parallel, MIMCH detector enterprises using channel estimator generates channel state information to restore the original data from the sad t baseband signal. 实际系统中,数据比特在映射之前可以先经过编码和交织,相应的在接收机输出数据之前要经过解交织和译码。 Practical systems, the data bits before mapping can go through coding and interleaving, the receiver corresponding to output the deinterleaved and decoded before the data. 该系统基带信号输入输出关系的数学表达式可以表示为: The mathematical expression system baseband signal input-output relationship can be expressed as:

y = Hx + s (1) y = Hx + s (1)

上式中,《=[^ & ... ^;f表示发射信号向量,W表示发射天线数目, In the above formula, "= [^ & ... ^; f represents a transmitted signal vector, W represents the number of transmission antennas,

—f表示转置,;c"表示从第"根发射天线发射的信号;£ = & s2…^f表示噪声向量,M表示接收天线数目,^表示第m根接收天线接收到的噪声; -f indicates transposition,; C "represents the first" signal transmitted by the transmitter antenna root; £ = & s2 ... ^ f denotes the noise vector, M is the number of receiving antennas, ^ m-th receive antenna denotes the received noise;

y=Ly, ^…h:r表示接收信号向量,^表示第m根接收天线接收到的信 y = Ly, ^ ... h: r represents the received signal vector, ^ represents the letter m th reception antenna to

号;H是MxiV维的矩阵,表示MIMO系统的等效基带信道矩阵;在进行MIMO 检测处理之前,首先要通过信道估计器获得信道矩阵的估计值,这里假设接收机可以无误差的估计步信道矩阵,为了方便描述,文中把信道矩阵的估计值仍记为H。 Number; H is MxiV dimensional matrix representing the equivalent baseband channel MIMO system matrix; prior MIMO detection process, first of all to obtain an estimate of the channel matrix through a channel estimator is assumed here that the receiver may be error-free estimate synchronization channel matrix, for convenience of description, herein the value of the channel estimate matrix are still referred to as H.

图2是MMSE-SIC检测器第Z次检测的流程图(/ = l,2,...,iV )。 FIG 2 is a flowchart of Z times detected MMSE-SIC detector (/ = l, 2, ..., iV). 该流程的步骤如下: The following process steps:

步骤l:计算加权向量to,.和其对应的发射天线序号^ (具体方法见图3)。 Step l: and calculating the weight vector corresponding to ,. transmitting antenna number ^ (specific methods see FIG. 3). 步骤2:计算、的判决值i「2(,y,.;)。这里,符号2W表示硬判决;当/ = 1 Step 2: calculate a decision value i "2 (., Y,;) where 2W represents the hard decision symbol; when / = 1

时,y:=y,当2^/《W时,yi由第/-l次检测的步骤3得到。 When, y: = y, when 2 ^ / "W, yi obtained from the first step / -l detected 3 times.

步骤3:如果/<^,将信号&的干扰从接收信号y,.中删除得到y,+。 Step 3: If / <^, the signal interference from the received signal & deleted obtained y ,. y, +. 即: which is:

y;+1=y;-、&;否则,输出判决值,结束算法。 y; + 1 = y; -, &; otherwise, outputs a decision value, the algorithm ends. 这里,、表示信道矩阵H的 Here ,, represents a channel matrix H

第&列。 The first & columns.

图3是计算加权向量cOi和其对应发射天线序号yt,的流程图(/ = 1,2,...,7V )。 3 is a flowchart cOi calculated weight vector and its corresponding transmitting antenna number yt, the (/ = 1,2, ..., 7V). 该流程的步骤如下: The following process steps:

步骤l:令z二l,递推计算扩展加权矩阵,^ (具体方法见图4)。 Step l: Order z = L, the recursive calculation of the weighting matrix expansion, ^ (specific methods see FIG. 4).

步骤2:抽取扩展加权矩阵W,.的前M列得到加权矩阵W,,抽取扩展加权矩 Step 2: extracting the weighting matrix W ,. extended front M columns obtained weighting matrix W ,, extracted extension weighted moments

阵^的后_ f +1列得到排序矩阵Z,。 Matrix ^ _ f +1 columns after sorting to obtain a matrix Z ,.

步骤3:取出Z,.的最小对角线元素对应的行序号《.,从W;中抽取第(.行得 Step 3: Remove the Z ,. smallest diagonal element corresponding to a row number ", from W is; extracted first (row too.

到第Z次检测的1 x M维的加权向量co,.,其对应的发射天线序号^为向量L,.中第 To 1 x M-dimensional weight vector of the co-time detection of the Z,., Its corresponding transmitting antenna number ^ first vector L ,.

A个元素的值。 A value elements. 这里,L,.表示删除向量[l 2…A/"]中值等于^^,…A—t的元 Here, L ,. means to delete vector [l 2 ... A / "] value equal ^^, ... A-t Element

素后得到的向量。 Vector obtained after prime.

步骤4:令/二z' + i,如果/^iV,由1^,—,递推计算扩展加4又矩阵W,(具体方 Step 4: Order / diethyl z '+ i, if / the iV ^, a ^ 1, -, 4 and recursive calculation of extended matrix W is applied, (particularly side

法见图5),转向步骤2;否则,结束算法。 Method see FIG. 5), to step 2; otherwise, the algorithm ends.

注:第7V次检测的加权向量a^的计算可由如下的简化算法得到: NOTE: 7V-time detection of the weight vector is calculated as a ^ a simplified algorithm can be obtained:

这里,cj表示噪声的标准差;—广表示共轭转置;IHI表示Frobemus范数。 Here, cj represents the noise standard differential; - represents a conjugate transpose wide; IHI represents Frobemus norm. 图4是计算扩展加权矩阵W,的流程图。 FIG 4 is a flowchart expansion calculated weighting matrix W, the. 该流程的步骤如下: 步骤l:令乂二l,计算R乂,hj2+a2)—'[h, o"-。 Step of the process is as follows: Step L: two orders qe l, calculate R qe, hj2 + a2) - '[h, o "-.

步骤2:令/=/ + 1,取出R.,的前M列构成矩阵灸.,,计算d,;6,.,h,, G =dg〃。 Step 2: Let / = / + 1, taken R., front M columns form a matrix calculation moxibustion ,, d,; 6,, h ,, G = dg〃... 这里,H.,表示信道矩阵H的前)-l列构成的矩阵。 Here, H., H denotes a channel matrix of pre) -l matrix columns.

,' / & / 7一1 J , '/ / 7 a 1 J

步骤3:计算R.二 Step 3: Calculation R. two

R , -GK R, -GK

j一i 7 ^ _/ j a i 7 ^ _ /

步骤4:当乂〈7V时,返回步骤2;当y二7V时,令W,二I^,结束算法。 Step 4: When qe <7V, the process returns to step 2; when y = 7V, so that W, two I ^, the algorithm ends. 图5是计算扩展力。 FIG 5 is a calculated extension force. 权矩阵^0、2,3,…,iV )的流程图。 Weight matrix ^ 0,2, 3, ..., a flow chart iV) of. 该流程的步骤如下步骤l:将,.—,按行拆分为向量p,.—i和矩阵A,一。 The process steps of Step l: a, .-, split into row vectors p, .- i and matrix A, a. 其中,p,.—,为W,—,的第〈,行,A.,为删除W.,的第《.,行后得到的矩阵。 Wherein, p, .-, as W, -, of <row, A, is deleted W., the first ", obtained after row matrix.

卜l Z—l /—1 Bu l Z-l / -1

步骤2:删除A,—,的第M +《,列得到矩阵V,.—,。 Step 2: Delete A, -, of M + ", the matrix column to give V, .- ,.

步骤3:删除p,—,的第M +《M个元素得到向量。 Step 3: Remove p, -, of M + "M vector elements obtained.

步骤4:计算,.=Vm — np,.—! ||-2 A;—lP「—,q,.—,,结束算法。 Step 4: calculating, = Vm - np, .- || -2 A; -lP "-, q, .- ,, algorithm ends.!.

图6和图7示出了本发明提供的检测器与传统MMSE-SIC检测器的两组性能比较结果。 Figures 6 and 7 illustrate two performance comparison result provided by the detector of the present invention and conventional MMSE-SIC detector. 图中横坐标表示的是发射数据的每比特能量与噪声功率谱密度的比值(EJN。),纵坐标表示的是误比特率(BER)。 FIG abscissa represents the ratio of the energy per bit to transmit data to the noise power spectral density (EJN.), The ordinate represents the bit error rate (BER). 系统的收发天线数目均为4,信道是独立同分布的MIMO平坦瑞利衰落信道,并作支设接收机可以无误差的估计出信道,检测器采用基于最小均方误差排序准则的排序方案。 Transceiver number of antenna systems are 4, the channel is a MIMO flat Rayleigh iid fading channel, and for estimating the channel support is provided receiver may be error-free, the detector based ordering scheme minimum mean square error sort criteria. 風6所仿真的系统采用QPSK调制,频谦效率为8bit/s/Hz;图7所仿真的系统采用16QAM 调制,频谱效率为16biVs/Hz。 Wind simulated system using QPSK modulation, frequency efficiency was modest 8bit / s / Hz; FIG. 7 of the simulated system uses 16QAM modulation, spectral efficiency 16biVs / Hz. 从蹈中可看出本发明提供的检测器性能与传统的MMSE-SIC检测器性能完全一致。 As can be seen from the tread of the present invention provides a conventional detector performance and the performance of MMSE-SIC detector identical.

下面分析本发明提供的MMSE-SIC检测器的运算复杂度。 The following calculation complexity of the present invention provides MMSE-SIC detector analysis. 这里以一次复数乘法的运算量为算法复杂度的单位,忽略加减法、比较、选择等相对简单的处理,只计算乘除法的复杂度。 Here a complex multiplication calculation amount in units of algorithmic complexity, ignoring subtraction, comparison, selection of a relatively simple process, only the calculation complexity of multiplication and division. 以发射天线数目AT和接收天线数目M相等时为例, 经过计算可知,本发明提供的检测器的复杂度大致为A^的水平,而传统的: MMSE-SIC检测器的复杂度则达到iV4的水平。 When the number of transmitting antennas to the receiving antenna AT and the number M is equal to, for example, through calculation shows that the complexity of the present invention provides a detector A ^ is substantially horizontal, whereas conventional: the complexity of the MMSE-SIC detector reached iV4 s level.

综上所述,本发明提供的检测器在不损失性能的前提下大幅度降低了运算复杂度。 In summary, the present invention provides a detector significantly reduces the computational complexity without loss of performance.

Claims (6)

1、一种适用于多入多出MIMO系统的低复杂度最小均方误差串行干扰删除MMSE-SIC检测器,在同一发送时隙内该检测器包含连续N次检测,其中N表示发射天线数目,其特征在于所述检测器的第i次检测包含如下步骤:a)根据以下递推算法计算第i次检测的扩展加权矩阵<overscore>W</overscore>i:当i=1时,<overscore>W</overscore>1=RN,RN由R1,R2,…,RN-1逐步递推得到;其中,当j=1时,当j=2,3,…,N时,这里,hj表示信道矩阵H的第j列,σ表示噪声的标准差,‖·‖表示Frobenius范数,(·)H表示共轭转置,表示由Rj-1的前M列构成的矩阵,M表示接收天线数目,βj=σαj,αj=(σ2+‖fj‖2+σ2‖dj‖2)-1,fj=hj-Hj-1dj,Hj-1表示信道矩阵H的前j-1列构成的矩阵,(·)T表示转置,当i=2,3,…,N时,<overscore>W</overscore>i的递推算法为:首先,将<overscore>W</overscore>i-1按行拆分为向量pi-1和矩 1, suitable for use in multiple input multiple output MIMO system low complexity MMSE MMSE-SIC serial interference cancellation detector, in the same transmission time slot comprises the detector detecting N consecutive times, where N denotes the transmit antenna number, characterized in that the i-th detector said detector comprising the steps of: a) calculating the i-th weighting matrix extended detected according to the following recursive algorithm <overscore> W </ overscore> i: when i = 1, <overscore> W </ overscore> 1 = RN, RN, ..., RN-1 obtained by the phase recursion R1, R2; wherein, when j = 1, when j = 2,3, ..., N, where hj denotes the j-th column of the channel matrix H, [sigma] represents the standard noise difference || || denotes the Frobenius norm, (·) H denotes conjugate transpose matrix representing consisting Rj-1 before and M columns, M represents the number of receiving antennas, βj = σαj, αj = (σ2 + ‖fj‖2 + σ2‖dj‖2) -1, fj = hj-Hj-1dj, j-1 before the column matrix configuration Hj-1 denotes a channel matrix H , (·) T denotes transposition, when i = 2,3, ..., N time, <overscore> W </ overscore> i recursive algorithm: first, the <overscore> W </ overscore> i-1 split into rows and moment vector pi-1 Λi-1,这里,pi-1为<overscore>W</overscore>i-1的第li-1行,Λi-1为删除<overscore>W</overscore>i-1的第li-1行后得到的矩阵;然后,删除Λi-1的第M+li-1列得到矩阵Vi-1;接着,删除pi-1的第M+li-1个元素得到向量qi-1;最后,计算b)根据<overscore>W</overscore>i得到第i次检测的最小均方误差MMSE加权矩阵Wi和排序矩阵Zi;c)取出Zi最小对角线元素对应的行序号li,取出Wi的第li行得到第i次检测的1×M维的加权向量ωi,这里,M表示接收天线数目;d)利用加权向量ωi恢复与其对应的第ki根发射天线发送的数据符号如果i≤N-1,则从接收数据中删除对其它未检测信号的干扰,这里,ki为ωi所对应的发射天线序号。 Λi-1, where, pi-1 is <overscore> W </ overscore> i-1 first li-1 line, Λi-1 is deleted <overscore> W </ overscore> i-1 first li-1 line after the obtained matrix; then, remove Λi-1 of the M + li-1 columns to give a matrix Vi-1; then remove pi-1 of the M + li-1 elements to obtain the vector qi-1; and finally, calculating b ) </ overscore> i obtain the minimum mean square error MMSE weighting matrix Wi and Zi i-th sort matrix detector according to <overscore> W; c) remove the Zi corresponding to the smallest diagonal element row number li, li of Wi removed the i-th row obtain detected 1 × M-dimensional weighting vector ωi, where, M is the number of receiving antennas; data symbols d) using the weighting vector corresponding to the recovery of [omega] i ki transmit antenna transmission if i≤N-1, remove interference from the received data on other non-detection signal, here, for the Ki ωi corresponding transmit antenna number.
2、 根据权利要求1所述的检测器,其特征在于,扩展加权矩阵的前M列即为第/次检测的MMSE加权矩阵W,.。 2. The detector according to claim 1, characterized in that, before the extended weight matrix of M columns that is / times MMSE weighting matrix W is detected, ..
3、 根据权利要求l所述的检测器,其特征在于,扩展加权矩阵W,的后_ / +!列即为第/次检测的排序矩阵Z;。 3, the detector according to claim l, characterized in that the expansion weighting matrix W, the _ / +! Column is the sort of the matrix Z / time detection ;.
4、 根据权利要求l所述的检测器,其特征在于,加权向量o^的计算公式为:〜=(11、112+"2)_1( 4. The detector as claimed in claim l, wherein the weight vector ^ o is calculated: = ~ (11,112+ "2) _1 (
5、 根据权利要求l所述的检测器,其特征在于,该检测器能够根据其它排序准则进行排序。 5. The detector as claimed in claim l, characterized in that the detector is capable of sorting to sort according to other criteria.
6、 根据权利要求l所述的检测器,其特征在于,该检测器适用于能够建模成MIMO系统的通信系统。 6. The detector as claimed in claim l, characterized in that the detector is applied to a communication system can be modeled as a MIMO system.
CN 200510080117 2005-06-30 2005-06-30 Low complexity detector for serial interference deletion in minimum mean square error CN100544327C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 200510080117 CN100544327C (en) 2005-06-30 2005-06-30 Low complexity detector for serial interference deletion in minimum mean square error

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 200510080117 CN100544327C (en) 2005-06-30 2005-06-30 Low complexity detector for serial interference deletion in minimum mean square error

Publications (2)

Publication Number Publication Date
CN1697430A CN1697430A (en) 2005-11-16
CN100544327C true CN100544327C (en) 2009-09-23

Family

ID=35349958

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 200510080117 CN100544327C (en) 2005-06-30 2005-06-30 Low complexity detector for serial interference deletion in minimum mean square error

Country Status (1)

Country Link
CN (1) CN100544327C (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100581091C (en) 2007-04-19 2010-01-13 北京邮电大学 Automatic retransfer request scheme based on the detection and merging joint optimization of the MIMO system
CN101409604B (en) 2007-10-09 2011-04-20 鼎桥通信技术有限公司 Method for eliminating serial interference in multi-input multi-output system
CN101247159B (en) 2008-03-26 2011-06-15 北京天碁科技有限公司 Activation code path detecting method for multi-antenna association detecting system

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1373581A (en) 1994-07-26 2002-10-09 交互数字技术公司 Spread spectrum interference eliminator system
CN1615597A (en) 2002-01-17 2005-05-11 皇家飞利浦电子股份有限公司 Multi-user detection in mobile terminal for reducing interference

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1373581A (en) 1994-07-26 2002-10-09 交互数字技术公司 Spread spectrum interference eliminator system
CN1615597A (en) 2002-01-17 2005-05-11 皇家飞利浦电子股份有限公司 Multi-user detection in mobile terminal for reducing interference

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
分组空时块编码结构及其信号检测算法. 周杰,管云峰,徐友云,罗汉文,葛建华.电讯技术,第2期. 2005

Also Published As

Publication number Publication date
CN1697430A (en) 2005-11-16

Similar Documents

Publication Publication Date Title
Jeganathan et al. Spatial modulation: Optimal detection and performance analysis.
Mesleh et al. Quadrature spatial modulation
Reynolds et al. Blind adaptive space-time multiuser detection with multiple transmitter and receiver antennas
CN100420178C (en) Method and apparatus for channel quality measurements
CN1211945C (en) System and method for processing radio signals received by multiple antenna vibrators
US7386076B2 (en) Space time encoded wireless communication system with multipath resolution receivers
JP4939888B2 (en) Wireless communication device
ES2425014T3 (en) Generating soft values ​​for low-complexity MIMO receiver JD-GRAKE
CN100414861C (en) Space-time coded transmissions within a wireless communication network
Molisch et al. Wideband wireless digital communication
CN1172472C (en) Receiving processing method of mobile communication system and receiving appts. thereof
CN100370718C (en) Joint detection receiving equipment and method without considering orthogonal code length
CN1284308C (en) Method and apparatus for processing a modulated signal using an equalizer and a rake receiver
CN1274093C (en) Method and apparatus for providing blind self-adaptive estimation and reception
JP5030279B2 (en) The wireless communication device and wireless communication method
CN101032108B (en) Method and detector for a novel channel quality indicator for space-time encoded mimo spread spectrum systems in frequency selective channels
Bai et al. Low complexity MIMO detection
US20050009476A1 (en) Virtual MIMO transmitters, receivers, systems and methods
CN101888259B (en) Method, apparatus and system used for CDMA communication
US20060268963A1 (en) Radio communication system that uses a MIMO receiver
KR20040085680A (en) Method for adaptive transmission and receiving in a wireless communication system with multiple antennas
CN1111864A (en) A system and method for joint demodulation of CDMA signals
CN1759617A (en) Weight generation method for multi-antenna communication systems utilizing RF-based and baseband signal weighting and combining
US20060135081A1 (en) System and method for data communication over multi-input, multi-output channels
Guo et al. A VLSI architecture of the Schnorr-Euchner decoder for MIMO systems

Legal Events

Date Code Title Description
C06 Publication
C10 Request of examination as to substance
C14 Granted
ASS Succession or assignment of patent right

Owner name: SHENZHEN TINNO WIRELESS TECHNOLOGY CO., LTD.

Free format text: FORMER OWNER: BEIJING POSTAL AND TELECOMMUNICATIONS UNIV.

Effective date: 20150702

C41 Transfer of the right of patent application or the patent right