WO2012069007A1 - 快速时变mimo系统中基于非码本预编码的传输方法 - Google Patents
快速时变mimo系统中基于非码本预编码的传输方法 Download PDFInfo
- Publication number
- WO2012069007A1 WO2012069007A1 PCT/CN2011/082936 CN2011082936W WO2012069007A1 WO 2012069007 A1 WO2012069007 A1 WO 2012069007A1 CN 2011082936 W CN2011082936 W CN 2011082936W WO 2012069007 A1 WO2012069007 A1 WO 2012069007A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- matrix
- channel
- signal
- permutation
- precoding
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
- H04L25/03006—Arrangements for removing intersymbol interference
- H04L25/03343—Arrangements at the transmitter end
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/0413—MIMO systems
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
- H04L25/03006—Arrangements for removing intersymbol interference
- H04L2025/0335—Arrangements for removing intersymbol interference characterised by the type of transmission
- H04L2025/03426—Arrangements for removing intersymbol interference characterised by the type of transmission transmission using multiple-input and multiple-output channels
Definitions
- the present invention relates to the field of communications technologies, and in particular, to a non-codebook precoding based transmission method in a fast time varying MIMO system. Background technique
- MEMO Multi-Input Multiple-Output
- FDD Frequency Division Duplexing
- TDD Time Division Duplexing
- the base station can obtain partial or complete channel information without feedback, and apply precoding technology.
- the research on precoding technology is mainly divided into codebook based precoding and non-codebook based precoding.
- the non-codebook based precoding technique can make the transmitted data better match the channel, resulting in higher beamforming gain.
- the Singukr Value Decomposition (SVD) algorithm is widely used in non-codebook-based precoding.
- the theoretical transmission rate of the SVD-based transmission and reception algorithm can reach the channel capacity.
- the MIMO channel is converted into independent single-input Single-Output (SISO) subchannels with different gains, and there is no dry-pull between the sub-channels.
- SISO independent single-input Single-Output
- the SIC Serial Interference Cancellation
- SNR signal-to-noise ratio
- the technical problem to be solved by the present invention is how to reduce interference between sub-channels in a MIMO system during SVD-based transmission in the case of rapid channel change (high speed mobility).
- the technical solution of the present invention provides a fast time-varying MIMO system based on non- a codebook precoding transmission method, at the receiving end, the method includes:
- the method further includes:
- S11 Obtain channel state information at time n, -f and perform singular-value decomposition on it, and obtain a precoding matrix u S12: according to the precoding matrix ⁇ , - the channel to be transmitted and the symbol modulated signal to be transmitted ⁇ And the inserted pilot is simultaneously precoded, and the precoded signal is sent to the t time transmission channel.
- step S2 further includes:
- the sorting-based QL decomposition process includes: reordering columns in the channel matrix based on determinant criteria to obtain a desired serial interference detection order; and, based on the modified Gram-Schmidt Gram Schmidt method Performing QL decomposition on the reordered channel matrix; and the reordering process includes:
- the method based on the - solution includes: reordering the columns of the old Gramm:-Schmidt 'schmidt orthogonal track matrix; and the sorting and the process include: , performing ⁇ ⁇ ,... Following steps;
- the technical solution of the present invention combines a transmitter precoding and a receiving end MIMO sub-signal detection algorithm.
- Channel estimation is performed on the pre-coded pilot signal at the receiving end, and the QL decomposition based on the ranking or the QL decomposition based on the minimum mean square error expansion is obtained, and a set of permutation matrix P, ⁇ matrix Q and pair are obtained.
- the lower triangular matrix L is used to perform serial interference elimination, and the detection of the MO subchannel is completed layer by layer.
- the permutation operation is performed by using the permutation matrix. Since the QL algorithm based on the sorting obtains the angular matrix L at the high: ⁇ good descending characteristics, guarantee
- Figure 1 is a fast time-varying ⁇ of the present invention
- Figure 2 is a fast time change of the present invention
- Transmission diagram where: send: the end uses feedback to obtain channel information
- a second embodiment of a non-codebook precoding based transmission method in a time-varying MIMO system The transmission diagram, wherein the sender obtains the letter by using the TDD channel reciprocity: ⁇ 3 ⁇ 4 ⁇ ;
- Figure 4 shows a comparison of the probability of the diagonal elements of the lower triangular matrix L obtained by the sorting-based decomposition algorithm and the unordered decomposition algorithm in descending order in the case of different channel time correlations;
- FIG. 5 is a diagram showing a non-codebook precoding based transmission method and a non-coded QL decomposition algorithm based non-codebook precoding in a fast time varying chirp system according to different channel time correlations and different signal to noise ratio conditions. The performance of the transmission method is compared with the bit error rate.
- FIG. 6 shows a non-codebook precoding based transmission method in the fast time varying chirp system of the present invention under the condition that the Doppler frequency shift is 300 Hz under different SNR conditions.
- SVD+QL-SORT, SVD+MMSE-SORT, SVD+MMSE-PSA the bit error rate of the transmission method
- SVD+QL bit error rate of the transmission method
- the core idea of the technical solution of the present invention is that the transmitting end performs singular value decomposition (SVD) using the obtained channel information to obtain a precoding matrix, and simultaneously performs precoding on the transmission signal and the inserted pilot according to the precoding matrix;
- the channel performs channel estimation according to the pilot that completes the precoding, and performs QL decomposition based on the ranking of the channel estimation result or QL decomposition based on the minimum mean square error spread to obtain a set of permutation matrix, ⁇ matrix Q and diagonal elements.
- the lower triangular matrix L of the sequence then, the completion of Q "multiplication with the received signal, and the use of the diagonal elements of the lower triangular matrix L can also be arranged in an orderly manner, the received signal Serial interference cancellation processing is performed, and the final MMO decoded signal is obtained.
- the operation of precoding at the transmitting end is equivalent to completing the process of sorting the lower triangular matrix L.
- the diagonal elements of the lower triangular matrix L have a high probability of being pressed. Sorted in descending order. However, under high-speed movement, the Doppler shift of the channel is large, and the channel changes rapidly. At this time, the descending triangle descending characteristics cannot be guaranteed.
- the sorting-based QL decomposition is adopted at the receiving end, which can still be guaranteed under the high-speed moving environment.
- the diagonal elements of the triangular matrix L are as shown in FIG. 1.
- the transmission method based on the non-codebook precoding in the time varying MIMO system according to the present invention includes the following steps at the transmitting end:
- S11 Obtain channel state information u at time T.
- the transmitting end may obtain channel state information I T through feedback.
- the transmitting end may obtain channel state information H by using channel reciprocity;
- S12 Perform singular value decomposition (SVD) on the acquired channel state information, and obtain a left ⁇ matrix right ⁇ matrix, ⁇ , and a diagonal matrix in which the elements on the diagonal are real numbers and are arranged in descending order;
- SVD singular value decomposition
- step 12 further includes storing the precoding matrix v; -f and its conjugate transpose
- S13 Unitary matrix guide in the transmission signal after channel coding and modulation symbols and pilot signal insertion while completing transmitting the precoding; precoded signal and to complete the current transmission into: H f channel in time!.
- a non-codebook precoding based transmission method in a time varying MIMD system includes the following steps at the receiving end:
- the step 1 further includes: ???resolving the channel estimation result, which is an equivalent channel information including a precoding matrix and a channel matrix of the time transmission transmission;
- a sorting-based QL decomposition is performed to generate a set of permutation matrices, and at the same time, the unitary matrix Q and the ordered lower-angle matrix L of the elements on the diagonal are obtained.
- the sorting based can be implemented in the following two ways.
- QL decomposition is performed after reordering the columns in the channel matrix; or, the channel matrix is reordered in the QL decomposition process.
- the process of reordering the columns of 0;1 3 ⁇ 4 ] includes:
- the first step initialization
- the detection order obtained is ⁇ 3 ⁇ 4, . ⁇ ,,3 ⁇ 4]; Determine the permutation matrix according to the order of the measurements.
- the modified Gram-Schmidt method can be used to perform QL decomposition on the rearranged channel matrix to obtain the unitary matrix Q and Lower triangular matrix L.
- the second implementation method further includes two sorting methods: diagonal of the lower triangular matrix L matrix in the process of QL decomposition
- the line elements are sub-optimized (QL SORT sorting scheme) and PSA (Post Sort Aigontlim).
- the sub-optimal ordering uses a modified gnschmidt orthogonalization method to change the order of the columns of the channel matrix in the process of decomposition to obtain a better detection order, which actually belongs to the QL SORT.
- the algorithm process is:
- a unitary matrix Q, a diagonally ordered lower triangular matrix L and a set of permutation matrices ⁇ ( ⁇ ⁇ ) ⁇ are obtained.
- the optimal post-order QL decomposition on the equivalent channel matrix 11 or the extended channel matrix, and analyzes the covariance matrix of the QL decomposition and the row of the exchange matrix L to implement the covariance matrix.
- the matrix obtained at this time is no longer a triangular matrix.
- the first row of the matrix should contain only one non-zero element.
- the triangle characteristics of the matrix can be regained by the Househoider transform:
- a row vector im can be mapped to a plane by the Householder transform method or a new vector bS> of equal length on the line, and the map is obtained by multiplying aa with the unitary matrix ⁇ . Multiplying the matrix j ⁇ L- 'right by the matrix ⁇ can force all other elements except the first element in the last row to be zero, without changing the norm of the row.
- the resulting optimal matrix has the following form:
- X is the transmitted signal, is the precoding matrix
- 11 ( , " 2 , ..., : 1 power is Gaussian white noise, ⁇ 3 ⁇ 4 ⁇ ⁇ , ⁇ ,..., ⁇ is Gaussian with power ⁇
- White noise R , H, is; channel state information at the moment; here, indicating transpose, (.f table, ⁇ .) indicates expectation:
- ⁇ denotes the elements of the first row and the second column of the lower triangular matrix L, which are the processed received signals of the first subchannel, and cfc denotes the decision process.
- FIG. 2 is a diagram showing the first type of the present invention.
- the sending end uses the feedback information, and the specific steps include:
- the transmitting end performs channel coding on the initial bit suffix and the transmitting signal b.
- channel coding modes such as ⁇ ⁇ code, LDPC code, etc. may be adopted;
- the transmitting end performs symbol modulation on the channel-coded super-level signal to obtain 4*- Symbol modulation such as QPSK, i6QAM, 64QAM, etc. can be used:
- the transmitting end inserts a pilot; preferably, the frequency division, the time division, the code division, etc. may be implemented;
- S204 Send The feedback from the receiver is the equivalent channel information of the packet-matrix and the channel matrix, where ⁇ '> ⁇
- the transmitting end uses the precoding matrix v; -7 to complete the precoding process for the inserted pilot and symbol modulated signals; that is, the precoding matrix and the symbol modulated modulated signal X and the pilot signal Multiply by; and send a signal;
- the receiving end A receives the antenna receiving signal ⁇ (.H- 1 ⁇ 4 f, and performs channel estimation on the pilot that completes the precoding, and obtains the signal estimation result, which is not the channel state information H but the transmitting precoding.
- the receiving end processes the Q and the received signal, and has:
- n ci - Q - ( , 2 ,.'.,3 ⁇ 4 "j is the Gaussian white noise of power (R s .
- transpose (.f means conjugate transpose, ⁇ £(.) means expectation;
- the receiving end completes the serial interference cancellation processing (SIC) on the processed received signal y, where the serial interference erasing process may be completed by layer-by-layer detection of the MIMO subchannel; the detecting sequence may be a sub-channel 1 to the subchannel ⁇ and the detected number of the first subchannel is:
- SIC serial interference cancellation processing
- the element indicating the row and the column of the lower triangular matrix L is the processed received signal of the ⁇ subchannel, indicating a decision process
- Step 216 The receiving end completes symbol demodulation of the signal after the MMO subchannel detection, where the demodulation mode corresponds to the symbol modulation mode of the transmitting end;
- the demodulated bit-level signal is subjected to channel decoding to obtain a decoded signal ⁇ ; wherein, the channel.
- Decoding side corresponds.
- the transmission model surface is the second embodiment of the present invention.
- the transmitting end obtains channel information by using the TDD system channel reciprocity, and the specific steps include:
- the transmitting end performs channel coding on the initial bit level and the transmitting signal b.
- a channel coding manner such as Turbo. LDPC code may be adopted;
- S302 The transmitting end performs symbol modulation on the channel-coded bit-level signal, and preferably obtains a modulation method such as QPSK or 16QAM 64QAM.
- S303 The transmitting end inserts a pilot; preferably, the frequency division, the time division, the code division, and the like may be implemented;
- S304 The transmitting end directly utilizes the channel reciprocity of TDD to obtain channel disturbance information at time r! ⁇ ;
- S305 The transmitting end performs singular value decomposition (SVD) on the channel state information, and obtains a left ⁇ matrix right ⁇ matrix and a diagonal matrix ⁇ f — f in the descending order.
- SVD singular value decomposition
- the transmitting end uses the right chirp matrix (the precoding matrix performs the precoding process on the pilot and symbol modulated transmission signals simultaneously; that is, the precoding matrix V_ and the symbol modulated transmission signal X and the pilot signal are simultaneously Multiply; and send a signal;
- the receiving end receives the antenna receiving signal ( ⁇ , .,., ), and performs channel estimation on the pilot that completes the precoding to obtain a channel estimation result: the channel state information 11 is obtained at this time ; Transmitting the equivalent channel information of the precoding matrix,
- ⁇ ( ⁇ 2 ,...,3 ⁇ 4) is the processed received signal, Is the receiving signal on the receiving antenna, ⁇ : : ⁇ , 2 ,...3 ⁇ 4; ⁇ is the transmitting signal, ⁇ is the precoding matrix, ⁇ ( , " 2 ,..., 3 ⁇ 4 ) is the Gaussian white of power Noise, fi ? ::: Q : ..., Aii f is Gaussian white noise with power (R ), where (.f denotes transpose, (.f denotes conjugate transpose, and £(.) denotes expectation;
- SIC serial interference cancellation processing
- Z represents an element of the row and column of the lower triangular matrix L, which is a processed received signal of the first subchannel, and ofec(.) represents a decision process;
- step S311 performing the anti-replacement operation on the signal after the interference cancellation operation, and obtaining the finally detected signal i l, .. ⁇ ) is a set of permutation matrices obtained in the sorted QL decomposition in step S212;
- S312 The receiving end completes symbol demodulation of the signal after detecting the subchannel, where the demodulation mode corresponds to the symbol modulation mode of the transmitting end;
- the receiving end performs channel decoding on the demodulated bit-level signal to obtain a decoded signal ⁇ . wherein the channel decoding mode corresponds to the channel coding mode of the transmitting end.
- the present invention combines a pre-coding and a receiving sub-channel detection algorithm at the transmitting end, wherein the transmitting end simultaneously pre-encodes the transmitting signal and the pilot, which is not estimated at the receiving end for the pilot channel.
- the channel state information ⁇ 3 ⁇ 4 but contains the equivalent channel information of the precoding matrix ⁇ ⁇ ⁇ ; the receiving end further 11 ⁇ or based on the minimum mean square error extended QL vine, get a set of permutation matrix ⁇ endeavor, ⁇ matrix Q and the diagonal element are the lower triangular matrix L of the real number.
- the received signal is processed by the ⁇ matrix Q, and on the basis of not amplifying the noise, the number of the first layer subchannel signals received from the front layer is eliminated.
- the lower triangular matrix L after reordering is in the high-speed moving environment.
- the medium pseudo-predicate can maintain better descending characteristics, guarantee the performance of the serial interference elimination technology, and reduce the interference between the sub-channels in the fast time-varying MIMO system;
- receiving algorithm aspect of the present invention can further suppress the influence of error propagation caused by the channel has been detected.
- FIG. 5 shows a transmission method based on an unsorted QL decomposition receiving algorithm (SVD ⁇ HQL) and a transmission method based on the sequential QL decomposition receiving algorithm of the present invention under different channel time correlation and different signal to noise ratio conditions.
- Implementation side using the aforementioned QL-SORT sequencing scheme).
- FIG. 6 shows an embodiment of a transmission method based on an unsorted QL decomposition receiving algorithm (SVD+QL) and a transmission method based on the sorting sub-receiving algorithm of the present invention when using ⁇ 1/3 encoding, QPSK modulation, SVIHQL SORT (For the channel matrix.
- SVD+QL- PSA unsorted QL decomposition receiving algorithm
- the SVD+QL-SORT scheme in the present invention can obtain greater performance gain than the SVD+QL scheme with little increase in complexity: Turbo 1/3 encoding, QPSK modulation, respectively, gain performance gain of L8dB and 3de .
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Power Engineering (AREA)
- Radio Transmission System (AREA)
Description
快速时变 l MO系统中基于非码本预编码的传输方法 技术领域
本发明涉及通信技术领域, 尤其涉及一种快速时变 MIMO系统中基于非码本预编 码的传输方法。 背景技术
随着多天线技术研究的深入, MEMO ( Multiple-Input Multiple-Output, 多输入多输出)系 统已经得到广泛的应用, 它可以对信道容量、 链路可靠性和覆盖范围等带来诸多好处, 尤其 是当发送端已知信道状态信息时, 可以通过发送端的预编码^入更大的性能增益。 在 FDD (Frequency Division Duplexing, 频分双工) MIMO系统中, 基站端可以通过反馈获得部分或完 整的信道信息,实现预编码技末。在 TDD (Time Division Duplexing,时分双工) MIMO系续,中, 基于信道互惠性的存在, 基站端可以在不需要反馈的情况下获得部分或完整的信道信息, 应 用预编码技术。
目前对预编码技术的研究主要分为基于码本的预编码和基于非码本的预编码。 当发送端 已知信遒状态信息时, 基于非码本的预编码技术可以使发送数据更好地匹配于信道, 带来更 高的波束成形增益。 在基于非码本的预编码中广泛应用奇异值分解 (Singukr Value Decomposition, SVD )算法,基于 SVD的发送接收算法在理论上的传输速率可以达到信道容 量。 在基于 SVD的传输中, MIMO信道被转换为独立的带有不同增益的并行单输入单输出 ( Single-Input Single-Output, SISO )子信道, 且子信道之间没有干挽。 然而实际中, 传统的 基于 SVD的传输会 入子信道之间的干扰, 这种干扰会大大降低基于 SVD传输的性能。
SIC ( successive interference cancellation, 行干扰删除 )技术是一种广泛应用的干扰删除 技术, 而其中的检测顺序对 SIC的性能至关重要。 因此, 若想更为有效地抑制 M1MO子信道 干扰, 需要尽可能地按照信噪比(SNR )从大到小的顺序完成接收端的 MIMO检瀏。 发明内容
(一) 要解决的技术问题
本发明要解决的技术问题是如何在信道变化迅速 (高速移动 )的情况下减小基于 SVD 的传输时 MIMO系统中各子信道之间的干扰。
(二) 技术方案
为解决上述技术问题,本发明的技术方案提供了一种快速时变 MIMO系统中基于非
码本预编码的传输方法, 在接收端, 所述方法包括:
S1: 接收由发送端发送的、 经同时预编码处理的信号^和导频, 根据所述导频完成信道 估计, 获得等效信道 H^ ;
S2: 对所述等效信道《^„进行基于棑序的 QL分解, 生成一组置换矩阵 I 并同时获得 酉矩阵 Q和对角线上元素有序的下三角矩阵 L;
S3: 根据所述酉矩阵 Q对信号 f进行处理, 获得处理后的接收信号 y H ;
S4: 根据所述下三角矩阵 L对 -所述处理后的接收信号 =(3 进行串行干扰消除, 获.得干 扰消除后的信号 ;
S5: 根据所述置换矩阵^对所述干扰消除后信号 进行反置换操作, 获得信号 i ;
S6: 对述信号 i完成解调和译码。
进一步地, 所述方法在发送端还包括:
S11 : 获取 n时刻的信道状态信息 ,—f并对其进行奇异-值分解., 获得预编码矩阵 u S12: 根据所述预编码矩阵 ν,— 对经信道编码和符号调制的待发送信号 χ以及插入的导 频同时 于预编码, 并将预编码后的信号送入 t时刻传输信道中。
进一步地, 所述步骤 S2还包括:
S21 : 根据最小均方误差准则对所述等效信道 进行信道扩展, 获得扩展后的信道 H, - [H, : σ ί ,, ] :
S22: 对所述扩展后的信道 '„ 进行基于排序的 QL分解, 生成一组置换矩阵 P„, 并同 时获得酉矩阵 Q和对角线上元素有序的下三角矩阵 L ; 其中, 为加性高斯白噪声功率的 均方根。
其中, 所述基于排序的 QL分解过程包括: 基于行列式准则对信道矩阵中的列进行重 新排序, 以获得期望的串行干扰检测次序; 以及, 基于修正的 Gram- Schmidt格拉姆施密 特方法对重新棑序后的信道矩阵进行 QL分解; 且所述重新排序的过程包括:
Q w A^ 获得期望的检测次序 其中, 为[^餘第 ./行第 t /列后构成的矩阵;
的
3
S223: 通过所述检 i定置换矩阵 。
其中, 所述基于- 解过程包括: 通舊正的格拉姆:-施密^ 'schmidt正交 道矩阵的列进行重新排序;且所述排序和 程包括: , 对 ^ Λ ,… ].执行以下步骤;
叟索具有最小范数的列: - ai¾ mm ||q If , 并用 ku # 化 q,^//^到单位长度;
】, 计算投影 /^ ::: ; 其中, 括: 在 QL分解的过程中获得逆矩阵 LT1后, 对所
:基于预置的置换》 j的置换操作和 householder变换操作,以对所述逆 矩阵 U中的列进行重新排序; 其中, 所述预置的置换准 L:' - Ρ,ΐ 1; 其中, I 为第 f 次置换后的新矩阵; 为第 f次置换的置换矩阵。
(三) 有益效果
本发明的技术方案结合了发送端预编码以及接收端 MIMO子信遒检测算法。 在接收端 对完成预编码的导频信号进行信道估计, 并对其结杲进行基于排序的 QL分解或基于最小均 方误差扩展后的 QL分解, 得到一组置换矩阵 P 、 酉矩阵 Q以及对角线上元素有序的下三角 矩阵 L ; 通过酉矩阵 Q对接收信号进行处理, 在不放大噪声的基础上, 消除了某一层子信道 信号受到的来自于前面层数子信道的干扰; 利用下三角矩阵 L执行串行干扰蒯除, 逐层完成 MO子信道的检测; 最后利用置换矩阵 执行反置换操作 由于基于排序的 QL算法获得 角矩阵 L在高: ί好的降序特性,保证
减小了快速时变 M¾ 系统中各子信道之间的干扰; 且基于最小均方误差扩展得 QL 分解还能对噪声起到抑制作. 从而大大提升了基于 SVD 的传输方法在实际应用中的可行 性。 附图说明
图 1是本发明的快速时变 ΜΙΜΟ
ii
输示意图, 其中. 送:端利用反馈获得信道信息;
:速时变 MIMO系统中基于非码本预编码的传输方法的第二实施例 专
输示意图, 其中, 发送端利用 TDD信道互惠性获得信: ϊ¾ϊ息;
图 4示出了不同信道时间相关性的情况下,基于排序的 分解算法与不排序的 分解 算法获得的下三角矩阵 L的对角线元素按降序排列的概率的对比;
图 5示出了不同信道时间相关性以及不同信噪比条件下,本发明的快速时变 ΜΙΜΟ系统 中基于非码本预编码的传输方法与基于不排序的 QL分解算法的非码本预编码的传输方法的 误码率的性能比较。
图 6示出了在多普勒频移为 300Hz的时候, 在不同的信噪比条件下, 加入 ΤΙιΛο编码条 件下, 本发明的快速时变 Μ Ο 系统中基于非码本预编码的传输方法( SVD+QL- SORT, SVD+MMSE- SORT, SVD+MMSE- PSA )与基于未排序的 QL分解算法的传输方法( SVD+QL ) 的误比特率。 具体实施方式
本发明提出的时变 MMO系统中基于非码本预编码的传输方法, 结合附图和实施例说 明如下。
本发明技术方案的核心思想在于,发送端利用获.得的信道信息进行奇异值分解( SVD ) 得到预编码矩阵, 并根据该預编码矩阵对发送信号和插入的导频同时完成预编码; 接收端根 据完成预编码的导频进行信道估计, 并对信道估计结果进行基于排序的 QL分解 ·或基于最小 均方误差 展后的 QL分解得到一组置换矩阵 、 酉矩阵 Q和对角线元素有序的下三角矩阵 L ; 接着, 完成 Q "与接收信号相乘的搡作, 并利用该下三角矩阵 L的对角线元素在快速 变 下也能有序排列的这一性质, 对接收信号进行串行干扰蒯除处理, 并得到最终 MMO译码后 的信号。
需要说明的是, 发射端预编码的操作相当于完成了对下三角矩阵 L排序的过程, 当信道 时变相对较慢的时候, 该下三角矩阵 L的对角线元素有很大概率是按降序排列的。 但在高速 移动下, 信道的多普勒频移较大, 信道变化较快, 此时下三角降序特性无法得到保证, 在接 收端采用基于排序的 QL分解, 能够在高速移动的环境下仍然保证下三角矩阵 L的对角线元 如图 i所示,根据本发明的时变 MMO系统中基于非码本预编码的传输方法在发送端 包括以下步骤:
S11: 获得 T时刻的信道状态信息 u 其中, 发送端可以通过反馈获得信道状态信息 I T 特殊的, 在 TDD系统中, 发送端可以利用信道互惠性获得信道状态信息 H ;
S12: 对获取的信道状态信息 进行奇异值分解(SVD ), 得到左酉矩阵 右酉矩阵 、—τ以及对角线上元素为实数且按降序排列的对角阵 ;
其中, 对于采用反缋方式获取信道状态信息的系统, 步骤 12还包括存储该预编码 矩阵 v;— f及其共轭转置
S13: 利用酉矩阵 对经信道编码和符号调制后的发送信号和插入发送信号内的导频同 时完成预编码; 并将完成预编码的信号送入当前传输!:时刻的信道 Hf中。
仍参照图 1 , 根据本发明的时变 MIMD系统中基于非码本预编码的传输方法在接收端 包括以下步骤:
S14: 根据完成预编码的导频进行信道估计, 获得信道估计结果 H^„ ; 其中, 可以采用 的信道估计算法包括: 最小二乘法、 最小均方误差法和基于傅立叶变换的信道估计算法等; 其中, 对于采用反馈方式获取信道状态信息的系统, 步骤 1还包括: 反镄该信道估 计结果 , 其为包括 时刻传输发送的预编码矩阵 和信道矩阵 的等效信道信息;
S15: 对信道估计结果 进行基于排序的 QL分解, 生成一组置换矩阵 , 并同时获 得酉矩阵 Q和对角线上元素有序的下三角矩阵 L;
该步骤还可 ¾包括: 根据 MMES(MininuSni Mean Square Error,最小均方误差)准则对上述信 道估计结杲 H,,^进行信道扩展, 获得扩展后的信道 = [H ;o;iMJ ; 然后再对扩展后 的信道 0;'。 进行基于排序的 QL分解, 生成一组置换矩阵 , 并同时获得酉矩阵 Q和对角 线上元素有序的下王角矩阵 L。
本步骤中, 无论是未扩展还是扩展后的信道, 都可以通过以下两种方式实现基于排序的
QL分解: 对信道矩阵中的列进行重新排序后再进行 QL分解; 或, 在 QL分解的过程中实 现信道矩阵的重新排序。
第一种实现方法, 通过交换信道矩阵的列向量 (Q SORT排序方案), 使得分解后的下 三角矩阵的对角线元素降序排列; 其中, 对 或其扩展后的信道 0( = [11 ; 0;1¾ ]的 列进行重新排序的过程包括:
第一步: 初始化
i = 1, D , = HHH, P, = [i,2"..,MJ ( 1 ) 第二步: 循环执行
/ = arg min {det(A; /)}
Λ'; ^Exr(P.] ,/) ( )
Pi†, -Dr(Pi5 ) 其中, 为^删除第 行第 _/列后构成的矩阵, 那么 的余子式为 det(A 。 采用上述排 序方法, 获得的检测顺序为^ ¾,.·,,¾]; 根据捨测次序确定置换矩阵 。 然后就可以通 修 正的 Gram- Schmidt (格拉姆-施密特)方法对重排后的信道矩阵^进行 QL分解, 得到酉矩阵 Q和下三角矩阵 L。
进一步地,对于信道估计得到的等效信道 或其扩展后的信道 = : a ], 上述第二种实现方法还包括两种排序方法: 在 QL分解的过程中对下三角矩阵 L矩阵的对角线 元素进行次优排序 (QL SORT排序方案)和 PSA ( Post Sort Aigontlim, 最优后排序)。
具体地, 次优排序采用修正的 gnschmidt (格拉姆-施密特)正交化方法, 在分解的过 程中改变信道矩阵的列的顺序以获得较优的检测次序, 其实际上仍属于 QL SORT排序方案。 算法从 Q = H,L = 0开始, 确定具有最小范数的列, 并将其与最右边未处理过的矢量交换; 在 其长度归一化为单位值后, 确定相应的对角线元素 , 将剩下的列投影到新矢量 确定非 对角线元素 其中矢量 表示矩阵 Q的第 ^列。 算法过程为:
初始化: :L 0,Q H
for μ - Μ ·· ·\
在 Q剩下的列中搜索最小范
k.u - argrnin||qv|j_
将 Q的第 /列和 交换, 确定
直 , lq
归一化 到单位长度
for v' l"'/i - 1
计算投影 n v end
end
(3)
通过上述算法获得一个酉矩阵 Q、 一个对角线元素有序的下三角矩阵 L和一组置换矩阵 μ(\<μ≤Νι)α
具体地, 最优后棑序 (QL- PSA方案)将等效信道矩阵11 或扩展后信道矩阵进行 QL分 解, 通过分析 QL分解的协方差矩阵, 交换矩阵 L的行, 来实现对协方差矩阵对角线元素的降 序排列, 以减少错误传播。 选择! ^
此时获得的矩阵不再是三角矩阵, 为保持矩阵的三角特性, 矩阵的首行应该只包含一 个非零元素。 可以通过 Househoider变换来重新获得矩阵的三角特性: 一个行矢量 im可 以通过 Householder变换方法映射为平面或者是直线上一个等长的新矢量 bS>, 映射通过 aa与酉矩阵 Θ相乘来的得到。 将矩阵 j^L- '右乘以有矩阵 Θ ,可以迫使最后一行中除了第 一个元素以外的所有其他元素为零, 而不改变行的范数。 假设在干扰抵消的过程中,
1)χ(Λ' - 1)的矩阵 上。 继续以 ίΐ行范数的比较 义置换矩阵 Ρ, 再 -进行' Householder变 换得到右 ¾阵©.,。 依次类推直到得到所有层的结果。 最终获得的最优矩阵具有如下 的形式:
I". ·ί (4)
由于所有的置换矩阵和 Householder矩阵都是奇异矩阵, 满足 pp' p 最终可得到 下三角矩阵为:
L Θ ■ (■>3L:P,J ■P
(5)
置换和 Ho- usdiolder变换至多进 次, 并且有以下等式:
B - QL=> B , --- --- BP; ■P
(6)
其中 =0Θ Θ^。
最终得到酉矩阵 和下三角矩障 Lepi和置换矩阵 =
S16: 利用酉矩 :理, 得
(7)
= Lx + ή .
其中, y-( ';, ½,·,¾) 是处理后的接收信号, f = (λ,.υΑ 是接收天线上 ^
X = 是发送信号, 是预编码矩阵, 11 = ( ,《2,..., : 1功率为 的高斯白噪声 , β¾ ^^ ^^,^,…,^^是功率为 σ 的高斯白噪声 (R
, H, 为;时刻的信道状态信息; 这里, 表示转置, (.f表 , ^.)表示求期望:
S17: 对处理后的接收信号 y执行串行干扰删除 SIC操作; 其中, 可以通过对 MIMO子
信道逐层检测完成所述串行干扰 1]除处理过程;
M! , 且检测得到的第. /个子信道的发送信号为 则有:
v3 ~ 3ix ―
x3 = dec
- dec ( 8 )
L 其中, ^表示所述下三角矩阵 L的第 行、 第2列的元素, 为第 个子信道的经处理后的接 收信号, cfc 表示判决过程。
S18: 根_据置换矩阵 Ρ 对干扰消除后的信号 进行反置换搡作; 对于不同的排序方 式的反置换过程相同: 将步骤 S15中获得的一组置换矩阵 Ρ„^ ,· .. )转置后、 依次右 乘于所述步骤 S17中获得的干扰消除后的信号 , 即:
— γ pH . , , ρΑ' .
X V; ΓΛ';-1 ( 9 ) 下面将结合附图以及具体实施例分别对利用反馈获得信道信息的快速时变 FDD MMO 系统以及利用上下行信道互惠性获得信道信息的 TDD MIMO系统中的基于非码本预编码的
如图 2所示为本发明第- ^型图。 本实施例中, 发送端利用反馈 信息, 具体步骤包括:
S201: 发送端对初始的比特缀.发送信号 b进行信道编码; 优选地, 可以采用 Ί Λο码、 LDPC码等信道编码方式;
S203: 发送端插入导频; 优选地, 可以釆用频分、 时分、 码分等方式实现;
S205: 发送端提取/ -- r时刻的传输过程中所存储的 V^,, 其为酉矩阵 的共轭转置;
S206: 发送 将反馈的信道信息 Hf—W与该酉矩阵 V^,的共轭转置 O目乘, 恢复获得 i - τ时刻的信道状态信息《 , 即 ^.;
S207: 发送—端对 Η(— τ进行 SVD, 得到左酉矩阵 I , 右酉矩阵 和对角线上元素为实数 且按降序排列的对角阵 U
S208: 发送端保存该预编码矩阵(右酉矩阵 ) 的共轭转置
S209:发送端利用该预编码矩阵 v;— 7对所插入的导频和符号调制后的发 号同时完成预 编码过程; 即将该预编码矩阵 与符号调制后的发遂信号 X和导频信号同 相乘; 并发送信 号;
S210: 接收端 A接收天线接收信号 ^(.H- ¼ f, 并对完成预编码的该导频进行信道 估计, 得到信号估计结果 此时得到的不是信道状态信息 H 而是包含有发送预编码矩 阵的等效信道信息 H;. ;
S211: 接收端反馈该等效信道信息 ;
S213: 接收端通过将 Q 与接收信号 的处理, 且有:
( 10 )
- Lx + ή ·, ,
其中, γ
里, 表示转置, (.f表示共轭转置, ·£(.)表示求期望;
S214: 接收端对处理后的接收信号 y完成串行干扰删除处理(SIC ), 其中, 可以通过对 MIMO子信道逐层检测完成所述串行干扰蒯除处理过程; 检测顺序可以为从子信道 1到子 信道 ^ 且检测得到的第 个子信道的发遊 号为 则有:
其中, 表示所述下三角矩阵 L的第 行、 第 列的元素, 为第^ /个子信道的经处理后的接 收信号, 表示判决过程;
步骤 215: 接收端对执行干扰消除操作后的信号 进行反置换操作, 得到最终检测到的 信号 i = ... _;;其中 ,Ρ = l,... l )是步骤 S212中基于排序的 QL分解巾获得的 组 置换矩阵;
步骤 216: 接收端完成 MMO子信道检测后信号的符号解调; 其中, 解调方式与发送端 的符号调制方式相对应;
如图 3所示为本发明第二实施例的传输模型面; 本实施例中,发送端利用 TDD系统信道 互惠性获得信道信息, 具体步骤包括:
S301: 发送端对初始的比特级.发送信号 b进行信道编码; 优选地, 可以采用 Turbo . LDPC码等信道编码方式;
S302: 发送端对信道编码后的比特级信号进行符号调制, 得到 ; 优选地, 可以采用 QPSK、 16QAM 64QAM等调制方式;
S303: 发送端插入导频; 优选地, 可以采用频分、 时分、 码分等方式实现;
S304: 发送端直接利用 TDD的信道互惠性, 获得卜 r时刻的信道扰态信息!^;
S305: 发送端对该信道状态信息 Η,—进行奇异值分解(SVD), 得到左酉矩阵 右酉 矩阵 和对角线上元素为实数且.按降序排列的对角阵∑f— f;
S306: 发送端利用该右酉矩阵 (预编码矩阵 对导频和符号调制后的发送信号同时完成 预编码过程; 即将该预编码矩阵 V_与符号调制后的发送信号 X和导频信号同时相乘; 并发送 信号;
S307: 接收端 接收天线接收信号 (ϋ,.,., ), 并对完成预编码的该导频进行信道 估计, 得到信道估 ϋ结果 : ; 此时得到的不是信道状态信息11;而是包含发送预编码矩阵的 等效信道信息 Η, ;
S308: 接收端刘 I道估计结果 Mf ^或其扩展后的矩阵 Hi>B 进行基于排序的 QL分解, 得到一组置换矩阵 , 并同时获得酉矩阵 Q和对角线上元素有序的下三角矩阵 L;
S309: 接收端通过将 与接收信号 f = ^. ,...,: 相乘实现对接收信号的处理, 且有:
:= „ 》 (12) 其中, ^(^2,...,¾) 是处理后的接收信号,
是接收天线上的接收信号, χ :::^, 2,...¾;ί是发送信号, ^是预编码矩阵, ^( ,《2,...,¾)是功率为 的高斯白噪声 , fi?::: Q : ..., Aiif是功率为 的高斯白噪声 (R
), 这 里, (.f表示转置, (.f表示共轭转置, £(.)表示求期望;
S310: 接收端对处理后的接收信号 y完成串行干扰删除处理(SIC), 其中, 可以通过对 MIMO子信道逐层检测完成所述串行干扰蒯除处理过程; 检测顺序可以为从子信道 ::〗到子 信道 = , 且检测得到的第 个子信道的发送信号为 则有:
i,::: dec 2i_
■·'?.— ?.ι·
其中, Z ,表示所述下三角矩阵 L的第 行、 第/列的元素, 为第 个子信道的经处理后的接 收信号, ofec(.)表示判决过程;
S312: 接收端完成 ΜίΜΟ子信道检测后信号的符号解调; 其中, 解调方式与发送端的符 号调制方式相对应;
S313: 接收端对解调后比特级信号进行信道译码, 得到译码后信号 β ; 其中, 信道译码 方式与发送端的信道编码方式相对应。
综上所述, 本发明结合发送端预编码和接收端 ΜΜΟ子信道检测算法, 其中, 发送端对 发送信号和导频同时进行预编码, 这祥在接收端对导频 畫道估计得到的不是信道状态信息 Η¾ , 而是包含发送预编码矩阵的等效信道信息 Η^Μ ; 接收端进一步对 11^或基于最小均方 误差扩展后的 QL藤,得到一组置换矩阵 Ρ„、酉矩阵 Q和对角线元素为实数的下三角矩阵 L。 通过酉矩阵 Q对接收信号进行处理, 在不放大噪声的基础上., 消除了某一层子信道信号受到 的来自于前靣层数子信道的干扰; 利用下三角矩阵 L扰行串行千扰删除, 逐层完成 ΜΪΜΟ子 信道的检测;最后利用置换矩阵 执行反置换操作。由于经过重新排序处理的下三角矩阵 L在 高速移动的环境中伪能保持较好的降序特性, 保证了串行干扰蒯除技术的性能, 减小了快速. 时变 MIMO系统中各子信道之间的干扰; 同时, 当对某一层子信道进行检测时, 在完全消除 未检测信道对其干扰的基础上, 本发明技术方案的接收算法可以进一步抑制已检测信道的误 码传播带来的影响。
下靣将给出本发明的传输方案与现有的其它传输方案的比较, 以使本发明的优势及特征 更加明显。 道时间相关性条件^ , 三角矩阵 L对角线元素按照降^排列的概率; 由图 5 可以看 , 采用本发明技术方案后, 在复杂度稍微增大的情况下, 降序棑列的概率有明显的提升。 图 5 示出了不同信道时间相关性、 不同信噪比的条件下, 基于未排序 QL分解接收算法的传输方 法 (SVD^HQL ) 与本发明的基于棑序 QL分解接收算法的传输方法的一个实施侧 (采用前述 QL- SORT排序方案)。 图 6示出了在采用 ΤυΛο 1/3编码, QPSK调制时, 基于未排序的 QL 分解接收算法的传输方法(SVD+QL ) 与本发明的基于排序 分 接收算法的传输方法的 实施例 SVIHQL SORT (对信道矩阵采.用前述 QL~SORT排序方案的实施例), SVD+QL- PSA
(对信道矩阵采用前述 QL-PSA的排序方案的实施倒), SVD+MMSE SORT (对信道矩阵进 行基于最小均方误差准则的扩展后,再执行前述 QL- SORT排序的实施例), SVD+MMSE- PSA
(对信遒矩阵进行基于最小均方误差准则的扩展后, 再执行前述 QL- PSA排序的实施例) 的 BER性能对比;由图 6中可以看出,本发明中的 SVD+QL- SORT方案以及 SVEH-MMSE-SORT 方案在复杂度增加不大的情况下,相比 SVD+QL方案能获得更大的性能增益: Turbo 1/3编码, QPSK调制下, 分别获得^ L8dB与 3de的性能增益。
以上实施方式仅用于说明本发明, 而并非对本发明的限制, 有关技术领域的普通技术人 员, 在不脱离本发明的精神和范围的情况下, 还可以做出各种变化和变型, 因此所有等同的 技术方案也属于本发明的范畴, 本发明的专利保护范围应由权利要求限定。
Claims
1、 一种快速时变 MIMO系统中基于非码本预编码的传输方法, 其特征在于, 对于 接收端,
S1 : 接收由发送端发送的、 经同时预编码处理的信号 ΐ和导频, 根据所述导频完成信道 估计, 获得等效信道 H ;
S2: 对所 ·述等效信道 ]¾„进行基于排序的 QL分解, 生成一组置换矩阵 并同时获得 酉矩阵 Q和对角线上元素有序的下三角矩阵 L;
S3: 根据所述酉矩阵 Q对信号 进行处理, 获得处理后的接收信号 = 9 ;
S4: 根据所述下三角矩阵 L对所述处理后的接收信号 y = Q 进行串行干扰消除, 获得干 扰消除后的信号 ip ;
S5: 根据所述置换矩阵 P„对所述千扰消除后信号 进行反置换操作, 获得信号 i ;
S6: 对所述信号 i完成解调和译码。
2、 如权利要求 1所述的快速时变 ΜίΜΟ系统中基于非码本预编码的传输方法, 其特
Α一 nr φ
'ΓΤ" 逸 ,
SI 1 : 获取 η时刻的信道状态信息 ΜΜ并对其进行奇异值分解, 获得预编码矩阵 υ S12: 根据所述预编码矩阵 ^ 对经信道编码和符号调制的待发送信号 Χ以及插入的导 频同时扰行预编码, 并将预编码后的信号送入 ί时刻传输信道中。
3、 如权利要求 1所述的快速时变 ΜΙΜΟ系统中基于非码本预编码的传输方法, 其特 征在于, 所述步骤 S2还包括:
S21 : 根据最小均方误差准则对所述等效信道 进行信道扩展, 获得扩展后的信道 , 1;Ί· = [Η,χ ^Ι^ ]
S22: 对所述扩展后的信道 进行基于排序的 QL分解, 生成一组置换矩阵 ^, 并同 时获得酉矩阵 Q和对角线上元素有序的下三角矩阵 L;
其中, σ„为加性高斯白噪声功率的均方根。
4、 如权利要求 i或 3所述的快速时变 ΜίΜΟ系统中基于非码本预编码的传输方法, 其特征在于, 所述基于排序的 过程包括: 基于行列式准则对信道矩阵中的列进行 重新排序, 以获得期望的串行干扰检测次序; 以及, 基于修正的 Gram- Schmidt格拉姆- 施密特方法对重新排序后的信道矩阵进行 QL分解。
5、 如权利要求 4所述的快速时变 MIMO系统中基于非码本预编码的传输方法, 其特征在 于, 所述重新排序的过程包括:
S221: '初台化: i - 1,□, HMH, I' ::[1,2,...,MJ
S222: ·'
- arg min {clet(A;. ,)}
P.,, -Dr(P,J)
A..5 i i + 1
获得期望的检测次序 [½ , ¾]; 其中, ^为^删除第 第 列后; f矩阵; S223: 通蕭 I确定置换矩阵 Ρ,,
6、 如权利要求〗或 3所述的快速时变 MIMO系统中基于非码本预编码的 W
其特征在于, 所述基于排序的(_ 过程包括: 通过修正的格拉姆—施密特 gram schmidt 正交化方法, 在 QL分解的过程中对信道矩阵的列进行重新排序
7、 如权利要求 6所述的快速时变 MIMO系统中基于非码本预编码
征在于, 所述排序和分解的过程包括:
822Γ: 初始化: L = 0,Q = , 对/^ 执行以下步骤;
S222': , :的列中搜索具有最小范数的列: ,, argmin 并用 ^替. 的 第/^列;
, 归一化 q : /^到单位长度;
其
S224': 对于! ^ΐ... — ·ί, 计算投影 _L 特 其中' qn Λ^; 为矩阵 Q中的第 列。
8、 如杈利要求 1或 3所述的快速时变 ΜΙΜΟ系统中基于非码本预编码的传输方法, 其特征在于, 所述基于排序的 QL分解过程包括: 在 QL分解的过程中获得逆矩阵 后, 对 所述逆矩!^交替执行多次基于预置的置换准则的置换搡作和 househoicter变换搡作, 以对所述 逆矩阵 L 中的列进行重新排序; 其中, 所述预置的置换准则包括: '; 其中, I 为 第 次置换后的新矩阵; R为第 i次置换的置换矩阵。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201010567209.8 | 2010-11-25 | ||
CN2010105672098A CN102013958A (zh) | 2010-11-25 | 2010-11-25 | 快速时变mimo系统中基于非码本预编码的传输方法 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2012069007A1 true WO2012069007A1 (zh) | 2012-05-31 |
Family
ID=43843990
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2011/082936 WO2012069007A1 (zh) | 2010-11-25 | 2011-11-25 | 快速时变mimo系统中基于非码本预编码的传输方法 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN102013958A (zh) |
WO (1) | WO2012069007A1 (zh) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102013958A (zh) * | 2010-11-25 | 2011-04-13 | 北京邮电大学 | 快速时变mimo系统中基于非码本预编码的传输方法 |
CN103973409A (zh) * | 2013-01-31 | 2014-08-06 | 富士通株式会社 | 预编码装置和包含该预编码装置的发射机 |
US9281885B2 (en) * | 2014-04-07 | 2016-03-08 | Imagination Technologies, Llc | Reordering of a beamforming matrix based on encoding |
CN112491826B (zh) * | 2020-11-13 | 2022-03-11 | 重庆大学 | 大规模mimo系统下基于回传方式的psa检测方法 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20080021323A (ko) * | 2006-09-04 | 2008-03-07 | 삼성전자주식회사 | 다중 안테나 통신시스템의 송신신호 검출을 위한 수신장치및 방법 |
CN101854237A (zh) * | 2010-06-01 | 2010-10-06 | 北京邮电大学 | 时变mimo系统中基于非码本预编码的传输方法 |
CN102013958A (zh) * | 2010-11-25 | 2011-04-13 | 北京邮电大学 | 快速时变mimo系统中基于非码本预编码的传输方法 |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101136721A (zh) * | 2007-09-30 | 2008-03-05 | 东南大学 | 基于次最优排序的混合判决反馈分层检测方法 |
CN101409604B (zh) * | 2007-10-09 | 2011-04-20 | 鼎桥通信技术有限公司 | 多输入多输出系统中的串行干扰消除方法 |
CN101594217A (zh) * | 2008-05-28 | 2009-12-02 | 中国移动通信集团公司 | 一种多用户多输入多输出系统的数据发送方法及装置 |
-
2010
- 2010-11-25 CN CN2010105672098A patent/CN102013958A/zh active Pending
-
2011
- 2011-11-25 WO PCT/CN2011/082936 patent/WO2012069007A1/zh active Application Filing
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20080021323A (ko) * | 2006-09-04 | 2008-03-07 | 삼성전자주식회사 | 다중 안테나 통신시스템의 송신신호 검출을 위한 수신장치및 방법 |
CN101854237A (zh) * | 2010-06-01 | 2010-10-06 | 北京邮电大学 | 时变mimo系统中基于非码本预编码的传输方法 |
CN102013958A (zh) * | 2010-11-25 | 2011-04-13 | 北京邮电大学 | 快速时变mimo系统中基于非码本预编码的传输方法 |
Non-Patent Citations (1)
Title |
---|
FANG SHU ET AL.: "Non-Unitary Precoding Base on Codebook for Multi-user MIMO System with Limited Feedback", JOURNAL OF ELECTRONICS & INFORMATION TECHNOLOGY, vol. 30, no. 10, October 2008 (2008-10-01), pages 2419 - 2422 * |
Also Published As
Publication number | Publication date |
---|---|
CN102013958A (zh) | 2011-04-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP4845738B2 (ja) | 線形プリコーディングされた信号のマルチアンテナ伝送方法、対応するデバイス、信号、および受信方法 | |
KR100979132B1 (ko) | 무선통신시스템에서 간섭 제거 장치 및 방법 | |
CN101841375A (zh) | 一种多输入多输出单载波分块传输系统的检测方法及装置 | |
CN101278500A (zh) | 无线通信设备 | |
WO2007143172A2 (en) | A method and apparatus for distributed space-time coding in wireless radio networks | |
CN101854237B (zh) | 时变mimo系统中基于非码本预编码的传输方法 | |
JP2009060616A (ja) | Mimoシステムに用いる低複雑度の信号検出方法及びその検出装置 | |
CN101483503B (zh) | 多输入多输出系统数据发送、接收方法及装置 | |
JP4381901B2 (ja) | 通信路推定及びデータ検出方法 | |
CN101888287B (zh) | 一种多输入多输出接收机信号检测方法及装置 | |
JP2008124843A (ja) | 無線受信装置 | |
Sakzad et al. | On complex LLL algorithm for integer forcing linear receivers | |
WO2012069007A1 (zh) | 快速时变mimo系统中基于非码本预编码的传输方法 | |
CN101227254A (zh) | 一种在多入多出系统中v-blast的检测方法 | |
CN102123114A (zh) | 用于mimo系统中的基于非码本预编码的传输方法 | |
WO2011032421A1 (zh) | 一种基于联合收发端信息的多路波束形成方法及系统 | |
US8576959B2 (en) | Receiver with prefiltering for discrete fourier transform-spread-orthogonal frequency division multiplexing (DFT-S-OFDM) based systems | |
US20040071235A1 (en) | Low complexity high performance decoder and method of decoding for communications systems using multidimensional signaling | |
CN108418619B (zh) | 一种信号检测方法及装置 | |
Nakajima et al. | Iterative FDIC using 2D-MMSE FDE for turbo-coded HARQ in SC-MIMO multiplexing | |
Honjo et al. | Computational complexity reduction of MLD based on SINR in MIMO spatial multiplexing systems | |
JP7541697B2 (ja) | Mimo復調方法および端末装置 | |
KR101342626B1 (ko) | 다중 입력 다중 출력 시스템에서 반복적 트리 검색에기반한 저 복잡도 신호 검출 장치 및 방법 | |
CN111064684B (zh) | 上行链路空间调制单载波频域联合均衡方法 | |
JP5535759B2 (ja) | 無線通信システム、送信装置、受信装置、通信方法、送信方法及び受信方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 11843698 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC DATED 17.10.2013 (FORM 1205A) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 11843698 Country of ref document: EP Kind code of ref document: A1 |