WO2022237028A1 - Low-complexity ml detection method - Google Patents
Low-complexity ml detection method Download PDFInfo
- Publication number
- WO2022237028A1 WO2022237028A1 PCT/CN2021/118574 CN2021118574W WO2022237028A1 WO 2022237028 A1 WO2022237028 A1 WO 2022237028A1 CN 2021118574 W CN2021118574 W CN 2021118574W WO 2022237028 A1 WO2022237028 A1 WO 2022237028A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- symbol
- detection
- constellation point
- complexity
- euclidean distance
- Prior art date
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 89
- 238000000034 method Methods 0.000 claims description 8
- 230000005540 biological transmission Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 abstract description 7
- 238000004422 calculation algorithm Methods 0.000 description 19
- 238000007476 Maximum Likelihood Methods 0.000 description 6
- 239000011159 matrix material Substances 0.000 description 4
- 238000007792 addition Methods 0.000 description 3
- 238000013139 quantization Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000005562 fading Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0045—Arrangements at the receiver end
- H04L1/0054—Maximum-likelihood or sequential decoding, e.g. Viterbi, Fano, ZJ algorithms
Definitions
- the invention relates to the technical field of communication modulation, in particular to a low-complexity ML detection method.
- SM spatial Modulation
- Common detection methods at the receiving end include maximum likelihood (Maximum Likelihood, ML) detection and suboptimal detection (Sub Optimal, SO).
- ML detection is optimal detection, which is an algorithm for joint detection of transmit antenna index and modulation symbol. Although this algorithm can obtain the optimal performance, it needs to search through the joint sample space of all possible transmit antenna indexes and constellation point symbols, and its complexity is very high. Especially in the case of multi-antenna SM systems or high-order constellation modulation, the complexity tends to rise exponentially.
- Suboptimal detection is a detection algorithm based on the maximum ratio combining MRC criterion. This algorithm screens out the most likely transmitting antenna index by establishing the reliability standard first, and then uses the constellation demodulation function to detect the index corresponding to the estimated transmitting antenna index. zodiac point symbols. Although this algorithm reduces computational complexity, its performance is not as good as ML detection.
- the patent document CN105245477B discloses a low-complexity differential detection algorithm.
- the bit filling scheme and the transmission bit calculation formula are used to construct a symbol matrix at the transmitting end of the differential space modulation system, and the received signal is obtained at the receiving end through differential transformation and transmission, and then Based on the formula structure of the differential maximum likelihood detection algorithm, the search method of column detection is adopted for the receiving matrix, and the signal judgment is performed column by column, and the serial numbers of the transmitting symbols and transmitting antennas are obtained.
- the bit and transmitting symbol, bit and transmitting antenna The corresponding relationship of the sequence numbers is inversely mapped to information bits.
- the method is obtained by combining the low-complexity detection algorithm under the space modulation with the formula structure formed by the maximum likelihood detection algorithm under the differential space modulation system.
- the research shows that by using the LC-DSM algorithm, the complexity of the differential maximum likelihood detection algorithm is greatly reduced under the premise of ensuring no loss in algorithm performance.
- Patent document CN109818663A discloses a low-complexity differential quadrature space modulation detection method (abbreviated as LC-DQSM algorithm), which divides the received signal matrix into signal vectors and considers the active antenna as one or Two cases are used to estimate the transmit antenna index sequence and transmit symbols step by step.
- the first case uses the idea of signal vector detection
- the second case uses the idea of improved block sorting minimum mean square error to detect.
- the simulation results show that, under the premise of ensuring the performance of the algorithm, the complexity of the LC-DQSM algorithm is greatly reduced compared with the maximum likelihood detection algorithm (ML).
- ML maximum likelihood detection algorithm
- the purpose of the present invention is to provide a low-complexity ML detection method to solve the problems raised in the above-mentioned background technology.
- a low-complexity ML detection method comprising the following steps:
- the constellation point symbol g j with noise n is calculated by the following formula:
- the estimated and detected constellation point symbol Q(g j ) argmin s ⁇ s
- 2 is obtained through the constellation demapper function Q( ⁇ ).
- the Euclidean distance d j
- 2 between the received signal y and the constellation point symbol x m through the channel gain h j is calculated by the following formula.
- the following method is used to determine whether the estimated detection of the transmitted signal x l and the symbol sequence number l is correct:
- N t is the transmitting antenna
- the method further includes: if the estimated detection of the transmitted symbol x l and the symbol sequence number 1 is inappropriate, repeating steps S2 to S4 until all N t transmitting antennas are traversed.
- a low-complexity ML detection method provided by the present invention filters out the most likely transmitted symbols by calculating the constellation point symbol without quantization processing, and defining the Euclidean distance standard between the constellation point symbol and all constellation point symbols, and then combines The channel state information and the received signal are used to estimate the antenna index for detection. On the premise that the detection performance is approximately unchanged, the complexity of ML detection is reduced, and the redundant calculation is reduced.
- Fig. 1 is a flow chart of a low-complexity ML detection method provided by the present invention
- Fig. 2 is a relationship diagram between BER performance and SNR of three detection methods under different modulations.
- the present invention provides a kind of low complexity ML detection method, comprises the following steps:
- the transmitting antenna and the receiving antenna of the spatial modulation system are set to be N t and N r respectively, and the channel is a flat Rayleigh fading channel.
- the signal received by the receiver is:
- ⁇ is the average received signal-to-noise ratio
- h j is the jth column of the channel matrix H
- n is additive Gaussian white noise, independent distribution and obeys the random variable with mean value 0 and variance 1.
- step S2 when the index of the activated transmitting antenna is j, what the receiving end actually obtains is the unquantized constellation point symbol g j , and the constellation point symbol g j mixed with noise n is calculated by the following formula:
- the quantization function that is, the constellation demapper function Q( ⁇ )
- the estimated and detected constellation point symbol x m is obtained, where the expression of x m is as follows:
- the embodiment of the present invention further discloses that by comparing the sizes of d j and d min , it is judged whether the estimated detection of the transmitted signal x l and the symbol sequence number l is correct, and the specific steps are as follows:
- N t is the transmitting antenna
- the method in this embodiment further includes: if the estimation and detection of the transmitted symbol x l and the symbol sequence number l are not suitable, repeating steps S2 to S4 until all N t transmitting antennas are traversed.
- the present invention further compares the computational complexity of the proposed low-complexity ML detection method with the ML detection method, because the number of complex additions is similar in these two detection methods, so the number of additions is not considered, only the number of complex multiplications is considered .
- the proposed low-complexity ML detection has the complexity of
- the proposed low-complexity ML detection method has lower computational complexity than the ML detection method. This is because ML detection requires all selected transmit antennas to combine all transmit symbols to estimate transmit information, which may bring a lot of unnecessary calculations.
- the proposed low-complexity ML detection utilizes the estimated symbols and corresponding channel state information to help decode the symbols to estimate the transmit antenna index and reduce redundant computation.
- FIG. 2 is a relationship curve between BER performance and SNR under 16QAM and 256QAM modulation for ML detection, proposed low-complexity ML detection and suboptimal detection methods.
- the transmit power, the number of transmit antennas and the number of receive antennas are set to 4 respectively.
- the proposed low-complexity ML detection and ML detection have approximate BER performance when the modulation is 16QAM, and both outperform the sub-optimal detection.
- BER 10 -6
- the proposed detection has a 2.5dB SNR gain over the sub-optimal detection.
- the BER performance of these three detection methods is also improved.
- the BER performance of all three detection methods decreases when the modulation is 256QAM, but the performance of the proposed low-complexity ML detection and ML detection are still similar, and all outperform the suboptimal detection.
Landscapes
- Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Digital Transmission Methods That Use Modulated Carrier Waves (AREA)
- Radio Transmission System (AREA)
Abstract
A low-complexity ML detection method, comprising the following steps: S1, initializing a transmitting antenna index j, a transmitting signal xl, a symbol serial number l, and a Euclidean distance di between a receiving signal y and the transmitting signal xl through a channel gain hj; S2, under the condition that signal state information hj is known, calculating a constellation point symbol gj mingled with noise n; S3, calculating Euclidean distances d between the constellation point symbol gj and all constellation point samples in a constellation sample space S, and selecting a minimum Euclidean distance dmin to obtain an estimated and detected constellation point symbol xm; and S4, calculating a Euclidean distance dj between the receiving signal y and the constellation point symbol xm through the channel gain hj, comparing the Euclidean distance dj with dmin, and determining, according to a comparison result, whether estimation and detection of the transmitting signal xl and the symbol serial number l are correct. On the premise that the detection performance is approximately unchanged, the complexity of ML detection is reduced, and redundant calculation is reduced.
Description
本发明涉及通信调制技术领域，尤其涉及一种低复杂度的ML检测方法。The invention relates to the technical field of communication modulation, in particular to a low-complexity ML detection method.
空间调制(Spatial Modulation，SM)技术的传输原理是将发送的信息比特中的一部分映射为天线索引，另一部分映射为传统的幅度相位调制符号。在接收端常见的检测方法有最大似然(Maximum Likelihood，ML)检测和次优检测(Sub Optimal，SO)。ML检测是最优检测，是一种联合检测发射天线索引和调制符号的算法。该算法虽可以获得最优的性能，但需要遍历搜索所有可能的发射天线索引和星座点符号的联合样本空间，其复杂度非常高。特别在多天线SM***或高阶星座调制的情况下，复杂度呈指数级上升趋势。另外在传输率较高情况下，运算开销很大，较难应用于实际的***中。次优检测是一种基于最大比合并MRC准则的检测算法，该算法通过先建立可靠性的标准筛选出最有可能的发射天线索引，再运用星座解调函数检测出与估计的发射天线索引对应的星座点符号。虽然该算法降低了运算复杂度，但性能不如ML检测。The transmission principle of spatial modulation (Spatial Modulation, SM) technology is to map part of the transmitted information bits into antenna indexes, and the other part into traditional amplitude-phase modulation symbols. Common detection methods at the receiving end include maximum likelihood (Maximum Likelihood, ML) detection and suboptimal detection (Sub Optimal, SO). ML detection is optimal detection, which is an algorithm for joint detection of transmit antenna index and modulation symbol. Although this algorithm can obtain the optimal performance, it needs to search through the joint sample space of all possible transmit antenna indexes and constellation point symbols, and its complexity is very high. Especially in the case of multi-antenna SM systems or high-order constellation modulation, the complexity tends to rise exponentially. In addition, when the transmission rate is high, the calculation overhead is very large, and it is difficult to apply it to the actual system. Suboptimal detection is a detection algorithm based on the maximum ratio combining MRC criterion. This algorithm screens out the most likely transmitting antenna index by establishing the reliability standard first, and then uses the constellation demodulation function to detect the index corresponding to the estimated transmitting antenna index. zodiac point symbols. Although this algorithm reduces computational complexity, its performance is not as good as ML detection.
在估计检测天线索引和星座点符号的过程中，所涉及到的复数乘法和加法的计算次数对检测算法的计算复杂度起主要的影响作用。检测性能与计算复杂度是一对矛盾的，但至关重要的两个指标，影响算法性能。因此研究出一种具有优良的检测性能且求解复杂度较低的检测算法具有重要意义。In the process of estimating the detection antenna index and the constellation point symbol, the calculation times of complex multiplication and addition involved play a major role in the calculation complexity of the detection algorithm. Detection performance and computational complexity are a pair of contradictions, but the two crucial indicators affect the performance of the algorithm. Therefore, it is of great significance to develop a detection algorithm with excellent detection performance and low solution complexity.
专利文献CN105245477B公开了一种低复杂度的差分检测算法，首先，在差分空间调制***发射端应用比特填充方案与传输比特计算公式构造符号矩阵，经过差分变换与发送在接收端得到接收信号，再以差分最大似然检测算法的算式结构为基础，对接收矩阵采取分列检测的搜索方式，逐列进行信号判决，得 到发射符号与发射天线的序号，最终根据比特与发射符号、比特与发射天线序号的对应关系逆映射为信息比特。该方法是将空间调制下低复杂度的检测算法与差分空间调制***下最大似然检测算法形成的算式结构相结合而得到的。研究表明通过使用LC-DSM算法，在保证算法性能没有损失的前提下，相对于差分最大似然检测算法复杂度得到了大大的降低。The patent document CN105245477B discloses a low-complexity differential detection algorithm. First, the bit filling scheme and the transmission bit calculation formula are used to construct a symbol matrix at the transmitting end of the differential space modulation system, and the received signal is obtained at the receiving end through differential transformation and transmission, and then Based on the formula structure of the differential maximum likelihood detection algorithm, the search method of column detection is adopted for the receiving matrix, and the signal judgment is performed column by column, and the serial numbers of the transmitting symbols and transmitting antennas are obtained. Finally, according to the bit and transmitting symbol, bit and transmitting antenna The corresponding relationship of the sequence numbers is inversely mapped to information bits. The method is obtained by combining the low-complexity detection algorithm under the space modulation with the formula structure formed by the maximum likelihood detection algorithm under the differential space modulation system. The research shows that by using the LC-DSM algorithm, the complexity of the differential maximum likelihood detection algorithm is greatly reduced under the premise of ensuring no loss in algorithm performance.
专利文献CN109818663A公开了一种低复杂度差分正交空间调制检测方法(简称为LC-DQSM算法)，该方法在将接收信号矩阵拆分为信号向量的基础上，通过考虑激活天线为一根或两根两种情况来分步估计发射天线索引序列和发射符号。其中第一种情况采用信号向量检测思想，第二种情况采用改进的分块排序最小均方误差思想检测。仿真结果表明，在保证算法性能的前提下，LC-DQSM算法相对于最大似然检测算法(ML)复杂度得到了大幅度的降低。Patent document CN109818663A discloses a low-complexity differential quadrature space modulation detection method (abbreviated as LC-DQSM algorithm), which divides the received signal matrix into signal vectors and considers the active antenna as one or Two cases are used to estimate the transmit antenna index sequence and transmit symbols step by step. The first case uses the idea of signal vector detection, and the second case uses the idea of improved block sorting minimum mean square error to detect. The simulation results show that, under the premise of ensuring the performance of the algorithm, the complexity of the LC-DQSM algorithm is greatly reduced compared with the maximum likelihood detection algorithm (ML).
发明内容Contents of the invention
本发明的目的在于提供一种低复杂度的ML检测方法，以解决上述背景技术中提出的问题。The purpose of the present invention is to provide a low-complexity ML detection method to solve the problems raised in the above-mentioned background technology.
本发明是通过以下技术方案实现的：一种低复杂度的ML检测方法，包括下列步骤：The present invention is achieved through the following technical solutions: a low-complexity ML detection method comprising the following steps:
S1、初始化发射天线索引j、发射信号x
l、符号序列号l以及接收信号y与发射信号x
l之间经过信道增益h
j之间的欧式距离d
i；
S1. Initialize the transmitting antenna index j, the transmitting signal x l , the symbol sequence number l and the Euclidean distance d i between the received signal y and the transmitted signal x l through the channel gain h j ;
S2、在已知信号状态信息h
j的条件下，计算夹杂噪声n的星座点符号g
j；
S2. Under the condition of known signal state information hj , calculate the constellation point symbol gj mixed with noise n ;
S3、计算星座点符号g
j与星座样本空间S中所有星座点样本之间的欧氏距离d，并选取其中最小的欧式距离d
min，通过星座解映射器函数，得到估计检测的星座点符号x
m；
S3. Calculate the Euclidean distance d between the constellation point symbol g j and all constellation point samples in the constellation sample space S, and select the smallest Euclidean distance d min , and obtain the estimated and detected constellation point symbol through the constellation demapper function x m ;
S4、计算接收信号y与所述星座点符号x
m之间进过信道增益h
j之间的欧式距 离d
j，比较欧式距离d
j与d
min，根据比较结果，判断发射信号x
l、符号序列号l的估计检测是否正确。
S4. Calculate the Euclidean distance d j between the received signal y and the constellation point symbol x m through the channel gain h j , compare the Euclidean distance d j and d min , and judge the transmitted signal x l and symbol according to the comparison result Whether the estimated detection of sequence number l is correct.
优选的，通过下式计算夹杂噪声n的星座点符号g
j：
Preferably, the constellation point symbol g j with noise n is calculated by the following formula:
其中，
为已知信道状态信息h
j的共轭转置，y为接收信号，
为信道增益h
j的F范数的平方值。
in, is the conjugate transpose of the known channel state information h j , y is the received signal, is the square value of the F norm of the channel gain h j .
优选的，通过下式计算星座点符号g
j与星座样本空间S中所有星座点样本之间的欧氏距离d：d＝||g
j-s||
2，其中s为星座点样本。
Preferably, the Euclidean distance d between the constellation point symbol g j and all the constellation point samples in the constellation sample space S is calculated by the following formula: d=||g j −s|| 2 , where s is the constellation point sample.
优选的，通过星座解映射器函数Q(·)，得到估计检测的星座点符号Q(g
j)＝argmin
s∈s||g
j-s||
2。
Preferably, the estimated and detected constellation point symbol Q(g j )= argmin s∈s ||g j −s|| 2 is obtained through the constellation demapper function Q(·).
优选的，通过下式计算接收信号y与所述星座点符号x
m之间进过信道增益h
j之间的欧式距离d
j＝||y-h
jx
m||
2。
Preferably, the Euclidean distance d j =||yh j x m || 2 between the received signal y and the constellation point symbol x m through the channel gain h j is calculated by the following formula.
优选的，通过下列方式判断发射信号x
l、符号序列号l的估计检测是否正确：
Preferably, the following method is used to determine whether the estimated detection of the transmitted signal x l and the symbol sequence number l is correct:
若d
j<d
min，则说明发射符号x
l、符号序列号l的估计检测正确，分别更新初始的x
l、l、d
i为x
m、j、d
j，此时若j＝N
t，则输出结果x
m、j、d
j，否则继续估计检测天线索引j+1；
If d j <d min , it means that the estimated detection of the transmitted symbol x l and the symbol sequence number l is correct, and the initial x l , l, and d i are respectively updated as x m , j, and d j , at this time, if j=N t , then output the results x m , j, d j , otherwise continue to estimate the detection antenna index j+1;
若d
j>d
min，则说明发射符号x
l、符号序列号l的估计检测不合适，此时若j＝N
t，则输出结果即初始值x
l、l、d
i。否则将继续估计下个天线索引j+1；
If d j >d min , it means that the estimation and detection of transmitted symbol x l and symbol sequence number l are not suitable. If j=N t at this time, the output results are the initial values x l , l, d i . Otherwise, continue to estimate the next antenna index j+1;
式中，N
t为发射天线。
In the formula, N t is the transmitting antenna.
所述方法还包括：若发射符号x
l、符号序列号l的估计检测不合适，重复步骤S2至S4，直到遍历所有的N
t发射天线为止。
The method further includes: if the estimated detection of the transmitted symbol x l and the symbol sequence number 1 is inappropriate, repeating steps S2 to S4 until all N t transmitting antennas are traversed.
与现有技术相比，本发明达到的有益效果如下：Compared with prior art, the beneficial effect that the present invention reaches is as follows:
本发明提供的一种低复杂度的ML检测方法,通过计算未经过量化处理星座点符号，并定义该星座点符号与所有星座点符号的欧式距离标准筛选出最有可能的发射符号，然后结合信道状态信息和接收信号获得估计检测的天线索引,在检测性能近似不变的前提下，降低了ML检测的复杂度,减少了冗余计算。A low-complexity ML detection method provided by the present invention filters out the most likely transmitted symbols by calculating the constellation point symbol without quantization processing, and defining the Euclidean distance standard between the constellation point symbol and all constellation point symbols, and then combines The channel state information and the received signal are used to estimate the antenna index for detection. On the premise that the detection performance is approximately unchanged, the complexity of ML detection is reduced, and the redundant calculation is reduced.
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的优选实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings that need to be used in the description of the embodiments. Obviously, the drawings in the following description are only preferred embodiments of the present invention. For those skilled in the art, other drawings can also be obtained based on these drawings without any creative effort.
图1为本发明提供的一种低复杂度的ML检测方法的流程图；Fig. 1 is a flow chart of a low-complexity ML detection method provided by the present invention;
图2为三种检测方法在不同调制下BER性能与SNR之间的关系图。Fig. 2 is a relationship diagram between BER performance and SNR of three detection methods under different modulations.
为了更好理解本发明技术内容，下面提供具体实施例，并结合附图对本发明做进一步的说明。In order to better understand the technical content of the present invention, specific embodiments are provided below, and the present invention is further described in conjunction with the accompanying drawings.
参见图1，本发发明提供了一种低复杂度的ML检测方法，包括下列步骤：Referring to Fig. 1, the present invention provides a kind of low complexity ML detection method, comprises the following steps:
S1、初始化发射天线索引j、发射信号x
l、符号序列号l以及接收信号y与发射信号x
l之间经过信道增益h
j之间的欧式距离d
i；
S1. Initialize the transmitting antenna index j, the transmitting signal x l , the symbol sequence number l and the Euclidean distance d i between the received signal y and the transmitted signal x l through the channel gain h j ;
S2、在已知信号状态信息h
j的条件下，计算夹杂噪声n的星座点符号g
j；
S2. Under the condition of known signal state information hj , calculate the constellation point symbol gj mixed with noise n ;
S3、计算星座点符号g
j与星座样本空间S中所有星座点样本之间的欧氏距离d，并选取其中最小的欧式距离d
min，通过星座解映射器函数，得到估计检测 的星座点符号x
m；
S3. Calculate the Euclidean distance d between the constellation point symbol g j and all constellation point samples in the constellation sample space S, and select the smallest Euclidean distance d min , and obtain the estimated and detected constellation point symbol through the constellation demapper function x m ;
S4、计算接收信号y与所述星座点符号x
m之间进过信道增益h
j之间的欧式距离d
j，比较欧式距离d
j与d
min，根据比较结果，判断发射信号x
l、符号序列号l的估计检测是否正确。
S4. Calculate the Euclidean distance d j between the received signal y and the constellation point symbol x m through the channel gain h j , compare the Euclidean distance d j and d min , and judge the transmitted signal x l and symbol according to the comparison result Whether the estimated detection of sequence number l is correct.
需要说明的是，本实施例中设定空间调制***的发射天线和接收天线分别为N
t、N
r，信道为平坦瑞利衰落信道。在已知所有的信道状态信息条件下，接收端接收信号为：
It should be noted that in this embodiment, the transmitting antenna and the receiving antenna of the spatial modulation system are set to be N t and N r respectively, and the channel is a flat Rayleigh fading channel. Under the condition that all the channel state information is known, the signal received by the receiver is:
其中：ρ为平均接收信噪比，h
j为信道矩阵H的第j列，n为加性高斯白噪声，独立分布且服从均值为0，方差为1的随机变量。
Among them: ρ is the average received signal-to-noise ratio, h j is the jth column of the channel matrix H, n is additive Gaussian white noise, independent distribution and obeys the random variable with mean value 0 and variance 1.
在步骤S2中，当激活的发射天线索引为j时，接收端实际得到的是未经过量化处理的星座点符号g
j，通过下式计算夹杂噪声n的星座点符号g
j：
In step S2, when the index of the activated transmitting antenna is j, what the receiving end actually obtains is the unquantized constellation point symbol g j , and the constellation point symbol g j mixed with noise n is calculated by the following formula:
其中，
为已知信道状态信息h
j的共轭转置，y为接收信号，
为信道增益h
j的F范数的平方值。
in, is the conjugate transpose of the known channel state information h j , y is the received signal, is the square value of the F norm of the channel gain h j .
在步骤S3中，通过下式计算星座点符号g
j与星座样本空间S中所有星座点样本之间的欧氏距离d：d＝||g
j-s||
2，其中s为星座点样本，对欧氏距离d进行排序，找出最小的欧式距离min||g
j-s||
2。通过量化函数即星座解映射器函数Q(·)，得到估计检测的星座点符号x
m，其中x
m的表达式如下：
In step S3, the Euclidean distance d between the constellation point symbol g j and all the constellation point samples in the constellation sample space S is calculated by the following formula: d=||g j -s|| 2 , where s is the constellation point sample , Sort the Euclidean distance d to find the smallest Euclidean distance min||g j -s|| 2 . Through the quantization function, that is, the constellation demapper function Q(·), the estimated and detected constellation point symbol x m is obtained, where the expression of x m is as follows:
在步骤S4中，在完成量化后，通过下式计算接收信号y与所述星座点符号x
m 之间进过信道增益h
j之间的欧式距离d
j＝||y-h
jx
m||
2。
In step S4, after the quantization is completed, calculate the Euclidean distance d j =||yh j x m || 2 between the received signal y and the constellation point symbol x m through the channel gain h j .
可选的，本发明实施例还进一步公开了通过比较d
j及d
min的大小，来判断发射信号x
l、符号序列号l的估计检测是否正确，其具体步骤如下：
Optionally, the embodiment of the present invention further discloses that by comparing the sizes of d j and d min , it is judged whether the estimated detection of the transmitted signal x l and the symbol sequence number l is correct, and the specific steps are as follows:
步骤S401，若d
j<d
min，则说明发射符号x
l、符号序列号l的估计检测正确，分别更新初始的x
l、l、d
i为x
m、j、d
j，此时若j＝N
t，则输出结果x
m、j、d
j，否则继续估计检测天线索引j+1；
Step S401, if d j <d min , it means that the estimated detection of the transmitted symbol x l and the symbol sequence number l is correct, and the initial x l , l, d i are respectively updated as x m , j, d j , if j =N t , then output the results x m , j, d j , otherwise continue to estimate the detection antenna index j+1;
步骤S402，若d
j>d
min，则说明发射符号x
l、符号序列号l的估计检测不合适，此时若j＝N
t，则输出结果即初始值x
l、l、d
i。否则将继续估计下个天线索引j+1；
Step S402, if d j >d min , it means that the estimation and detection of transmitted symbol x l and symbol sequence number l are not suitable. If j=N t at this time, the output results are initial values x l , l, d i . Otherwise, continue to estimate the next antenna index j+1;
式中，N
t为发射天线。
In the formula, N t is the transmitting antenna.
可选的，本实施例的所述方法还包括：若发射符号x
l、符号序列号l的估计检测不合适，重复步骤S2至S4，直到遍历所有的N
t发射天线为止。
Optionally, the method in this embodiment further includes: if the estimation and detection of the transmitted symbol x l and the symbol sequence number l are not suitable, repeating steps S2 to S4 until all N t transmitting antennas are traversed.
本发明进一步比较了所提出的低复杂度的ML检测方法与ML检测方法的计算复杂度，因为在这两种检测法中复数加法的次数类似，因此不考虑加法的次数，仅考虑复数乘法次数。The present invention further compares the computational complexity of the proposed low-complexity ML detection method with the ML detection method, because the number of complex additions is similar in these two detection methods, so the number of additions is not considered, only the number of complex multiplications is considered .
得到ML检测的计算复杂度为C
ML＝2N
tN
r+2N
tM+M，所提出的低复杂度ML检测的复杂度为
通过比较2种检测方法的复杂度可知，所提出的低复杂度的ML检测方法比ML检测方法的计算复杂度要低。这是因为ML检测要求所有选定的发射天线结合所有发射符号来估计发射信息，这可能带来很多不必要的计算。而所提出的低复杂度ML检测利用估计的符号和相应的信道状态信息来帮助解码符号，来估计发射天线索引，减少了冗余计算。
The computational complexity of the obtained ML detection is C ML =2N t N r +2N t M+M, and the proposed low-complexity ML detection has the complexity of By comparing the complexity of the two detection methods, it can be seen that the proposed low-complexity ML detection method has lower computational complexity than the ML detection method. This is because ML detection requires all selected transmit antennas to combine all transmit symbols to estimate transmit information, which may bring a lot of unnecessary calculations. Whereas the proposed low-complexity ML detection utilizes the estimated symbols and corresponding channel state information to help decode the symbols to estimate the transmit antenna index and reduce redundant computation.
本发明的效果可通过以下对比实验说明，具体如下：图2为ML检测、提 出的低复杂度ML检测和次优检测方法在16QAM和256QAM调制下BER性能与SNR之间的关系曲线。在本次仿真中，将发射功率定、发射天线数和接收天线数分别设置为4。从图中可知，当调制为16QAM时所提出的低复杂度ML检测和ML检测具有近似的BER性能，并且均优于次优检测。当BER＝10
-6时，提出的检测比次优检测高2.5dB的SNR增益。随着SNR的提高，这三种检测方法的BER性能也得到了提高。但是随着调制阶数的增加，星座图的规模也在增加，会对信号检测性能产生影响。当调制为256QAM时这三种检测方法的BER性能均会降低，但所提出的低复杂度ML检测和ML检测的性能仍近似，并且均优于次优检测。当BER＝10
-2时，提出的检测方法比次优检测方法高8dB的SNR增益。
The effect of the present invention can be illustrated by the following comparative experiments, specifically as follows: FIG. 2 is a relationship curve between BER performance and SNR under 16QAM and 256QAM modulation for ML detection, proposed low-complexity ML detection and suboptimal detection methods. In this simulation, the transmit power, the number of transmit antennas and the number of receive antennas are set to 4 respectively. It can be seen from the figure that the proposed low-complexity ML detection and ML detection have approximate BER performance when the modulation is 16QAM, and both outperform the sub-optimal detection. When BER = 10 -6 , the proposed detection has a 2.5dB SNR gain over the sub-optimal detection. Along with the improvement of SNR, the BER performance of these three detection methods is also improved. However, as the modulation order increases, the scale of the constellation diagram also increases, which will affect the signal detection performance. The BER performance of all three detection methods decreases when the modulation is 256QAM, but the performance of the proposed low-complexity ML detection and ML detection are still similar, and all outperform the suboptimal detection. When BER= 10-2 , the proposed detection method has 8dB SNR gain over the sub-optimal detection method.
以上所述仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内，所做的任何修改、等同替换、改进等，均应包含在本发明保护的范围之内。The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the present invention. within the scope of protection.
Claims (7)
- 一种低复杂度的ML检测方法，其特征在于，包括下列步骤：A low-complexity ML detection method is characterized in that it comprises the following steps:S1、初始化发射天线索引j、发射信号x l、符号序列号l以及接收信号y与发射信号x l之间经过信道增益h j之间的欧式距离d i； S1. Initialize the transmitting antenna index j, the transmitting signal x l , the symbol sequence number l and the Euclidean distance d i between the received signal y and the transmitted signal x l through the channel gain h j ;S2、在已知信号状态信息h j的条件下，计算夹杂噪声n的星座点符号g j； S2. Under the condition of known signal state information hj , calculate the constellation point symbol gj mixed with noise n ;S3、计算星座点符号g j与星座样本空间S中所有星座点样本之间的欧氏距离d，并选取其中最小的欧式距离d min，通过星座解映射器函数，得到估计检测的星座点符号x m； S3. Calculate the Euclidean distance d between the constellation point symbol g j and all constellation point samples in the constellation sample space S, and select the smallest Euclidean distance d min , and obtain the estimated and detected constellation point symbol through the constellation demapper function x m ;S4、计算接收信号y与所述星座点符号x m之间进过信道增益h j之间的欧式距离d j，比较欧式距离d j与d min，根据比较结果，判断发射信号x l、符号序列号l的估计检测是否正确。 S4. Calculate the Euclidean distance d j between the received signal y and the constellation point symbol x m through the channel gain h j , compare the Euclidean distance d j and d min , and judge the transmitted signal x l and symbol according to the comparison result Whether the estimated detection of sequence number l is correct.
- 根据权利要求1所述的一种低复杂度的ML检测方法，其特征在于，通过下式计算夹杂噪声n的星座点符号g j： A kind of low-complexity ML detection method according to claim 1, it is characterized in that, calculate the constellation point symbol gj of inclusion noise n by following formula:
其中，为已知信道状态信息h j的共轭转置，y为接收信号，
为信道增益h j的F范数的平方值。 in,
is the conjugate transpose of the known channel state information h j , y is the received signal,is the square value of the F norm of the channel gain h j . - 根据权利要求2所述的一种低复杂度的ML检测方法，其特征在于，通过下式计算星座点符号g j与星座样本空间S中所有星座点样本之间的欧氏距离d：d＝||g j-s|| 2，其中s为星座点样本。 A kind of low-complexity ML detection method according to claim 2, it is characterized in that, calculate the Euclidean distance d between the constellation point symbol g j and all constellation point samples in the constellation sample space S by the following formula: d= ||g j -s|| 2 , where s is the constellation point sample.
- 根据权利要求3所述的一种低复杂度的ML检测方法，其特征在于，通过星座解映射器函数Q(·)，得到估计检测的星座点符号Q(g j)＝argmin s∈S||g j-s|| 2。 A low-complexity ML detection method according to claim 3, characterized in that, through the constellation demapper function Q(·), the constellation point symbol Q(g j )=argmin s∈S | |g j -s|| 2 .
- 根据权利要求4所述的一种低复杂度的ML检测方法，其特征在于，通过下式计算接收信号y与所述星座点符号x m之间进过信道增益h j之间的欧式距离d j＝||y-h jx m|| 2。 A kind of low-complexity ML detection method according to claim 4, it is characterized in that, calculate the Euclidean distance d between the received signal y and the constellation point symbol x m through the channel gain h j by the following formula j = ||yh j x m || 2 .
- 根据权利要求5所述的一种低复杂度的ML检测方法，其特征在于，通过下列方式判断发射信号x l、符号序列号l的估计检测是否正确： A kind of low-complexity ML detection method according to claim 5, is characterized in that, judge whether the estimated detection of transmission signal x l , symbol serial number l is correct by the following manner:若d j<d min，则说明发射符号x l、符号序列号l的估计检测正确，分别更新初始的x l、l、d i为x m、j、d j，此时若j＝N t，则输出结果x m、j、d j，否则继续估计检测天线索引j+1； If d j <d min , it means that the estimated detection of the transmitted symbol x l and the symbol sequence number l is correct, and the initial x l , l, and d i are respectively updated as x m , j, and d j , at this time, if j=N t , then output the results x m , j, d j , otherwise continue to estimate the detection antenna index j+1;若d j>d min，则说明发射符号x l、符号序列号l的估计检测不合适，此时若j＝N t，则输出结果即初始值x l、l、d i。否则将继续估计下个天线索引j+1； If d j >d min , it means that the estimation and detection of transmitted symbol x l and symbol sequence number l are not suitable. If j=N t at this time, the output results are the initial values x l , l, d i . Otherwise, continue to estimate the next antenna index j+1;式中，N t为发射天线。 In the formula, N t is the transmitting antenna.
- 根据权利要求6所述的一种低复杂度的ML检测方法，其特征在于，所述方法还包括：若发射符号x l、符号序列号l的估计检测不合适，重复步骤S2至S4，直到遍历所有的N t发射天线为止。 A low-complexity ML detection method according to claim 6, characterized in that the method further comprises: if the estimated detection of the transmitted symbol x l and the symbol sequence number l is not suitable, repeating steps S2 to S4 until Traversing through all N t transmit antennas.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110522651.7A CN113541862A (en) | 2021-05-13 | 2021-05-13 | ML detection method with low complexity |
| CN202110522651.7 | 2021-05-13 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2022237028A1 true WO2022237028A1 (en) | 2022-11-17 |
Family
ID=78095366
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2021/118574 WO2022237028A1 (en) | 2021-05-13 | 2021-09-15 | Low-complexity ml detection method |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113541862A (en) |
| WO (1) | WO2022237028A1 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115426059B (en) * | 2022-08-31 | 2024-04-16 | 西南交通大学 | Low-complexity signal detection method based on power separation receiver |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011080094A1 (en) * | 2009-12-30 | 2011-07-07 | Telecom Paris Tech | Decoding method using an augmented point network for a multi-source system |
| CN103987068A (en) * | 2014-05-30 | 2014-08-13 | 电子科技大学 | Low-complexity spatial modulation system detecting method |
| CN105099974A (en) * | 2015-08-28 | 2015-11-25 | 中国人民解放军理工大学 | Low-complexity spatial modulation soft-decision detection method |
| CN106911374A (en) * | 2017-01-16 | 2017-06-30 | 重庆邮电大学 | A kind of low complexity soft exports spatial modulation detection method |
| CN108989262A (en) * | 2018-08-08 | 2018-12-11 | 中国计量大学 | A kind of low complex degree Incoherent Spatial method for modulation detection based on APSK modulation |
| CN109286587A (en) * | 2018-11-01 | 2019-01-29 | 长安大学 | A kind of how active generalized space method for modulation detection |
| CN109818663A (en) * | 2018-12-21 | 2019-05-28 | 中国计量大学 | A kind of low complex degree difference quadrature spatial modulation detection method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170288933A1 (en) * | 2016-03-30 | 2017-10-05 | Intel IP Corporation | Wireless signal receiver |
| CN107979396A (en) * | 2016-10-25 | 2018-05-01 | 索尼公司 | Communicator and method for Multi-User Dimension modulation |
-
2021
- 2021-05-13 CN CN202110522651.7A patent/CN113541862A/en active Pending
- 2021-09-15 WO PCT/CN2021/118574 patent/WO2022237028A1/en active Application Filing
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011080094A1 (en) * | 2009-12-30 | 2011-07-07 | Telecom Paris Tech | Decoding method using an augmented point network for a multi-source system |
| CN103987068A (en) * | 2014-05-30 | 2014-08-13 | 电子科技大学 | Low-complexity spatial modulation system detecting method |
| CN105099974A (en) * | 2015-08-28 | 2015-11-25 | 中国人民解放军理工大学 | Low-complexity spatial modulation soft-decision detection method |
| CN106911374A (en) * | 2017-01-16 | 2017-06-30 | 重庆邮电大学 | A kind of low complexity soft exports spatial modulation detection method |
| CN108989262A (en) * | 2018-08-08 | 2018-12-11 | 中国计量大学 | A kind of low complex degree Incoherent Spatial method for modulation detection based on APSK modulation |
| CN109286587A (en) * | 2018-11-01 | 2019-01-29 | 长安大学 | A kind of how active generalized space method for modulation detection |
| CN109818663A (en) * | 2018-12-21 | 2019-05-28 | 中国计量大学 | A kind of low complex degree difference quadrature spatial modulation detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113541862A (en) | 2021-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106911374A (en) | A kind of low complexity soft exports spatial modulation detection method | |
| CN109547077B (en) | Signal transmission method, signal reception method, communication device, and storage medium | |
| US9008236B2 (en) | Communication signal decoding and soft demapping | |
| CN108989262A (en) | A kind of low complex degree Incoherent Spatial method for modulation detection based on APSK modulation | |
| WO2022237028A1 (en) | Low-complexity ml detection method | |
| US20060251181A1 (en) | Method and system for signal detection using a reduced transmitter constellation | |
| CN109286587B (en) | Multi-active generalized spatial modulation detection method | |
| US9166740B1 (en) | Low complexity distance metrics for maximum likelihood receivers | |
| Cogen et al. | Hexagonal quadrature amplitude modulation aided spatial modulation | |
| CN105099974B (en) | A kind of spatial modulation soft decision detection method of low complex degree | |
| US20020007257A1 (en) | Reliable symbols as a means of improving the performance of information transmission systems | |
| KR101015727B1 (en) | Maximum likelihood detection apparatus and method in wireless communication system | |
| US20090262871A1 (en) | Receiver and method for detecting signal in multiple antenna system | |
| CN107968758B (en) | System detection method based on orthogonal space modulation of MPSK signal | |
| Dai et al. | Reducing the complexity of quasi-ML detectors for MIMO systems through simplified branch metric and accumulated branch metric based detection | |
| CN109167648B (en) | Candidate constellation point set generation method and MIMO spatial multiplexing detection method | |
| CN103595508B (en) | The method and soft de-mapped method of fixed point process are carried out to receiving symbol | |
| CN109547074B (en) | Lattice reduction assisted ML-SIC signal detection method based on ZF criterion | |
| CN112260729A (en) | Signal detection method and device, electronic equipment and readable storage medium | |
| Gou et al. | Low-complexity sphere decoding for quadrature spatial modulation | |
| CN109639618A (en) | A kind of low complexity soft output demodulation method suitable for high-order quadrature amplitude modulation signal | |
| CN111131107A (en) | Self-adaptive soft demodulation method based on 5G downlink shared channel state | |
| CN112260726B (en) | Signal detection method and device, electronic equipment and readable storage medium | |
| CN108900450A (en) | ESL system, wireless communication system and its receiving end and signal acceptance method | |
| CN113067617B (en) | Independent enhanced generalized orthogonal space modulation method |
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: 21941595 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 21941595 Country of ref document: EP Kind code of ref document: A1 |