CN109167748B - Partial maximum likelihood detection method based on energy sorting - Google Patents

Abstract

The invention discloses a partial maximum likelihood detection method based on energy sorting, which adopts a step demodulation mode, firstly, the number of subcarriers adopts minimum mean square error equalization, then the energy value of each subcarrier is obtained, and sorting is carried out, the number p of selected candidate subcarriers is set, and then the partial subcarrier number, the antenna number and the constellation symbol are detected by adopting the maximum likelihood. The subcarrier blocks can make up the error retransmission defect when a receiving end demodulates, the subcarrier serial number can be demodulated by comparing the power of the balanced symbols on the subcarriers, the number of traversed combinations can be reduced by adding part of the maximum likelihood detection algorithm, the complexity is reduced to a certain extent, and the setting of the P value can achieve good compromise between the error rate and the complexity of the system.

Description

A Partial Maximum Likelihood Detection Method Based on Energy Ranking

technical field

The invention belongs to the technical field of wireless communication, and in particular relates to a partial maximum likelihood detection method based on energy sorting.

Background technique

Traditional information transmission resources such as airspace, frequency domain and time domain can no longer meet the demands of 5G's increasing information transmission rate. Therefore, the joint space and frequency domain transmission information at the same time has become an emerging resource in 5G, but the increase in the number of antennas and the number of sub-carriers also brings challenges to detection, and index modulation occurs accordingly. Index modulation makes full use of the index of the transmission medium, such as transmit antenna, sub-carrier, time slot or linear block code, and modulates the information bits by some mapping rules. Since the power consumption of index bit transmission is very small, index modulation utilizes a feasible compromise between spectral efficiency (SE) and energy efficiency (EE) or diversity gain and multiplexing gain, It has great potential for green communication in the future fifth generation network. The index modulation techniques mainly include spatial index modulation and carrier index modulation. The idea of both is to use the index modulation technique to reduce interference and introduce index bits to make up for the loss of spectral efficiency. The difference is that spatial index modulation is used to select antennas, and subcarrier index modulation is used to select subcarriers. Spatial index modulation is spatial modulation (Spacial Modulation, SM), which is a new type of multiple input multiple output (Multiple Input Multiple Output, MIMO) transmission technology. Technology. SM is a new type of two-dimensional modulation modulation method. It selects an antenna from a group of antennas to activate and transmit data through index bits. At the receiving end, the index bit information is detected by judging the position of the activated antenna. The received symbols are demodulated to obtain modulation bit information. Due to the unique transmission characteristics of spatial modulation, that is, only one antenna is activated, there is no multi-antenna interference, and the detection complexity and RF link cost at the receiving end are also reduced. The introduced index bits can compensate for the fact that only one antenna is activated. The problem of reducing the spectral efficiency caused by the antenna. The idea of spatial modulation is applied to the multi-carrier system to obtain frequency domain index modulation, namely OFDM-IM. Compared with the traditional Orthogonal Frequency Division Multiplexing (OFDM), the OFDM-IM system has more advantages. At the same time, as long as the number of activated subcarriers is appropriate, the spectral efficiency of the OFDM-IM system will greatly exceed that of the OFDM system.

However, the receiving end of the SM-OFDM-IM system has higher requirements for channel independence, synchronization, etc., and the real-time performance is worse, and the symbol information to be detected also changes accordingly. In this system, the receiving end needs to detect three parts Bit information: antenna index bit information, subcarrier index bit information and modulation bit information, with higher complexity. Therefore, designing a detection method with low complexity and good BER performance is an important content of index modulation.

At present, there are few detection methods for the SM-OFDM-IM system, and the existing method can only detect the sequence number and constellation symbol of the activated sub-carrier, but does not detect the activated antenna sequence number. And it cannot be directly applied to the system of the present invention.

SUMMARY OF THE INVENTION

The technical problem to be solved by the present invention is to provide a partial maximum likelihood detection method based on energy sorting, which reduces the search range of ML, and can achieve a balance between complexity and bit error rate performance. compromise.

The present invention adopts the following technical solutions:

A partial maximum likelihood detection method based on energy sorting, which adopts a step-by-step demodulation method. First, the sub-carrier sequence number is equalized by the minimum mean square error, and then the energy value of each sub-carrier is obtained, and then sorted, and the selected candidate is set. The number of sub-carriers is p, and then the maximum likelihood is used to detect part of the sub-carrier numbers, antenna numbers and constellation symbols.

Specifically, it includes the following steps:

S1. Perform MMSE equalization on the received sub-block y _g ;

能量值，并对能量值进行排序，能量最大的子载波最有可能为激活子载波；S2. Calculate the signal obtained in step S1

energy value, and sort the energy value, the sub-carrier with the largest energy is most likely to be the active sub-carrier;

S3. Set P to be the number of candidate subcarriers in each subblock, and P=1, 2, . . . n, where n is the number of subcarriers in each subblock;

S4. Perform maximum likelihood detection on the selected P candidate subcarriers, all antennas and constellation modulation, and use the group with the smallest Euclidean distance as the final decision result;

S5, g=g+1, repeating steps S1-S4 to obtain the detection results of the G sub-blocks.

为：Wherein, in step S1, the equalized signal

for:

G _MMSE = (H ^H H+σ ² I) ^-1 H ^H

Among them, G _MMSE is the weight matrix, σ ² is the noise variance, I is the unit diagonal matrix, H is the channel, X is the transmitted symbol, and W is the noise symbol.

Among them, the g-th receiving sub-block y _g is:

y _g =H _g X _g +W _g

Among them, g=1,2,...G, G is the total number of sub-blocks, N _t is the total number of transmitting antennas, N _r is the total number of receiving antennas, the number of sub-carriers in each sub-block is n=N/G, H is the channel, X is the transmitted symbol, W is the noise symbol, and the dimension is N _t ×n.

的能量值

为：Among them, in step S2, the signal

energy value of

for:

进行排序如下：Among them, for the obtained energy value of each subcarrier

Sort as follows:

Among them, e ₁ , e ₂ ,..., e _N are the index values from small to large after the energy values are sorted.

Wherein, in step S4, first, the energy detection is used to narrow the traversal range of the subcarriers, and a primary selection is performed, and on this basis, partial maximum likelihood detection is performed to narrow the traversal range of the ML.

Among them, the Euclidean distance D is:

为估计的第g个子块的发送符号，H为信道，g＝1,2，…，G，F为范数。in,

is the estimated transmission symbol of the gth sub-block, H is the channel, g=1, 2, . . . , G, and F are the norm.

Specifically, the distributed modulation is as follows: based on one having N _t transmitting antennas and N _r receiving antennas, N sub-carriers are divided into G sub-carrier blocks, the length of the sub-carrier blocks is n=N/G, and k is selected among them. The sub-carriers are activated and data is sent, the sub-carrier configuration is (n, k), and the modulation mode is M-order modulation.

Specifically, for each subcarrier block, the index bit p ₁ for activating an antenna is:

The subcarrier index bits _p2 are:

_The constellation sign bit p3 is:

p ₃ =k log ₂ M

The number of bits p carried by one SM-OFDM-IM block is:

p=p ₁ +p ₂ +p ₃

The transmission rate R is

如下：Frequency domain transmit symbols of the gth subblock

as follows:

表示第g个子块在第i个发射天线第j个子载波上的发送符号，i＝1,2,…,N_t，j＝1,2,…,n；Among them, g=1,...G,

represents the transmitted symbol of the gth subblock on the jth subcarrier of the ith transmit antenna, i=1, 2,...,N _t , j=1, 2,...,n;

为：Assuming that the wireless channel remains unchanged during the transmission of the SM-OFDM-IM symbol, the received signal in the frequency domain is obtained

for:

表示第g个子块第j个子载波上的信道矩阵，其服从分布CN(0,1)，

表示第g个子块中的接收信号和高斯白噪声。in,

represents the channel matrix on the jth subcarrier of the gth subblock, which obeys the distribution CN(0,1),

represents the received signal and Gaussian white noise in the gth sub-block.

Compared with the prior art, the present invention at least has the following beneficial effects:

The invention is a partial maximum likelihood detection method based on energy sorting, which reduces the search range of traditional ML and reduces the complexity of the detection algorithm of the receiving end.

Further, according to the selection of the P value, detection algorithms with different bit error rate performances can be obtained, which can achieve a good compromise between the system bit error rate and the complexity.

Further, in order to minimize the mean square error between the estimated value based on the received data and the target data, the estimated value of the transmitted symbols on each subcarrier is obtained through frequency domain equalization.

Further, by comparing the powers of the equalization symbols on the subcarriers, it is considered that the corresponding subcarriers with high power are active subcarriers. Therefore, the sequence number of the activated subcarrier can be demodulated.

Further, the partial maximum likelihood detection algorithm can demodulate the accurate subcarrier serial number, antenna index serial number and constellation symbol, which reduces the number of combinations that need to be traversed, and reduces the complexity to a certain extent.

Further, in the transmission structure when the subcarriers are not divided into blocks, there is a major defect of possible error transmission in the demodulation of the receiving end, and this defect can be compensated by dividing the subcarriers into blocks.

To sum up, the subcarrier block of the present invention can make up for the error retransmission defect during demodulation at the receiving end, and the subcarrier serial number can be demodulated by comparing the power of the equalization symbol on the subcarrier, and the added part of the maximum likelihood detection algorithm can be used. Reducing the number of traversed combinations reduces the complexity to a certain extent, and the setting of the P value can achieve a good compromise between the system bit error rate and the complexity.

The technical solutions of the present invention will be further described in detail below through the accompanying drawings and embodiments.

Description of drawings

Fig. 1 is the bit error rate curve diagram of different P values in the partial maximum likelihood detection algorithm based on energy sorting in the SM-OFDM-IM system of the present invention;

Fig. 2 is the SM-OFDM-IM system model diagram in the present invention;

Fig. 3 is the bit error rate comparison diagram of the system of the present invention and the initial system under the same spectral efficiency;

4 is a comparison diagram of the bit error rate of the partial maximum likelihood detection algorithm based on energy sorting in the SM-OFDM-IM system of the present invention, the classical algorithm ML, and the partial maximum likelihood detection algorithm using ZF equalization.

Detailed ways

The present invention provides a partial maximum likelihood detection method based on energy sorting, which mainly combines spatial index modulation and frequency domain index modulation to form a space-frequency joint index modulation system, namely SM-OFDM-IM, which combines spatial index modulation with spatial domain index modulation. The antenna index and the subcarrier index in the frequency domain can have the advantages of both.

A partial maximum likelihood detection method based on energy sorting of the present invention adopts the idea of step-by-step demodulation. First, the sub-carrier sequence number is equalized by the minimum mean square error, and then the energy value of each sub-carrier is obtained, and the order is set. Set the value of P as the number of selected candidate sub-carriers, and then use the maximum likelihood to detect part of the sub-carrier numbers, antenna numbers and constellation symbols.

Based on a network with N _t transmit antennas and N _r receive antennas, N subcarriers are divided into G subcarrier blocks, and the length of the subcarrier block is n=N/G, and k subcarriers are selected to activate and transmit data. The carrier configuration is (n, k), and the modulation mode is M-order modulation.

For each subcarrier block, the index bit p1 that activates _one antenna is:

The subcarrier index bits _p2 are:

_The constellation sign bit p3 is:

p ₃ =k log ₂ M

So the number of bits p that a SM-OFDM-IM block can carry is:

p=p ₁ +p ₂ +p ₃

The transmission rate R is

如下bits per channel use, bpcu, according to the above steps, the frequency domain transmission symbol of the gth sub-block

as follows

表示第g个子块在第i个发射天线第j个子载波上的发送符号，i＝1,2,…,N_t，j＝1,2,…,n。Among them, g=1,...G,

Represents the transmitted symbol of the gth subblock on the jth subcarrier of the ith transmit antenna, i=1, 2,...,N _t , j=1, 2,...,n.

为：Assuming that the wireless channel remains unchanged during the transmission of the SM-OFDM-IM symbol, the received signal in the frequency domain is obtained

for:

表示第g个子块第j个子载波上的信道矩阵，其服从分布CN(0,1)，

表示第g个子块中的接收信号和高斯白噪声。in,

represents the channel matrix on the jth subcarrier of the gth subblock, which obeys the distribution CN(0,1),

represents the received signal and Gaussian white noise in the gth sub-block.

The received signal is detected in units of each sub-block, and the g-th received sub-block y _g is assumed to have a dimension of N _t ×n, where

y _g =H _g X _g +W _g

In the formula, g=1, 2, ..., G, G is the total number of sub-blocks, N _t is the total number of transmitting antennas, N _r is the total number of receiving antennas, the number of sub-carriers in each sub-block is n=N/G, and H is channel, X is the transmitted symbol, and W is the noise symbol.

Specific steps are as follows:

S1. Perform MMSE equalization on y _g

MMSE is a linear detection algorithm, which is an improved result based on the zero-forcing detection algorithm. Considering the influence of noise on detection, the weight matrix is designed as follows:

G _MMSE = (H ^H H+σ ² I) ^-1 H ^H

where σ ² is the noise variance, and I is the unit diagonal matrix.

For the MMSE signal detection algorithm, the receiving end needs to know the statistical information of the noise variance σ ² , and the equalized signal obtained by using the obtained frequency domain received signal is:

求能量值，即

得到每个子载波的能量值

并对其进行排序，得到

认为能量最大的子载波最有可能为激活子载波；S2, for the signal after equalization

Find the energy value, that is

Get the energy value of each subcarrier

and sort it to get

It is considered that the sub-carrier with the largest energy is most likely to be the active sub-carrier;

S3. Set the value of P, where P is the number of candidate sub-carriers for each sub-block, P=1, 2,...n;

The complexity of using energy detection is low, and its detection performance is poor due to the interference between multiple antennas and the influence of white Gaussian noise.

S4. Perform maximum likelihood detection on the selected P candidate subcarriers, all antennas and constellation modulation

为根据上述步骤估计的第g个子块的发送符号，欧氏距离最小的组合即为最终判决结果；in,

For the transmission symbol of the gth sub-block estimated according to the above steps, the combination with the smallest Euclidean distance is the final decision result;

S5, g=g+1, repeating the above steps to obtain the detection results of the G sub-blocks.

Firstly, the energy detection is used to narrow the traversal range of subcarriers, and the primary selection is carried out.

In order to make the purposes, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments These are some embodiments of the present invention, but not all embodiments. The components of the embodiments of the invention generally described and illustrated in the drawings herein may be arranged and designed in a variety of different configurations. Thus, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the invention as claimed, but is merely representative of selected embodiments of the invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

The block diagram of the transmitting end and the receiving end of the SM-OFDM-IM system adopted in the present invention is shown in Figure 2. The specific modulation process of the transmitter is as follows: first, the antenna to be activated is selected according to the antenna index OFDM-IM modulation is performed on the carrier block, and the subcarrier to be activated is selected according to the subcarrier index bit and the constellation symbol is modulated. The receiving end is opposite to the transmitting end.

Different from the detection of the traditional MIMO-OFDM system, the receiving end of the SM-OFDM-IM system needs to detect the subcarrier index bits, the antenna index bits and the constellation modulation bits. Index modulation is mainly divided into joint detection and step detection. The joint detection method uses the maximum likelihood criterion to jointly detect the sub-carrier index, antenna index and constellation points to improve the BER performance of the method. The idea of step-by-step detection is to detect the sub-carrier index, antenna index and constellation point separately. However, the detection methods of SM and OFDM-IM systems cannot be used directly. The existing energy detection methods for frequency domain index modulation systems can achieve a bit error rate of 10 ^-4 when the signal-to-noise ratio is 40dB, with good performance. However, when it is directly applied to the SM-OFDM-IM system, the performance is very poor, and the bit error rate is always maintained at about 10 ^-2 . On this basis, a partial ML detection method based on energy sorting is proposed. The bit error rate simulation As shown in Figure 1, the complexity of the method can be greatly reduced under the premise of a slight loss of bit error rate performance.

It can be seen from Figure 1 that as the value of P becomes larger and larger, the search range becomes larger, the complexity of the detection method becomes higher and higher, and the bit error rate performance becomes better and better. When P=4, it is the maximum likelihood ratio. However, the performance of bit error rate is the best, and the complexity is higher.

Table 1 shows the computational complexity of complex multiplication for different detection methods under each sub-block

It can be seen from the above table that the computational complexity of the partial maximum likelihood detection based on the energy ranking detector is lower than that of the traditional maximum likelihood detection method, because the method proposed in the present invention reduces the search range of the maximum likelihood detection by introducing energy detection. Compared with the partial maximum likelihood method using ZF equalization, the complexity of the method proposed in the present invention is slightly higher, which is caused by the difference in selecting equalization matrices.

Please refer to Fig. 2 for MATLAB simulation of SM-OFDM-IM. The number of Monte Carlo simulations is selected as 10 ⁷ , and the noise is Gaussian white noise. In order to ensure that OFDM has the same spectral efficiency as OFDM-IM and SM-OFDM-IM systems, the total number of sub-carriers in the OFDM-IM system is 128 and divided into 64 sub-blocks. The configuration of sub-carriers in each sub-block is (2, 1), using BPSK modulation; adding space modulation on the basis of OFDM-IM to form space-frequency joint index modulation, the parameters of the system are set to a total of 64 sub-carriers, divided into 32 sub-blocks, the antenna system is 2 × 2, The subcarriers of each subblock are configured as (2,1), using QPSK modulation.

Please refer to Figure 3, which is the bit error rate curve diagram of OFDM, OFDM-IM, and SM-OFDM-IM systems. According to the simulation curve, it can be seen that under the same spectral efficiency, after adding the antenna index, the bit error rate At 10 ^-3 , the SM-OFDM-IM system achieves gains of 30dB and 15dB, respectively, compared to OFDM and OFDM-IM. This is because, compared with the traditional OFDM technology, SM-OFDM-IM adds subcarrier index information and antenna index information, and selects some subcarriers and some antennas to transmit data. While other subcarriers and other antennas remain silent, the sparseness of frequency domain data reduces the sensitivity of the system to frequency offset and the impact of inter-subcarrier interference on transmission performance. Interference.

In the case of high signal-to-noise ratio, the bit error rate performance of SM-OFDM-IM system is better than OFDM, which indicates that SM-OFDM-IM has better achievable rate. Compared with the bit rate of the transmitting end, SM-OFDM-IM will reduce the spectral efficiency due to the muting of part of the subcarriers and part of the antenna, but the introduced index bit information can make up for this problem. SM-OFDM-IM is more flexible in parameter configuration due to its unique system settings, and is a more universal multi-carrier system than OFDM.

Please refer to Figure 4. The parameter settings of the SM-OFDM-IM system remain unchanged, and different detection methods are compared. Figure 4 shows the BER performance of different detection methods of SM-OFDM-IM. It can be seen from the figure that when P = 3, the detection method proposed by the present invention achieves better bit error rate performance than the ML method using ZF equalization, because the MMSE considers the influence of noise on detection, but compared with the maximum likelihood detection method, the BER performance is better than that of the maximum likelihood detection method. Poor, the complexity is lower because the increased energy detection narrows the search range for maximum likelihood detection.

The above content is only to illustrate the technical idea of the present invention, and cannot limit the protection scope of the present invention. Any changes made on the basis of the technical solution according to the technical idea proposed by the present invention all fall within the scope of the claims of the present invention. within the scope of protection.

Claims (4)

1. A partial maximum likelihood detection method based on energy sorting is characterized in that a step-by-step demodulation mode is adopted, firstly, the number of subcarriers is balanced by the minimum mean square error, then the energy value of each subcarrier is obtained, sorting is carried out, the number P of selected candidate subcarriers is set, and then the partial subcarrier number, the antenna number and the constellation symbol are detected by the maximum likelihood;

the distribution modulation specifically comprises: based on one having N_tRoot transmitting antenna, N_rDividing N sub-carriers into G sub-carrier blocks according to a receiving antenna, selecting k sub-carriers to activate and send data, wherein the sub-carriers are configured as (N, k), and the modulation mode is M-order modulation; the step-by-step demodulation mode comprises the following steps:

s1, receiving sub-block y_gMMSE equalization is performed, and the equalized signal

Comprises the following steps:

G_MMSE＝(H^HH+σ²I)^-1H^H

wherein G is_MMSEAs a weight matrix, σ²I is the unit diagonal matrix, H is the channel,

for transmitting symbols, W is a noise symbol and dimension N_t×n；

S2, calculating the signal obtained in the step S1

Energy value, and sorting the energy values, the subcarrier with the highest energy is most probably the activation subcarrier, signal

Energy value of

Comprises the following steps:

s3, setting P as the number of candidate subcarriers in each subblock, where P is 1,2, … n, and n is the number of subcarriers in each subblock;

s4, carrying out maximum likelihood detection on the selected P candidate subcarriers, all antennas and constellation modulation, taking the group with the minimum Euclidean distance as a final judgment result, firstly using energy detection to reduce the traversal range of the subcarriers, carrying out initial selection, carrying out partial maximum likelihood detection on the basis, reducing the traversal range of ML, wherein the Euclidean distance D is as follows:

wherein,

h is the estimated transmission symbol of the G-th sub-block, G is 1,2, …, G, and F are norms;

s5, when G is G +1, steps S1 to S4 are repeated to obtain the detection results of G subblocks.

2. The partial maximum likelihood detection method based on energy sorting of claim 1, wherein in step S1, the g-th received sub-block y_gComprises the following steps:

y_g＝H_gX_g+W_g

wherein G is 1,2, … G, and G is the total number of sub-blocks.

3. The partial maximum likelihood detection method based on energy sorting as claimed in claim 1, wherein in step S2, the obtained energy value of each sub-carrier is used

The ordering is as follows:

wherein e is₁,e₂,…,e_NAnd sorting the energy values to obtain index values from small to large.

4. The partial maximum likelihood detection method based on energy sorting according to claim 1, characterized in that, for each subcarrier block, the index bit p of one antenna is activated₁Comprises the following steps:

subcarrier index bit p₂Comprises the following steps:

wherein,

selecting the combination condition of k subcarriers from n subcarriers;

constellation symbol bit p₃Comprises the following steps:

p₃＝klog₂M

the bit number p carried by one SM-OFDM-IM block is:

p＝p₁+p₂+p₃

a transmission rate R of

Frequency domain transmission symbol of g sub-block

The following were used:

wherein G is 1, … G,

denotes the symbol sent by the g-th sub-block on the j-th sub-carrier of the ith transmitting antenna, i is 1,2, …, N_t，j＝1,2,…,n；

If the wireless channel is kept unchanged in the transmission process of the SM-OFDM-IM symbol, the obtained frequency domain receiving signal

Comprises the following steps:

wherein,

representing the channel matrix on the jth subcarrier of the jth subblock, subject to a distribution CN (0,1),

representing the received signal and the Gaussian white in the g-th sub-blockNoise.

Images