A two-step detection method, system and apparatus based on maximum likelihood criterion

Abstract

A two-step detection method, system and device based on maximum likelihood criterion belong to wireless communication technology field, the method includes detecting and sorting all possible symbol constellation combinations through maximum likelihood detection algorithm, and selecting the symbol constellation combination with the minimum distance value as the preliminary decoding result based on the maximum likelihood criterion to complete the first step detection; setting a detection threshold, and finding all other distance values within the detection threshold range from the distance value corresponding to the preliminary decoding result based on the distance value sorting obtained in the first step detection, and storing the corresponding symbol constellation combination in the alternative set; using a different measurement standard from the first step maximum likelihood detection algorithm, and performing the second step detection on the symbol constellation combinations in the alternative set based on the maximum likelihood criterion to find the symbol constellation combination with the minimum measurement value as the target decoding signal. The application aims to solve the "error code platform" problem and effectively improve the accuracy and reliability of decoding.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of wireless communication, and particularly relates to a two-step detection method, system and device based on a maximum likelihood criterion. BACKGROUND

[0002] With the continuous development of wireless communication, limited spectrum resources gradually become a bottleneck to meet the growing demand for communication capacity. Multiple-input multiple-output (MIMO) technology can achieve higher data throughput and communication quality without increasing transmit power or system bandwidth by multiplexing multiple data streams in a spatial manner to multiple users (MU), and is widely considered as a promising wireless communication technology. However, MU-MIMO systems face many challenges in practical applications, among which multi-user interference is an important factor that seriously restricts the performance of wireless transmission, and needs to be handled by precoding technology.

[0003] From a traditional point of view, interference is generally considered to be a harmful factor that limits the performance of wireless communication systems, and most existing linear precoding schemes still aim to suppress or eliminate interference. These traditional precoding schemes often only rely on precoding weight matrices designed based on channel state information, without considering data symbol information. Unlike this, a symbol-level precoding (SLP) scheme based on interference utilization can convert the interference signals that are not conducive to wireless transmission into useful signals by adaptively adjusting the amplitude and phase of the interference signals using channel state information and additional data symbol information, not only fully solving the multi-user interference problem in the downlink transmission of the MU-MIMO system, but also utilizing the power of the interference signals to improve the energy efficiency of the system, thereby providing significant performance gain.

[0004] It is worth noting that in the downlink transmission process of the MU-MIMO system, when multi-level modulation such as high-order quadrature amplitude modulation (QAM) is used, for the symbol-level transmit precoding processing based on interference utilization at the base station, the traditional maximum likelihood detection (MLD) receiver will have a relatively serious "error floor" in the decoding process. This is mainly because in a complex signal space, although the traditional MLD algorithm attempts to directly determine the original transmitted signal in one decoding operation, it often encounters multiple candidate signals similar to it, which makes these signals easy to be confused in detection, thereby causing decoding errors. SUMMARY

[0005] The application aims to provide a two-step detection method, system and device based on maximum likelihood criterion, which effectively improves decoding accuracy and reliability and further improves communication performance of the system by solving the "error floor" problem encountered by symbol-level transmit precoding based on interference utilization in MU-MIMO system downlink transmission under high-order modulation.

[0006] In order to achieve the above-mentioned purpose, the application has the following technical solutions:

[0007] In the first aspect, a two-step detection method based on maximum likelihood criterion is provided, comprising:

[0008] The maximum likelihood detection algorithm is used to detect and sort all possible symbol constellation combinations, and based on the maximum likelihood criterion, the symbol constellation combination with the minimum distance value is selected as the preliminary decoding result to complete the first step detection.

[0009] The detection threshold is set, and based on the distance value sorting obtained in the first step detection, all other distance values within the detection threshold range from the distance value corresponding to the preliminary decoding result are found, and the corresponding symbol constellation combinations are stored in the candidate set.

[0010] A different metric standard from the maximum likelihood detection algorithm of the first step detection is used, and the symbol constellation combinations in the candidate set are detected in the second step based on the maximum likelihood criterion to find the symbol constellation combination with the minimum metric value as the target decoding signal.

[0011] As a preferred scheme, the two-step detection method based on maximum likelihood criterion is applicable to a multi-stream MU-MIMO downlink system, which includes a base station equipped with N T root transmit antennas, and K multi-antenna user terminals, each equipped with root receive antennas, and satisfies The number of data streams received by user k is denoted as satisfies The modulated input signal at the base station is represented as:

[0012]

[0013] wherein, denotes the symbol data stream to be sent to user k;

[0014] When the number of user receive antennas is greater than the number of received data streams, the solution of the precoding matrix cannot be directly completed by using the original channel.

[0015] As a preferred solution, the two-step detection method based on the maximum likelihood criterion further comprises obtaining an equivalent channel matrix of the user by singular value decomposition of the channel matrix of the user, combining the equivalent channel matrices of all users to obtain an equivalent channel of the MU-MIMO system, and performing symbol-level transmit precoding matrix calculation by using the equivalent channel of the MU-MIMO system.

[0016] As a preferred solution, the step of obtaining an equivalent channel matrix of the user by singular value decomposition of the channel matrix of the user, combining the equivalent channel matrices of all users to obtain an equivalent channel of the MU-MIMO system comprises the following steps:

[0017] The channel matrix of the user k is denoted as H k, and the channel matrix of the user k is denoted as H k. The singular value decomposition is performed as follows:

[0018]

[0019] wherein Λ k is a singular value matrix of H k , and are unitary matrices;

[0020] Based on the number of received data streams of the user k , the U k is blocked to obtain two sub-matrices and with dimensions of and respectively. The expression is as follows:

[0021]

[0022] In the above formula, corresponds to a sub-matrix composed of the maximum singular value vectors of H k , and is used to construct the equivalent channel matrix of the user k, and the expression is as follows:

[0023]

[0024] The H eff,k of all users are combined to obtain the equivalent channel H eff of the entire MU-MIMO system, at this time, the original MU-MIMO system is equivalent to a low-dimensional MU-MIMO system with a channel matrix H eff .

[0025] As a preferred solution, the two-step detection method based on the maximum likelihood criterion further comprises constructing, through the equivalent channel of the MU-MIMO system, a symbol-level transmit precoding optimization problem based on interference exploitation of the base station under the constraint of total available transmit power, obtaining a symbol-level transmit precoding matrix by solving the symbol-level transmit precoding optimization problem based on interference exploitation, and obtaining the precoding signal of the base station according to the symbol-level transmit precoding matrix.

[0026] As a preferred solution, the step of constructing, through the equivalent channel of the MU-MIMO system, a symbol-level transmit precoding optimization problem based on interference exploitation of the base station under the constraint of total available transmit power comprises the following steps:

[0027] For 16QAM modulation, the considered constellation point Decomposing along the direction of the detection threshold, we obtain

[0028]

[0029] Decomposing the l-th decoded signal stream of the k-th user along the direction of the detection threshold, we obtain

[0030]

[0031] where and are two newly introduced column vectors in the 16QAM constellation; and are two detection thresholds parallel to the constellation points, respectively; and are two real scaling factors representing the severity of the interference affecting the decoded signal stream; for the k-th user, the above expressions can be expressed in a compact form as follows:

[0032]

[0033] The 16QAM constellation points are decomposed into two sets, where the set represents the constellation points that can exploit interference, and if the noiseless decoded signal stream is located in the constructive region, the following expression is satisfied:

[0034]

[0035] In designing the precoding matrix based on interference exploitation, the base station can exploit only the intra-user data stream interference while setting the inter-user interference to zero, i.e., satisfying the following expression:

[0036] H k W i = 0, k ≠ i

[0037] The symbol-level transmit precoding optimization problem based on interference exploitation is expressed as follows:

[0038]

[0039] C2∶H k W i = 0, k≠i,

[0040]

[0041] where p0represents the total available transmit power of the base station;

[0042] The symbol-level transmit precoding matrix is obtained by solving the symbol-level transmit precoding optimization problem based on interference exploitation, and the precoded signal of the base station is obtained according to the symbol-level transmit precoding matrix, and the method comprises the following steps:

[0043] The symbol-level transmit precoding optimization problem based on interference exploitation is solved by using a convex optimization toolbox to obtain the symbol-level transmit precoding matrix and the output signal of the base station after precoding processing is obtained

[0044] x = Ws

[0045] Based on the above settings, the received signal of the kth user is which is expressed as follows:

[0046] y k = H k x + n k = H k Ws + n k

[0047] where represents the channel matrix between the base station and the kth user, and each item in H k obeys a standard complex Gaussian distribution; represents an additive Gaussian white noise vector with a mean of 0 and a variance of σ 2 .

[0048] As a preferred solution, the detection and sorting of all possible symbol constellation combinations by the maximum likelihood detection algorithm are completed, and the symbol constellation combination with the minimum distance value is selected as the preliminary decoding result based on the maximum likelihood criterion, and the first step detection comprises the following steps:

[0049] The distance value of all possible symbol constellation combinations is sorted by the maximum likelihood detection algorithm, and the signal r k,1As a preliminary decoding result, the expression is calculated as follows:

[0050]

[0051] As a preferred solution, the second step detection of the symbol constellation combination in the candidate set based on the maximum likelihood criterion and the minimum metric value of the symbol constellation combination as the target decoding signal includes the following steps:

[0052] The second step detection of the symbol constellation combination in the candidate set based on the maximum likelihood criterion and the minimum metric value of the symbol constellation combination as the target decoding signal includes the following steps: k,2 The expression is calculated as follows:

[0053]

[0054] Wherein, G k = H k W k represents the effective user channel; γ represents a regularization factor, and the introduction of the regularization factor is to avoid the unsolvable problem caused by the rank-one precoding matrix.

[0055] In the second aspect, a two-step detection system based on the maximum likelihood criterion is provided, including:

[0056] The preliminary decoding module is configured to detect and sort all possible symbol constellation combinations by the maximum likelihood detection algorithm, and select the symbol constellation combination with the minimum distance value as the preliminary decoding result based on the maximum likelihood criterion, to complete the first step detection.

[0057] The candidate set construction module is configured to set a detection threshold, find all other distance values within the detection threshold range based on the distance value sorting obtained in the first step detection, and store the corresponding symbol constellation combinations in the candidate set.

[0058] The target decoding signal detection module is configured to use a different metric standard from the maximum likelihood detection algorithm in the first step detection, and perform the second step detection of the symbol constellation combination in the candidate set based on the maximum likelihood criterion, to find the symbol constellation combination with the minimum metric value as the target decoding signal.

[0059] In the third aspect, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the two-step detection method based on the maximum likelihood criterion.

[0060] Compared with the prior art, the present application has at least the following beneficial effects:

[0061] The two-step detection method based on the maximum likelihood criterion of the present application adds a re-detection step on the basis of the traditional maximum likelihood detection algorithm, and the purpose is to improve the accuracy and reliability of decoding through secondary screening. The method of the present application firstly performs preliminary detection and distance value sorting on all possible symbol constellation combinations based on the maximum likelihood criterion through the traditional maximum likelihood detection algorithm, selects the symbol constellation combination with the minimum distance value as the preliminary decoding result, and the detection of the first step aims to maximize the similarity between the decoded signal and the transmitted signal, so as to facilitate the subsequent second step detection. Then, a reasonable detection threshold is set to further screen out the candidate signals similar to the preliminary decoding result, and put them into the alternative set. Finally, a different metric from the traditional maximum likelihood detection algorithm is used to perform secondary detection on the symbol constellation combinations in the alternative set, and the symbol constellation combination corresponding to the minimum metric value in the secondary detection is determined as the final decoding signal based on the maximum likelihood criterion, and the detection of the second step aims to maximize the decoding accuracy in the complex signal space under high-order modulation. The method of the present application effectively solves the "error floor" problem faced by the MU-MIMO downlink system under high-order modulation when using symbol-level transmit precoding based on interference utilization, and has significant performance gain. BRIEF DESCRIPTION OF DRAWINGS

[0062] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show part of the embodiments of the present application, and other related drawings can also be obtained by those skilled in the art without creative labor.

[0063] Figure 1 The system block diagram of the MU-MIMO of the embodiment of the present application.

[0064] Figure 2 The principle diagram of the interference utilization precoding based on "symbol scaling" under 16QAM modulation of the embodiment of the present application.

[0065] Figure 3 The method of the embodiment of the present application adopts 16QAM modulation mode, and when N T = 16, , the bit error rate performance diagram under different regularization factors γ values;

[0066] Fig. 4(a) is the bit error rate performance diagram under different regularization factors γ values when the method of the embodiment of the present application adopts 16QAM modulation mode, and N T = 16, Fig. 4(a) is a schematic diagram of bit error rate varying with signal-to-noise ratio under different threshold values δ when the method of the embodiment of the application adopts 16QAM modulation mode, N = 16, K = 1;

[0067] Fig. 4(b) is a schematic diagram of bit error rate varying with signal-to-noise ratio when the threshold value is δ = 0.01 when the method of the embodiment of the application adopts 16QAM modulation mode, N = 16, K = 1; T

[0068] Fig. 5(a) is a schematic diagram of bit error rate varying with signal-to-noise ratio under different threshold values δ when the method of the embodiment of the application adopts 16QAM modulation mode, N = 16, K = 2; T

[0069] Fig. 5(b) is a schematic diagram of bit error rate varying with signal-to-noise ratio when the threshold value is δ = 0.01 when the method of the embodiment of the application adopts 16QAM modulation mode, N = 16, K = 2; T

[0070] Fig. 6(a) is a schematic diagram of bit error rate varying with signal-to-noise ratio under different threshold values δ when the method of the embodiment of the application adopts 16QAM modulation mode, N = 16, K = 4; T

[0071] Fig. 6(b) is a schematic diagram of bit error rate varying with signal-to-noise ratio when the threshold value is δ = 0.01 when the method of the embodiment of the application adopts 16QAM modulation mode, N = 16, K = 4; T

[0072] Fig. 7(a) is a schematic diagram of bit error rate varying with signal-to-noise ratio under different threshold values δ when the method of the embodiment of the application adopts 16QAM modulation mode, N = 16; T

[0073] Fig. 7(b) is a schematic diagram of bit error rate varying with signal-to-noise ratio when the threshold value is δ = 0.01 when the method of the embodiment of the application adopts 16QAM modulation mode, N = 16. T DETAILED DESCRIPTION

[0074] ​​​​​​​​​​​​​​With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described, obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, other embodiments can also be obtained by those skilled in the art without creative labor.

[0075] Please refer to Figure 1 The two-step detection method based on the maximum likelihood criterion in the embodiments of the present application is suitable for a multi-stream MU-MIMO downlink system, which includes a base station equipped with N T root transmit antennas, and K multi-antenna user terminals, wherein each user is equipped with root receive antennas, and satisfies The number of data streams received by the user k is denoted as which satisfies Then the modulated input signal at the base station can be expressed as:

[0076]

[0077] wherein, denotes the symbol data stream to be sent to the user k.

[0078] In the MU-MIMO system downlink transmission process, since the receiving end user is equipped with multiple receive antennas, it is possible that the number of user receive antennas is not equal to the number of received data streams. When the number of user receive antennas is greater than the number of received data streams, the original channel cannot be directly used to solve the precoding matrix.

[0079] The two-step detection method based on the maximum likelihood criterion in the embodiments of the present application mainly includes the following steps:

[0080] All possible symbol constellation combinations are detected and sorted by the maximum likelihood detection algorithm, and based on the maximum likelihood criterion, the symbol constellation combination with the minimum distance value is selected as the preliminary decoding result to complete the first step detection;

[0081] The detection threshold is set, and based on the distance value sorting obtained in the first step detection, all other distance values within the detection threshold range from the distance value corresponding to the preliminary decoding result are found, and the corresponding symbol constellation combination is stored in the candidate set;

[0082] A different metric standard is used than the maximum likelihood detection algorithm in the first step detection, and the symbol constellation combinations in the candidate set are detected in the second step based on the maximum likelihood criterion, and the symbol constellation combination with the minimum metric value is found as the target decoding signal.

[0083] In one possible implementation, the two-step detection method based on the maximum likelihood criterion of this invention further includes applying singular value decomposition to the user's channel matrix to obtain the equivalent channel matrix, combining the equivalent channel matrices of all users to obtain the equivalent channel of the MU-MIMO system, and using the equivalent channel of the MU-MIMO system to calculate the symbol-level transmit precoding matrix.

[0084] Specifically, this embodiment of the invention utilizes singular value decomposition to perform an equivalent transformation of the original channel, thereby obtaining an equivalent low-dimensional MU-MIMO channel. The specific steps are as follows:

[0085] First, consider the channel matrix for user k. Using singular value decomposition, that is:

[0086]

[0087] Among them, Λ k It is H k The singular value matrix, and They are all unitary matrices.

[0088] Next, based on the number of data streams received by the user For U k Divide into blocks, and obtain the dimensions as follows: and Two submatrices and Right now:

[0089]

[0090] In fact, in the above formula Corresponding to H k The submatrix formed by the maximum singular value vectors is the channel vector most likely to be used to carry the data symbols that user k is interested in. Therefore, it is possible to utilize... The equivalent channel matrix of user k The constructed form is represented as:

[0091]

[0092] H of all users eff,k Combining these elements yields the equivalent channel H of the entire system. eff At this point, the original MU-MIMO system is equivalent to a channel matrix H. eff The low-dimensional MU-MIMO system can then directly utilize the equivalent channel H eff Perform symbol-level emission precoding matrix calculation.

[0093] In a possible implementation, the two-step detection method based on the maximum likelihood criterion in the embodiment of the present application further comprises constructing, by means of the equivalent channel of the MU-MIMO system, a symbol-level transmission precoding optimization problem based on interference utilization of the base station under the constraint of total available transmission power, obtaining a symbol-level transmission precoding matrix by solving the symbol-level transmission precoding optimization problem based on interference utilization, and obtaining the precoding signal of the base station according to the symbol-level transmission precoding matrix.

[0094] Specifically, taking 16QAM modulation as an example, refer to Figure 2 , the considered constellation point is decomposed along the direction of the detection threshold to obtain

[0095]

[0096] Similarly, the kth decoded signal stream of the lth user is decomposed along the direction of the detection threshold to obtain

[0097]

[0098] wherein, and are two column vectors newly introduced in the 16QAM constellation; and are two detection thresholds parallel to the respective constellation points; and are two real scaling factors, representing the severity of the influence of the interference on the decoded signal stream. For the convenience of representation, for the kth user, the above formula can be expressed in a compact form, i.e.

[0099]

[0100] Specifically, the 16QAM constellation points are decomposed into two sets, wherein the set represents the constellation points that can perform interference utilization, i.e. Figure 2 the real part of the "B" point, the imaginary part of the "C" point, and the real part and the imaginary part of the "D" point, etc.; the set J represents the constellation points that cannot perform interference utilization, i.e. Figure 2 the real part of the "C" point, the imaginary part of the "B" point, and the real part and the imaginary part of the "A" point, etc. Thus, under the 16QAM modulation, if the noiseless decoded signal stream is located in the constructive region, i.e. the green region in Figure 2 , it is satisfied that

[0101]

[0102] Further, since the MLD receivers operate independently at each user in the MU-MIMO downlink system, they only rely on their own channel matrix and precoding matrix when decoding, and cannot obtain information of other users.

[0103] Therefore, when designing the precoding matrix based on interference exploitation, the only available interference for the base station is the inter-stream interference within the user, while the inter-user interference needs to be zeroed, i.e., satisfying:

[0104] H k W i = 0, k≠i

[0105] According to the above derivation, when the base station pre-processes the transmitted signal through the symbol-level precoding (SLP) technique based on interference exploitation, the optimization problem can be expressed as:

[0106]

[0107] C2: H k W i = 0, k≠i

[0108]

[0109] where p0 represents the total available transmit power of the base station.

[0110] Obviously, is a typical convex optimization problem, which can be directly solved by a convex optimization toolbox such as CVX.

[0111] Further, after obtaining the symbol-level precoding (SLP) matrix through the above optimization problem, the output signal of the base station after precoding processing can be obtained as i.e.,

[0112] x = Ws

[0113] Based on the above settings, the received signal of the kth user can be expressed as:

[0114] y k = H k x + n k = H k Ws + n k

[0115] where represents the channel matrix between the base station and the kth user, and each item in H K obeys a standard complex Gaussian distribution; represents an additive Gaussian white noise vector with a mean of 0 and a variance of σ 2 .

[0116] In one possible implementation, for the k-th user, the received signal y is processed at the receiver using a two-step maximum likelihood detection (MLD) algorithm. k Decoding is performed to obtain the final demodulated signal.

[0117] Specifically, the first step of the detection process is as follows:

[0118] All were analyzed using the traditional maximum likelihood detection MLD algorithm. The possible symbol constellation combinations are sorted by distance values, and the signal r corresponding to the minimum distance value is selected based on the maximum likelihood criterion. k,1 As a preliminary decoding result, namely:

[0119]

[0120] Specifically, the second step of the detection process is as follows:

[0121] Set a suitable detection threshold δ. For the distance values ​​obtained in the first step that have been sorted in ascending order, find all other distance values ​​whose minimum difference from the distance value is within the range of δ, and store them in a candidate set. The symbol constellation combination corresponding to this set is the candidate signal s that is similar to the preliminary decoding result and is easily confused. i (Where i≠k). Based on this, a different metric than the traditional Maximum Likelihood Detection (MLD) algorithm is adopted, and a secondary detection is performed on the symbol constellation combinations in the candidate set based on this metric. Also based on the maximum likelihood criterion, the calculated metric values ​​are sorted in ascending order; the symbol constellation combination ranked first is the final decoded signal r. k,2 ,Right now:

[0122]

[0123] Among them, G k =H k W k γ represents the effective user channel; γ represents the regularization factor, which is introduced to avoid the unsolvable problem caused by the rank-one precoding matrix.

[0124] The following simulation experiments further illustrate the advantages of the two-step detection method based on the maximum likelihood criterion of this invention:

[0125] The proposed algorithm was simulated using Monte Carlo simulation.

[0126] The test conditions are as follows:

[0127] In the downlink of a multi-stream MU-MIMO system, the base station is equipped with N TThe root transmit antenna, and the receiving end is K multi-antenna users, wherein the number of receiving antennas of user k is recorded as The number of receiving data streams is recorded as Here, the transmit power budget of each time slot is set to p0=1W.

[0128] The simulation results are shown in Figure 3 , Figure 4(a), Figure 4(b), Figure 5(a), Figure 5(b), Figure 6(a), Figure 6(b), Figure 7(a) and Figure 7(b), wherein "SLP+Traditional-MLD" means that the transmit signal is preprocessed at the transmitting end using symbol-level precoding, and the received signal is decoded at the receiving end using the traditional MLD algorithm; "SLP+TwoSteps-MLD" means that the transmit signal is preprocessed at the transmitting end using symbol-level precoding, but the decoding operation at the receiving end is performed by the two-step MLD algorithm; "BD+Traditional-MLD" means that the transmit signal is preprocessed at the transmitting end using traditional block diagonalization precoding, and the received signal is decoded at the receiving end using the traditional MLD algorithm; "BD+TwoSteps-MLD" means that the transmit signal is preprocessed at the transmitting end using traditional block diagonalization precoding, but the decoding operation at the receiving end is performed by the two-step MLD algorithm.

[0129] Please refer to Figure 3 , Figure 3 The bit error rate (BER) performance of the proposed "SLP+TwoSteps-MLD" scheme under different regularization factors γ is shown. Here, the simulation parameters are set to N T =16, and the threshold is set to δ=0.01. It is observed that the BER curves under different γ values are nearly overlapped, that is, the value of the regularization factor γ is not the main factor affecting the bit error rate BER performance, so the regularization factor of all subsequent simulations is selected as the regularization factor γ=1.

[0130] Referring to FIG. 4(a), FIG. 4(b) and FIG. 5(a), FIG. 5(b), specifically, FIG. 4(a), FIG. 4(b) and FIG. 5(a), FIG. 5(b) show the BER performance under different simulation parameter configurations when the number of users K = 2. It can be observed from FIG. 4(a) that when the threshold is set as δ = 0.01, the BER performance under all signal-to-noise ratios is better, thus, in the parameter configuration of FIG. 4(a), the threshold is fixed as δ = 0.01, and thus the simulation result of FIG. 4(b) is obtained. Similarly, the simulation result of FIG. 5(b) is also obtained when the threshold is fixed as δ = 0.01. It can be observed that for the symbol-level precoding SLP algorithm, the two-step MLD algorithm at the receiving end can effectively improve the "error floor" problem of the traditional MLD receiver under high signal-to-noise ratio. In particular, in the two simulation scenarios, the proposed "SLP + TwoSteps-MLD" scheme has obvious performance gain compared with other schemes. In addition, the difference between FIG. 4(a), FIG. 4(b) and FIG. 5(a), FIG. 5(b) lies in the number of receiving antennas of the user. In FIG. 4(a), FIG. 4(b), the number of receiving antennas of the user is equal to the number of received data streams, while in FIG. 5(a), FIG. 5(b), the number of receiving antennas of the user is greater than the number of received data streams. It can be known by comparison that when the number of antennas equipped by the user is greater than the number of data streams, the system can utilize the extra antenna degrees of freedom to improve the communication performance. Therefore, under the same signal-to-noise ratio, FIG. 5(a), FIG. 5(b) realizes lower bit error rate BER than FIG. 4(a), FIG. 4(b).

[0131] Referring to FIG. 6(a), FIG. 6(b) and FIG. 7(a), FIG. 7(b), specifically, FIG. 6(a), FIG. 6(b) and FIG. 7(a), FIG. 7(b) show the BER performance under different simulation parameter configurations when the number of users K = 4. Similarly, according to the performance comparison under different thresholds δ in FIG. 6(a) and FIG. 7(a), the threshold is fixed as δ = 0.01 in the two simulation scenarios. Also, it can be observed from FIG. 6(b) and FIG. 7(b) that the "SLP + TwoSteps-MLD" scheme has the optimal bit error rate BER performance. In particular, for the scenario in which the base station adopts the symbol-level precoding SLP algorithm, the proposed two-step maximum likelihood detection MLD algorithm can overcome the "error floor" problem of the traditional maximum likelihood detection MLD receiver to the greatest extent, especially under high signal-to-noise ratio, the improvement effect is more obvious. In addition, it can be observed that since the number of data streams in FIG. 7(a), FIG. 7(b) is less than the number of receiving antennas, thus under the same signal-to-noise ratio, FIG. 7(a), FIG. 7(b) has lower bit error rate BER than FIG. 6(a), FIG. 6(b).

[0132] Another embodiment of the present application also provides a two-step detection system based on the maximum likelihood criterion, comprising:

[0133] The preliminary decoding module is configured to detect and sort all possible symbol constellation combinations by a maximum likelihood detection algorithm, and select a symbol constellation combination with the minimum distance value as a preliminary decoding result based on a maximum likelihood criterion, thereby completing the first-step detection.

[0134] The alternative set construction module is configured to set a detection threshold, find all other distance values within the detection threshold range from the distance value corresponding to the preliminary decoding result based on the distance value sorting obtained in the first-step detection, and store the corresponding symbol constellation combinations in an alternative set.

[0135] The target decoding signal detection module is configured to perform a second-step detection on the symbol constellation combinations in the alternative set based on a maximum likelihood criterion by using a different metric from the maximum likelihood detection algorithm in the first-step detection, and find a symbol constellation combination with the minimum metric value as a target decoding signal.

[0136] Another embodiment of the present application further provides an electronic device, which comprises a memory configured to store at least one instruction, and a processor configured to execute the instruction stored in the memory to implement the two-step detection method based on the maximum likelihood criterion.

[0137] Another embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the two-step detection method based on the maximum likelihood criterion.

[0138] For example, the instruction stored in the memory can be divided into one or more modules / units, which are stored in a computer readable storage medium and executed by the processor to complete the two-step detection method based on the maximum likelihood criterion. The one or more modules / units can be a series of computer readable instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program in the server.

[0139] The electronic device can be a smart phone, a notebook, a palm computer, a cloud server and other computing devices. The electronic device can include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the electronic device can further include more or less components, or combine certain components, or different components, for example, the electronic device can further include an input / output device, a network access device, a bus, etc.

[0140] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0141] The memory can be an internal storage unit of the server, such as a hard disk or a memory of the server. The memory can also be an external storage device of the server, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory can also include both the internal storage unit and the external storage device of the server. The memory is used to store the computer readable instructions and other programs and data required by the server. The memory can also be used to temporarily store data that has been output or will be output.

[0142] It should be noted that the information interaction and execution process between the above module units are based on the same concept as the method embodiments, and the specific functions and technical effects brought by them can be referred to the method embodiment part, which will not be repeated here.

[0143] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional units and modules is taken as an example for illustration, and in actual application, the above functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit or module in the embodiment can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of software functional unit. In addition, the specific names of each functional unit or module are only for easy distinction, and do not limit the protection scope of the present application. The specific working process of the unit or module in the system can refer to the corresponding process in the foregoing method embodiment, which will not be repeated here.

[0144] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the above-mentioned embodiment methods can be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned various method embodiments can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms. The computer readable medium at least includes any entity or device capable of carrying the computer program code to the photographing device / terminal equipment, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium. For example, U disk, mobile hard disk, magnetic disk or optical disk, etc.

[0145] In the above embodiments, the description of each embodiment has its own focus, and the parts not described or recorded in a certain embodiment can be referred to the relevant description of other embodiments.

[0146] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A two-step detection method based on maximum likelihood criterion, characterized in that, Comprise: Through the maximum likelihood detection algorithm, all possible symbol constellation combinations are detected and sorted, and based on the maximum likelihood criterion, the symbol constellation combination with the minimum distance value is selected as the preliminary decoding result, completing the first step detection; Set the detection threshold, and based on the distance value sorting obtained in the first step detection, find all other distance values within the detection threshold range from the distance value corresponding to the preliminary decoding result, and store the corresponding symbol constellation combinations in the candidate set; A different metric standard is used than the maximum likelihood detection algorithm in the first step detection, and based on the maximum likelihood criterion, the symbol constellation combinations in the candidate set are detected in the second step, and the symbol constellation combination with the minimum metric value is found as the target decoding signal; Also includes constructing a symbol-level transmission precoding optimization problem based on interference utilization of the base station under the constraint of the total available transmission power through the equivalent channel of the MU-MIMO system, obtaining the symbol-level transmission precoding matrix by solving the symbol-level transmission precoding optimization problem based on interference utilization, and obtaining the precoding signal of the base station according to the symbol-level transmission precoding matrix; The step of constructing a symbol-level transmission precoding optimization problem based on interference utilization of the base station under the constraint of the total available transmission power through the equivalent channel of the MU-MIMO system comprises the following steps: For 16QAM modulation, for the considered constellation points Decomposing along the direction of the detection threshold, we get: For the first The first user's The decoded signal stream is decomposed along the direction of the detection threshold to obtain: In the formula, as well as These are two newly introduced column vectors in the 16QAM constellation; and These are two detection thresholds, each parallel to the corresponding constellation point. and These are two real-valued scaling factors, representing the severity of interference affecting the decoded signal stream; for the... For each user, the above expression can be represented in a compact form as follows: The 16QAM constellation points are divided into two sets, where set represents the constellation points where interference exploitation is possible, provided that the noiseless decoded signal stream is located in the constructive region, satisfying the following expression: When designing the precoding matrix based on interference utilization, the only interference that the base station can utilize is the intra-user data stream interference, and the inter-user interference is set to zero, i.e. satisfies the following expression: The symbol-level transmission precoding optimization problem based on interference utilization is expressed as the following expression: In the formula, denotes the total available transmit power of the base station; The step of obtaining the symbol-level transmission precoding matrix by solving the symbol-level transmission precoding optimization problem based on interference utilization, and obtaining the precoding signal of the base station according to the symbol-level transmission precoding matrix comprises the following steps: Solve the symbol-level transmit precoding optimization problem based on interference exploitation by using a convex optimization toolbox to obtain a symbol-level transmit precoding matrix , and obtain an output signal pre-processed at the base station : Based on the above settings, the first user's received signal is represented by the following equation: wherein denotes a channel matrix between the base station and the user and each entry in denotes an additive white Gaussian noise vector with mean and variance at the receiving end. The step of completing the first step detection by detecting and sorting all possible symbol constellation combinations through the maximum likelihood detection algorithm, and selecting the symbol constellation combination with the minimum distance value as the preliminary decoding result based on the maximum likelihood criterion comprises the following steps: The distance values are sorted for all possible symbol constellation combinations by a maximum likelihood detection algorithm, and the signal corresponding to the minimum distance value is selected based on a maximum likelihood criterion The expression is calculated as a preliminary decoding result as follows:​ The step of finding the symbol constellation combination with the minimum metric value as the target decoding signal by detecting the symbol constellation combinations in the candidate set based on the maximum likelihood criterion using a different metric standard than the maximum likelihood detection algorithm in the first step detection comprises the following steps: A different metric is used in the second step than in the first step of the maximum likelihood detection algorithm, and the symbol constellation combinations in the candidate set are detected again based on this metric. The metric values calculated are sorted in ascending order based on the maximum likelihood criterion, and the symbol constellation combination ranked first is the target decoded signal The expression is calculated as follows: wherein represents an effective user channel; represents a regularization factor, which is introduced to avoid the non-soluble problem caused by the rank-one precoding matrix.

2. The two-step detection method based on maximum likelihood criterion according to claim 1, characterized in that, A downlink system suitable for multi-stream MU-MIMO comprising a base station equipped with Nt roots transmit antennas, and Nt users equipped with Nt receive antennas, and satisfying ; the number of data streams received by a user is denoted by , satisfying , the modulated input signal at the base station is represented by wherein, represents a stream of symbol data to be transmitted to a user ; When the number of user receiving antennas is greater than the number of receiving data streams, the original channel cannot be directly used to solve the precoding matrix.

3. The two-step detection method based on maximum likelihood criterion according to claim 2, characterized in that, Also includes using singular value decomposition on the channel matrix of the user to obtain the equivalent channel matrix, combining all equivalent channel matrices of the users to obtain the equivalent channel of the MU-MIMO system, and using the equivalent channel of the MU-MIMO system to calculate the symbol-level transmission precoding matrix.

4. The two-step detection method based on maximum likelihood criterion according to claim 3, characterized in that, The step of using singular value decomposition on the channel matrix of the user to obtain the equivalent channel matrix, combining all equivalent channel matrices of the users to obtain the equivalent channel of the MU-MIMO system comprises the following steps: The user is prompted by the following formula The channel matrix of the user Singular value decomposition is used: wherein is a singular value matrix of and are unitary matrices; Based on the user receiving data stream number to block, get two sub-matrix and dimension respectively and , expression as follows: The maximum singular value vector of the matrix corresponding to is denoted as The equivalent channel matrix of the user is constructed as follows: ​ Combining all users to get the equivalent channel of the whole MU-MIMO system At this time, the original MU-MIMO system is equivalent to a low-dimensional MU-MIMO system with a channel matrix ​ 5. A two-step detection system based on maximum likelihood criterion, characterized in that, Comprise: The preliminary decoding module is configured to detect and sort all possible symbol constellation combinations by a maximum likelihood detection algorithm, and select a symbol constellation combination with the minimum distance value as a preliminary decoding result based on a maximum likelihood criterion, thereby completing first-step detection; The candidate set construction module is configured to set a detection threshold, find all other distance values within the detection threshold range from the distance value corresponding to the preliminary decoding result based on the distance value sorting obtained in the first-step detection, and store the corresponding symbol constellation combinations in a candidate set; The target decoding signal detection module is configured to perform second-step detection on the symbol constellation combinations in the candidate set based on a maximum likelihood criterion by using a different metric from the maximum likelihood detection algorithm in the first-step detection, and find a symbol constellation combination with the minimum metric value as a target decoding signal. The equivalent channel of the MU-MIMO system is used to construct a symbol-level transmission precoding optimization problem based on interference utilization of the base station under the constraint of total available transmission power, and a symbol-level transmission precoding matrix is obtained by solving the symbol-level transmission precoding optimization problem based on interference utilization, and a precoding signal of the base station is obtained according to the symbol-level transmission precoding matrix. The equivalent channel of the MU-MIMO system is used to construct a symbol-level transmission precoding optimization problem based on interference utilization of the base station under the constraint of total available transmission power, and a symbol-level transmission precoding matrix is obtained by solving the symbol-level transmission precoding optimization problem based on interference utilization, and a precoding signal of the base station is obtained according to the symbol-level transmission precoding matrix. For 16QAM modulation, for the considered constellation points Decomposing along the direction of the detection threshold, we get: For the first The first user's The decoded signal stream is decomposed along the direction of the detection threshold to obtain: In the formula, as well as These are two newly introduced column vectors in the 16QAM constellation; and These are two detection thresholds, each parallel to the corresponding constellation point. and These are two real-valued scaling factors, representing the severity of interference affecting the decoded signal stream; for the... For each user, the above expression can be represented in a compact form as follows: The 16QAM constellation points are divided into two sets, where set represents the constellation points for which interference exploitation is possible, provided that the noiseless decoded signal stream is located in the constructive region, satisfying the following expression: When designing the precoding matrix based on interference utilization, the interference that can be utilized by the base station is only the data stream interference between users, and the inter-user interference is set to zero, that is, the following expression is satisfied: The symbol-level transmission precoding optimization problem based on interference utilization is expressed as the following expression: In the formula, denotes the total available transmit power of the base station; The equivalent channel of the MU-MIMO system is used to construct a symbol-level transmission precoding optimization problem based on interference utilization of the base station under the constraint of total available transmission power, and a symbol-level transmission precoding matrix is obtained by solving the symbol-level transmission precoding optimization problem based on interference utilization, and a precoding signal of the base station is obtained according to the symbol-level transmission precoding matrix. Solve the symbol-level transmit precoding optimization problem based on interference exploitation by using a convex optimization toolbox to obtain a symbol-level transmit precoding matrix , and obtain an output signal pre-processed at the base station : Based on the above settings, the first user's received signal is expressed by the following equation: wherein denotes the channel matrix between the base station and the user and each entry in is subject to a standard complex Gaussian distribution; denotes an additive white Gaussian noise vector at the receiving end with mean and variance . The equivalent channel of the MU-MIMO system is used to construct a symbol-level transmission precoding optimization problem based on interference utilization of the base station under the constraint of total available transmission power, and a symbol-level transmission precoding matrix is obtained by solving the symbol-level transmission precoding optimization problem based on interference utilization, and a precoding signal of the base station is obtained according to the symbol-level transmission precoding matrix. The distance values are sorted for all possible symbol constellation combinations by a maximum likelihood detection algorithm and the signal corresponding to the smallest distance value is selected based on a maximum likelihood criterion The expression is calculated as a preliminary decoding result as follows:​ The equivalent channel of the MU-MIMO system is used to construct a symbol-level transmission precoding optimization problem based on interference utilization of the base station under the constraint of total available transmission power, and a symbol-level transmission precoding matrix is obtained by solving the symbol-level transmission precoding optimization problem based on interference utilization, and a precoding signal of the base station is obtained according to the symbol-level transmission precoding matrix. A different metric is used in the second step than in the first step of the maximum likelihood detection algorithm, and the symbol constellation combinations in the candidate set are detected again based on this metric. The metric values calculated are sorted in ascending order based on the maximum likelihood criterion, and the symbol constellation combination ranked first is the target decoded signal The expression is calculated as follows: wherein represents an effective user channel; represents a regularization factor, which is introduced to avoid the non-soluble problem caused by the rank-one precoding matrix.

6. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 5. The computer program is executed by the processor to implement the two-step detection method based on the maximum likelihood criterion in any one of claims 1 to 4.

Images