A pilot-derived signal detection method and related apparatus

Abstract

The application discloses a pilot-derived signal detection method and related device, and belongs to the technical field of wireless communication, which comprises the following steps: constructing a full combination candidate point set and a random candidate point set based on each received pilot point in the subspace, constructing a total candidate point set based on the full combination candidate point set and the random candidate point set; calculating the estimation result of the interference noise covariance matrix of the communication system based on the total candidate point set; and detecting the received signal based on the estimation result of the interference noise covariance matrix, wherein the method and the related device can accurately estimate the covariance and improve the precision of signal detection.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology and relates to a signal detection method and related apparatus based on pilot frequency derivation. Background Technology

[0002] As mobile communication systems evolve towards higher speeds, higher reliability, lower latency, and wider connectivity, multi-antenna technology has become a key means to improve system capacity and coverage performance. By configuring more antennas on the base station side, finer beamforming can be formed in the spatial domain, allowing multiple terminal users to be served simultaneously on the same spectrum resources, achieving higher spectral efficiency and stronger coverage gain. Meanwhile, MIMO (Multiple-Input Multiple-Output) technology, by spatially multiplexing multiple users in parallel on the same time-frequency resources, is widely considered a key technology for improving spectral efficiency.

[0003] However, in multi-user and multi-cell scenarios, the receiver needs to recover the desired user data under strong interference. The main challenges include: 1) Stronger interference from the same frequency and resources. Parallel transmission by multiple users results in multiple signals being superimposed on the same time-frequency resources. The receiver must not only suppress multi-user interference within the same cell but also suppress co-channel interference from neighboring cells, making the interference statistics more complex and faster-changing. 2) Limited pilot resources lead to insufficient estimation. Due to the limited channel coherence time and bandwidth, the pilot resources available for estimation are often limited. In multi-user and multi-cell conditions, pilot overhead increases more significantly with the number of users, and practical systems often have to reduce the number of pilots or reuse pilots. 3) Pilot reuse causes pilot pollution. In non-cooperative multi-cell systems, to save overhead, the same or related pilot sequences are reused in different cells, causing channel estimation to be interfered with by users in other cells and significantly limiting the performance of multi-antenna systems. The above factors together result in the fact that in multi-user, multi-cell, multi-antenna systems, interference and noise statistics, especially covariance, are difficult to estimate accurately and stably under finite pilot conditions, which in turn affects the interference suppression and detection performance of the receiver.

[0004] In multi-antenna receivers, common linear interference suppression techniques rely on the statistical characteristics of interference plus noise to construct receiver weights. Therefore, the quality of covariance estimation often determines the effectiveness of interference suppression. A common engineering approach is to first subtract the known pilot signals from the received signal to obtain residuals, and then use these residuals to form a covariance estimate. Besides this, there are shrinkage estimation methods. Under conditions of small samples and high dimensionality, shrinking the sample covariance towards a structured target matrix can significantly reduce the mean square error and improve the condition number. Alternatively, data detection results can be used to further assist in covariance calculation.

[0005] The estimation accuracy is significantly insufficient when pilot samples are insufficient. Covariance estimation obtained directly from pilot residuals can achieve near-unbiasedness in a statistical sense under ideal assumptions. However, the problem lies in the fact that when the number of pilot samples is small and the covariance dimension increases with the number of receiving antennas, the fluctuation of the estimation error will increase significantly. When the number of observations is insufficient to support stable estimation of high-dimensional covariance, large estimation errors, ill-conditioned matrices, and even rank deficiency may occur.

[0006] Regularization methods such as shrinkage estimation are insufficient to fundamentally improve accuracy with small samples and suffer from engineering adaptability issues. Shrinkage estimation essentially makes the matrix more usable but cannot address the root cause of statistical unreliability due to insufficient samples, which in turn leads to poor signal detection accuracy. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide a signal detection method and related apparatus based on pilot derivatives. This method and related apparatus can accurately estimate the covariance and improve the accuracy of signal detection.

[0008] To achieve the above objectives, this invention discloses a signal detection method based on pilot derivatives, comprising:

[0009] Obtain the pilot residual matrix of the communication system, project the pilot residual matrix of the communication system onto the principal subspace, and obtain the projection position of the pilot residual matrix of the communication system in the principal subspace;

[0010] Based on the projection position of the pilot residual matrix of the communication system in the main subspace, a symmetric point set matrix is ​​constructed, and the symmetric point set matrix is ​​projected onto the subspace to form a number of receiving pilot points in the subspace.

[0011] Based on each receiving pilot point in the subspace, a full combination candidate point set and a random candidate point set are constructed, and a total candidate point set is constructed based on the full combination candidate point set and the random candidate point set;

[0012] The estimation result of the interference noise covariance matrix of the communication system is calculated based on the total candidate point set;

[0013] The received signal is detected based on the estimation result of the interference noise covariance matrix.

[0014] Furthermore, the process of projecting the pilot residual matrix of the communication system into the principal subspace is as follows:

[0015] The pilot residual matrix of the communication system is subjected to singular value decomposition to obtain the left singular vector matrix and the non-negative diagonal matrix of SVD (Singular Value Decomposition).

[0016] An energy sequence is constructed based on the diagonal elements of the non-negative diagonal matrix;

[0017] Set an energy accumulation ratio threshold, and determine the minimum subspace dimension r based on the energy accumulation ratio threshold and the energy sequence;

[0018] Obtain the first r columns of the left singular vector matrix of the SVD, and project the pilot residual matrix of the communication system into the principal subspace based on the first r columns of the left singular vector matrix of the SVD.

[0019] Furthermore, the process of constructing a full set of candidate points and a random set of candidate points based on each receiving pilot point in the subspace is as follows:

[0020] Calculate the subspace squared Euclidean distance from each receiving pilot point to other receiving pilot points;

[0021] Based on the squared Euclidean distance of the subspace, each receiving pilot point is sorted from smallest to largest to obtain the sorting result of each receiving pilot point;

[0022] Set the difference vector and the upper limit of the neighbor pool size used for random combination. Based on the difference vector, the upper limit of the neighbor pool size used for random combination. The sorting results of each receiving pilot point are used to construct a random combination neighbor set matrix;

[0023] Determine the upper limit of the number of neighbors used for the entire combination. Based on the difference vector, the upper limit of the number of neighbors used for the full combination. The sorting results of each received pilot point are used to construct a full combination neighbor set matrix;

[0024] Define a coefficient vector, and generate a set of candidate points for the full combination based on the coefficient vector and the full combination neighbor set matrix;

[0025] A random candidate point set is generated based on the random combination of the neighbor set matrix.

[0026] Furthermore, the process of generating a random candidate point set based on the random combination neighbor set matrix is ​​as follows:

[0027] Generate each receiving pilot point separately A number of random candidate points;

[0028] Determine the order of each random candidate point ;

[0029] Draw from the neighbor pool used for random combination using a method of random sampling without replacement. Each index is used to form an index set for each receiving pilot point;

[0030] Generate a binary coefficient vector for each receiving pilot point based on the index set of each receiving pilot point;

[0031] Random candidate points for each receiving pilot point are generated based on the binary coefficient vector of each receiving pilot point, and a random candidate point set is constructed based on the random candidate points of each receiving pilot point.

[0032] Furthermore, the process of calculating the estimation result of the interference noise covariance matrix of the communication system based on the total candidate point set is as follows:

[0033] The candidate points in the total candidate point set are filtered to obtain the filtered point set. ;

[0034] when hour, Given the number of points in the filtered point set, the estimated result of the interference noise covariance matrix of the communication system is... for:

[0035]

[0036] when When, the estimation result of the interference noise covariance matrix of the communication system is... for:

[0037]

[0038] in, Indicates the number of pilot resources. This represents the pilot residual matrix of the communication system.

[0039] Furthermore, the process of filtering candidate points in the total candidate point set is as follows:

[0040] Calculate the square of the Euclidean distance between each candidate point in the total candidate point set;

[0041] Construct a set of Euclidean distance squares based on the squares of the Euclidean distances to each candidate point;

[0042] The candidate points are sorted according to the square of the Euclidean distance to obtain the sorting result;

[0043] Set quantile parameters, and determine a threshold based on the quantile parameters and the set of squared Euclidean distances;

[0044] Based on the sorting results, candidate points of a threshold number are selected from the total candidate point set.

[0045] Furthermore, the process of detecting the received signal based on the estimation result of the interference noise covariance matrix is ​​as follows:

[0046] The detection weight matrix is ​​determined based on the estimation result of the interference noise covariance matrix. ;

[0047] Based on the detection weight matrix The estimated value of the received signal is determined, and a constellation decision is made on the estimated value of the received signal to obtain the detected signal.

[0048] This invention discloses a signal detection system based on pilot derivatives, comprising:

[0049] The acquisition module is used to acquire the pilot residual matrix of the communication system, project the pilot residual matrix of the communication system into the principal subspace, and obtain the projection position of the pilot residual matrix of the communication system in the principal subspace.

[0050] The projection module is used to construct a symmetric point set matrix based on the projection position of the pilot residual matrix of the communication system in the main subspace, and project the symmetric point set matrix onto the subspace to form a plurality of receiving pilot points in the subspace.

[0051] The construction module is used to construct a full combination candidate point set and a random candidate point set based on each receiving pilot point in the subspace, and to construct a total candidate point set based on the full combination candidate point set and the random candidate point set;

[0052] The calculation module is used to calculate the estimation result of the interference noise covariance matrix of the communication system based on the total candidate point set;

[0053] The detection module is used to detect the received signal based on the estimation result of the interference noise covariance matrix.

[0054] The present invention discloses a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the pilot-derived signal detection method.

[0055] The present invention discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the pilot-derived signal detection method.

[0056] The present invention has the following beneficial effects:

[0057] In specific operation, the pilot-derived signal detection method and related apparatus of the present invention construct a full combination candidate point set and a random candidate point set based on each receiving pilot point in the subspace. Based on the full combination candidate point set and the random candidate point set, a total candidate point set is constructed to increase the number of candidate points and fundamentally solve the statistical unreliability problem caused by insufficient samples. Then, based on the total candidate point set, the estimation result of the interference noise covariance matrix of the communication system is calculated, thereby obtaining a more stable and accurate estimation result of the interference plus noise statistical matrix, which improves the detection performance and robustness of the linear interference suppression receiver, and thus improves the accuracy of signal detection. Attached Figure Description

[0058] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 This is a flowchart of the method of the present invention;

[0060] Figure 2a This is a simulation result image from simulation experiment one;

[0061] Figure 2b This is another simulation result image from Simulation Experiment 1;

[0062] Figure 3 This is a distribution diagram of candidate points in simulation experiment two;

[0063] Figure 4 This is a system structure diagram of the present invention. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0065] The accompanying drawings illustrate various structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0066] Considering uplink multi-antenna reception scenarios, the receiver has Each user has a root receiving antenna. Given the transmitting antenna and spatial flow, and the simultaneous superposition of interfering users from three neighboring cells at the receiving end, the communication system model is as follows:

[0067]

[0068] in, Represents the received signal matrix; Represents the desired user channel matrix; This indicates that the user is expected to send a symbol matrix; Indicates the first Channel matrix from interfering users to the receiver; Indicates the first The transmitted symbol matrix of the interfering users; This represents the Gaussian white noise matrix at the receiving end.

[0069] To detect the received signal, the receiver uses the IRC (Interference Rejection Combining) algorithm for interference suppression. IRC is a linear detection method that utilizes the statistical characteristics of interference and noise. The receiver constructs a detection weight matrix. Perform linear transformation on the received signal The symbol matrix to be sent to the desired user is obtained. The estimation of the detection weight matrix The expression is:

[0070]

[0071] in, Represents the interference noise covariance matrix; This represents the power of the Gaussian white noise; Represents a unit array.

[0072] In summary, the core premise of IRC is to accurately obtain the interference noise covariance matrix, which is generally achieved by transmitting a known pilot signal. Specifically, during the pilot phase, it is expected that the user will transmit a known pilot signal. The receiving end receives the pilot signal. ,Right now:

[0073]

[0074] in, Indicates the first The pilot signal sent by the interfering user.

[0075] In the expected user channel matrix Given the information, construct the pilot residual matrix. for:

[0076]

[0077] Among them, the pilot residual matrix Each column in Each of these can be considered as an observation sample of superimposed interference noise, from which the interference noise covariance matrix can be estimated, thus obtaining the interference noise covariance matrix. estimation results :

[0078]

[0079] in, It represents the mathematical expectation.

[0080] However, since the number of pilot signals that can be transmitted in a practical communication system is limited, existing technologies do not address the interference noise covariance matrix. The estimated results are inaccurate.

[0081] Example 1

[0082] To solve the above problems, refer to Figure 1 This invention discloses a signal detection method based on pilot derivatives, comprising the following steps:

[0083] 1) Obtain the pilot residual matrix of the communication system, project the pilot residual matrix of the communication system into the principal subspace, and obtain the projection position of the pilot residual matrix of the communication system in the principal subspace;

[0084] The specific operation of step 1) is as follows:

[0085] 11) Obtain the pilot residual matrix of the communication system ;

[0086] 12) Project the pilot residual matrix of the communication system into the principal subspace;

[0087] The specific operation of step 12) is as follows:

[0088] 121) Perform singular value decomposition on the pilot residual matrix of the communication system to obtain the left singular vector matrix and the non-negative diagonal matrix of SVD;

[0089] Specifically, determine the number of pilot samples; when the number of pilot samples... When it is less than the preset value, for example Then the interference noise covariance matrix is ​​calculated directly. estimation results for:

[0090]

[0091] When the number of pilot samples If the value is greater than or equal to the preset value, then the pilot residual matrix is... Perform singular value decomposition ,in, Denotes the left singular vector matrix of SVD. Represents a non-negative diagonal matrix. The diagonal elements are singular values. , This represents a right singular vector matrix.

[0092] 122) Based on the aforementioned non-negative diagonal matrix Construct an energy sequence from the diagonal elements in the sequence;

[0093] Specifically, based on the non-negative diagonal matrix The singular values ​​in the sequence are used to construct the energy sequence, where the first singular value in the sequence is the first singular value. element for:

[0094]

[0095] in, For pilot resources, The number of elements in the energy sequence.

[0096] 123) Set an energy accumulation ratio threshold, and determine the minimum subspace dimension r based on the energy accumulation ratio threshold and the energy sequence;

[0097] Specifically, given a cumulative energy ratio threshold The minimum subspace dimension is determined based on the energy sequence and the energy accumulation ratio threshold. , so that:

[0098] .

[0099] 124) Obtain the first r columns of the left singular vector matrix of the SVD, and project the pilot residual matrix of the communication system into the principal subspace based on the first r columns of the left singular vector matrix of the SVD.

[0100] Take the first r columns of the left singular vector matrix of the SVD. for:

[0101]

[0102] The pilot residual matrix U is used Projecting onto the principal subspace, the projection position of the pilot residual matrix U onto the principal subspace is... for:

[0103] .

[0104] 2) Based on the projection positions of the pilot residual matrix of the communication system in the principal subspace, construct a symmetric point set matrix. The symmetric point set matrix The projection is applied to a subspace, forming several receiving pilot points in the subspace;

[0105] The specific operation of step 2) is as follows:

[0106] The projection position of the pilot residual matrix U of the communication system in the principal subspace Taking symmetry, we obtain the matrix of symmetric point sets. for:

[0107]

[0108] in, ; This represents the number of symmetric pilot residual matrices after expanding the symmetric points; the symmetric point set matrix. The first in Listed as Projecting the symmetric point set matrix P onto the subspace yields the coordinate representation of the symmetric point set matrix P in the subspace. :

[0109]

[0110] Among them, the symmetric point set matrix P has the first... List Coordinate representation in subspace .

[0111] 3) Construct a full combination candidate point set and a random candidate point set based on each receiving pilot point in the subspace, and construct a total candidate point set based on the full combination candidate point set and the random candidate point set;

[0112] The specific process of step 3) is as follows:

[0113] 31) Parameter settings;

[0114] set up This is the upper limit for the number of neighbors used for the entire combination. ; This is the upper limit of the neighbor pool size used for random combination. ; The number of candidate points randomly generated for each pilot point; This represents the upper limit of the number of difference vectors superimposed in each random combination. The complete set of coefficients uses the set of ternary coefficients. Randomly combine the set of binary coefficients .

[0115] 32) Calculate the subspace squared Euclidean distance from each receiving pilot point to other receiving pilot points;

[0116] Let the indices of all receiving pilot points in the subspace be . For the first One receiving pilot point The corresponding subspace receiving pilot signal is In the coordinates of the subspace, for all but the first... All receiving pilot points other than the one receiving pilot point Calculate all the received pilot points To the receiving pilot point square Euclidean distance of the subspace for:

[0117]

[0118] in, For the first The receiving pilot point to the first The subspace squared Euclidean distance between the receiving pilot points.

[0119] 33) Sort each receiving pilot point from smallest to largest based on the square Euclidean distance of the subspace to obtain the sorting result of each receiving pilot point;

[0120] For all the receiving pilot points Sort the corresponding subspaces according to their squared Euclidean distance in ascending order to obtain the sorted index sequence:

[0121]

[0122] satisfy ;

[0123] 34) Define a difference vector, and based on the difference vector, set an upper limit for the size of the nearest neighbor pool used for random combination. The sorting results of each receiving pilot point are used to construct a random combination neighbor set matrix;

[0124] 35) Determine the upper limit of the number of neighbors used for the entire combination. Based on the difference vector, the upper limit of the number of neighbors used for the full combination. The sorting results of each received pilot point are used to construct a full combination neighbor set matrix;

[0125] Specifically, let the difference vector be... Construct a full combinatorial neighbor set matrix and a random combinatorial neighbor set matrix, wherein the first... The complete neighborhood set matrix of each receiving pilot point ;No. A random combination of neighbor set matrix of receiving pilot points .

[0126] 36) Define a coefficient vector, and generate a set of candidate points for the entire combination based on the coefficient vector and the matrix of the nearest neighbor set of the entire combination;

[0127] Let the coefficient vector be for:

[0128]

[0129] Each component satisfies The number of coefficient vectors is Because of the common Each dimension can be taken as a single dimension. Three values, and by performing all combinations of these values, a total of A vector of coefficients. For example, when When, the set of coefficients is There are 3 types in total; when When, the coefficient combination is There are 9 types in total; when At that time, there are a total of kind.

[0130] List all combination coefficients into a matrix :

[0131]

[0132] Then choose the first one. When receiving pilot points, the generated full set of candidate points is obtained. for:

[0133]

[0134] in, Indicates the selection of the first When there are 1 receiving pilot point, the full set of candidate points is generated.

[0135] 37) Generate a set of random candidate points based on the aforementioned random combination neighbor set matrix;

[0136] The specific operation of step 37) is as follows:

[0137] 371) Generate each receiving pilot point separately A number of random candidate points;

[0138] To supplement a larger neighborhood coverage, generate pilot signals for each receiving pilot point. A random candidate point, .

[0139] 372) Determine the order of each random candidate point. ;

[0140] Determine the first The order of random candidate points for:

[0141]

[0142] in, Indicates the generation of the first The order when there are a number of random candidate points. It indicates a uniform distribution.

[0143] 373) Draw from the neighbor pool used for random combination using a method of random sampling without replacement. Each index is used to form an index set for each receiving pilot point;

[0144] Specifically, from The method of sampling without replacement is used. Each index forms an index set:

[0145]

[0146] in, To generate the first The set of indices for a random candidate point. Represents a set The base number.

[0147] 374) Generate a binary coefficient vector for each receiving pilot point based on the index set of each receiving pilot point.

[0148] Specifically, the generated binary coefficient vector is:

[0149]

[0150] in, To generate the first A binary coefficient vector for a random candidate point A binary coefficient vector The first in Each element.

[0151] 375) Generate random candidate points for each receiving pilot point based on the binary coefficient vector of each receiving pilot point, and construct a random candidate point set based on the random candidate points of each receiving pilot point;

[0152] Specifically, construct random candidate points:

[0153]

[0154] in, Indicates the selection of the first When the first receiving pilot point is reached, the generated first... A random candidate point, express All lines, Each column represents a random candidate point, which is then combined column by column to obtain the first column. A set of random candidate points for receiving pilot points .

[0155] 38) Construct a total candidate point set based on the full set of candidate points and the random set of candidate points;

[0156] The symmetric point set matrix is The generated full combination of candidate points Generated random candidate points They were unified and merged into a total candidate point set. for:

[0157]

[0158] 4) Calculate the estimation result of the interference noise covariance matrix of the communication system based on the total candidate point set;

[0159] The specific operation of step 4) is as follows:

[0160] The candidate points in the total candidate point set are filtered to obtain the filtered point set. , This represents the number of points in the filtered point set.

[0161] Specifically, calculate each candidate point The square of the Euclidean distance:

[0162]

[0163] Form a set of squared Euclidean distances:

[0164]

[0165] Set quantile parameters For example, 0.8, set a threshold. for of Quantiles, after sorting:

[0166]

[0167] in, for The number of elements in the middle, The bottom right subscript in the table represents the position after sorting by Euclidean distance, so we take:

[0168]

[0169] in, This indicates rounding up, thus obtaining the filtered point set. for:

[0170]

[0171] remember The number of points after filtering.

[0172] Using the filtered point set Covariance estimation is performed to obtain the estimated result of the interference noise covariance matrix, where the filtered point set is:

[0173]

[0174] when At that time, the estimation result of the interference noise covariance matrix is ​​obtained. for:

[0175]

[0176] when or Then, take the estimated results of the basic sample covariance and the interference noise covariance matrix. for:

[0177]

[0178] 5) Detect the received signal based on the estimation result of the interference noise covariance matrix;

[0179] The specific operation of step 5) is as follows:

[0180] 51) Determine the detection weight matrix based on the estimation result of the interference noise covariance matrix. ;

[0181] The estimation results of the interference noise covariance matrix As The detection weight matrix is ​​obtained. for:

[0182]

[0183] in, This is the regularization parameter, typically set to... , for An identity matrix of order 1;

[0184] During the signal reception phase, the weighting matrix of the received signal... for:

[0185]

[0186] in, For the data that the user is expected to send, The data were sent to three interfering users in the neighboring cell.

[0187] 52) Based on the detection weight matrix Determine the estimated value of the received signal, perform constellation decision on the estimated value of the received signal, and obtain the detected signal;

[0188] Specifically, based on the detection weight matrix The received signal is estimated and detected to obtain an estimated value of the received signal. for:

[0189]

[0190] Among them, the estimated value of the received signal A constellation judgment is performed to obtain the detected signals.

[0191] Simulation Experiment 1

[0192] Employing an uplink multi-user MIMO receive model, the desired spatial stream number at the user transmitter is... Number of antennas at the receiving end The desired user and three interfering users from neighboring cells all use BPSK modulation. At each signal-to-noise ratio (SNR) point, 120 independent time slots are simulated. Each time slot independently generates the transmitted symbol vector, channel, and noise, and transmits 20,000 bits of data per time slot. The number of bits transmitted in all time slots and the number of erroneous bits are counted to obtain the final bit error rate (BER). Two sets of parameters are simulated. The first set includes the number of pilot signals transmitted. The interference ratio (INR) is 15dB; the number of pilot signals transmitted in the second group is... The interference noise ratio (INR) is 5 dB. Within each time slot, the desired channel and the three interfering channels are generated using independent Rayleigh block fading. To ensure fair comparisons under different signal-to-noise ratios (SNR) and interference noise ratios (INR), a fixed total noise power is used, and the INR and SNR are set by scaling the channels.

[0193] In the simulation, the power of the channel matrix is ​​set according to different signal-to-noise ratios (SNR), and four different weight matrices are calculated during the pilot transmission and reception phases:

[0194] Method 1: Treat both interference and noise as white noise.

[0195] Method 2: Calculate the covariance estimate directly using the existing pilot signals.

[0196] Method 3: Use the true covariance as the ideal performance benchmark.

[0197] Method 4: Using the present invention;

[0198]

[0199] in, This is the covariance matrix calculated in this invention.

[0200] During the signal reception phase, the received signal is:

[0201]

[0202] in, For the data that the user is expected to send, The data were sent by three interfering users, and the received signal was detected using the four methods described above:

[0203]

[0204] in, This is an estimate of the signal. For the first The weighting matrix of the method is obtained, and then the estimated signal is subjected to BPSK decision, and the bit error rate is calculated.

[0205] Bit error rate curves for four methods at different signal-to-noise ratios, such as Figure 2aand Figure 2b As shown, in summary Figure 2a and Figure 2b As can be seen, the present invention continues to maintain a lower BER compared to the pilot sample estimation algorithm, and shows a trend of moving closer to the optimal solution, indicating that the present invention can stably improve the covariance estimation and detection performance under different interference intensities and pilot configurations.

[0206] Simulation Experiment 2

[0207] The base station is configured with two receiving antennas, one desired user and two interfering users, each with only one transmitting antenna. A real-number system is used, and each interfering user employs 8PAM (Pulse Amplitude Modulation). Assuming the pilot received signal is noise-free, candidate points are generated using this invention. Due to the relatively low order, only full combination is used, not random combination; for each pilot, only the two nearest pilots are selected for full combination.

[0208] Figure 3 The diagram illustrates the candidate point distribution generated using neighbor differential combination when two interferences exist, the system is a real number system, the signal is modulated with 8PAM, and the pilot signal is noise-free. Figure 3 As shown, the candidate points generated by this invention are quite comprehensive.

[0209] Example 2

[0210] refer to Figure 4 The pilot-derived signal detection system of the present invention includes:

[0211] The acquisition module is used to acquire the pilot residual matrix of the communication system, project the pilot residual matrix of the communication system into the principal subspace, and obtain the projection position of the pilot residual matrix of the communication system in the principal subspace.

[0212] The projection module is used to construct a symmetric point set matrix based on the projection position of the pilot residual matrix of the communication system in the main subspace, and project the symmetric point set matrix onto the subspace to form a plurality of receiving pilot points in the subspace.

[0213] The construction module is used to construct a full combination candidate point set and a random candidate point set based on each receiving pilot point in the subspace, and to construct a total candidate point set based on the full combination candidate point set and the random candidate point set;

[0214] The calculation module is used to calculate the estimation result of the interference noise covariance matrix of the communication system based on the total candidate point set;

[0215] The detection module is used to detect the received signal based on the estimation result of the interference noise covariance matrix.

[0216] The module division in this embodiment is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in each embodiment of this application can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0217] Example 3

[0218] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the pilot-derived signal detection method, for example, including: acquiring the pilot residual matrix of a communication system; projecting the pilot residual matrix of the communication system into a principal subspace to obtain the projection position of the pilot residual matrix of the communication system in the principal subspace; constructing a symmetric point set matrix based on the projection position of the pilot residual matrix of the communication system in the principal subspace; projecting the symmetric point set matrix into a subspace to form a plurality of receiving pilot points in the subspace; constructing a full combination candidate point set and a random candidate point set based on each receiving pilot point in the subspace; constructing a total candidate point set based on the full combination candidate point set and the random candidate point set; calculating the estimation result of the interference noise covariance matrix of the communication system based on the total candidate point set; and detecting the received signal based on the estimation result of the interference noise covariance matrix. The memory may include RAM, such as high-speed random access memory, or it may also include non-volatile memory, such as at least one disk storage device. The processor, network interface, and memory are interconnected through an internal bus, which may be an industry standard architecture bus, a peripheral component interconnection standard bus, an extended industry standard architecture bus, etc. The bus may be divided into address bus, data bus, control bus, etc.

[0219] Example 4

[0220] A computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the pilot-derived signal detection method. For example, the method includes: obtaining the pilot residual matrix of a communication system; projecting the pilot residual matrix of the communication system into a principal subspace to obtain the projection position of the pilot residual matrix of the communication system in the principal subspace; constructing a symmetric point set matrix based on the projection position of the pilot residual matrix of the communication system in the principal subspace; projecting the symmetric point set matrix into a subspace to form a plurality of receiving pilot points in the subspace; constructing a full combination candidate point set and a random candidate point set based on each receiving pilot point in the subspace; constructing a total candidate point set based on the full combination candidate point set and the random candidate point set; calculating an estimation result of the interference noise covariance matrix of the communication system based on the total candidate point set; and detecting the received signal based on the estimation result of the interference noise covariance matrix. Specifically, the computer-readable storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. The volatile memory may include random access memory (RAM) and / or cache memory, etc. The non-volatile memory may include read-only memory (ROM), hard disk, flash memory, optical disk, magnetic disk, etc.

[0221] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and disclosure of the invention. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein.

[0222] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

[0223] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the present invention. Any simple modifications, alterations, or equivalent structural changes made to the above embodiments based on the technical essence of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A signal detection method based on pilot derivatives, characterized in that, include: Obtain the pilot residual matrix of the communication system, project the pilot residual matrix of the communication system onto the principal subspace, and obtain the projection position of the pilot residual matrix of the communication system in the principal subspace; Based on the projection position of the pilot residual matrix of the communication system in the main subspace, a symmetric point set matrix is ​​constructed, and the symmetric point set matrix is ​​projected onto the subspace to form a number of receiving pilot points in the subspace. Based on each receiving pilot point in the subspace, a full combination candidate point set and a random candidate point set are constructed, and a total candidate point set is constructed based on the full combination candidate point set and the random candidate point set; The estimation result of the interference noise covariance matrix of the communication system is calculated based on the total candidate point set; The received signal is detected based on the estimation result of the interference noise covariance matrix. The process of projecting the pilot residual matrix of the communication system into the principal subspace is as follows: Singular value decomposition (SVD) is performed on the pilot residual matrix of the communication system to obtain the left singular vector matrix and the non-negative diagonal matrix of SVD. An energy sequence is constructed based on the diagonal elements of the non-negative diagonal matrix; Set an energy accumulation ratio threshold, and determine the minimum subspace dimension r based on the energy accumulation ratio threshold and the energy sequence; Obtain the first r columns of the left singular vector matrix of the SVD, and project the pilot residual matrix of the communication system into the principal subspace based on the first r columns of the left singular vector matrix of the SVD.

2. The signal detection method based on pilot derivatives according to claim 1, characterized in that, The process of constructing a full set of candidate points and a random set of candidate points based on each receiving pilot point in the subspace is as follows: Calculate the subspace squared Euclidean distance from each receiving pilot point to other receiving pilot points; Based on the squared Euclidean distance of the subspace, each receiving pilot point is sorted from smallest to largest to obtain the sorting result of each receiving pilot point; Set the difference vector and the upper limit of the neighbor pool size used for random combination. Based on the difference vector, the upper limit of the neighbor pool size used for random combination. The sorting results of each receiving pilot point are used to construct a random combination neighbor set matrix; Determine the upper limit of the number of neighbors for the entire combination. Based on the difference vector, the upper limit of the number of neighbors used for the full combination. The sorting results of each received pilot point are used to construct a full combination neighbor set matrix; Define a coefficient vector, and generate a set of candidate points for the full combination based on the coefficient vector and the full combination neighbor set matrix; A random candidate point set is generated based on the random combination of the neighbor set matrix.

3. The signal detection method based on pilot derivatives according to claim 2, characterized in that, The process of generating a random candidate point set based on the random combination neighbor set matrix is ​​as follows: Generate each receiving pilot point separately A number of random candidate points; Determine the order of each random candidate point ; Draw from the neighbor pool used for random combination using a method of random sampling without replacement. Each index is used to form an index set for each receiving pilot point; Generate a binary coefficient vector for each receiving pilot point based on the index set of each receiving pilot point; Random candidate points for each receiving pilot point are generated based on the binary coefficient vector of each receiving pilot point, and a random candidate point set is constructed based on the random candidate points of each receiving pilot point.

4. The signal detection method based on pilot derivatives according to claim 1, characterized in that, The process of calculating the estimation result of the interference noise covariance matrix of the communication system based on the total candidate point set is as follows: The candidate points in the total candidate point set are filtered to obtain the filtered point set. ; when hour, Given the number of points in the filtered point set, the estimated result of the interference noise covariance matrix of the communication system is... for: when When, the estimation result of the interference noise covariance matrix of the communication system is... for: in, Indicates the number of pilot resources. This represents the pilot residual matrix of the communication system.

5. The signal detection method based on pilot derivatives according to claim 4, characterized in that, The process of filtering candidate points in the total candidate point set is as follows: Calculate the square of the Euclidean distance between each candidate point in the total candidate point set; Construct a set of Euclidean distance squares based on the squares of the Euclidean distances to each candidate point; The candidate points are sorted according to the square of the Euclidean distance to obtain the sorting result; Set quantile parameters, and determine a threshold based on the quantile parameters and the set of squared Euclidean distances; Based on the sorting results, candidate points of a threshold number are selected from the total candidate point set.

6. The signal detection method based on pilot derivatives according to claim 1, characterized in that, The process of detecting the received signal based on the estimation result of the interference noise covariance matrix is ​​as follows: The detection weight matrix is ​​determined based on the estimation result of the interference noise covariance matrix. ; Based on the detection weight matrix The estimated value of the received signal is determined, and a constellation decision is made on the estimated value of the received signal to obtain the detected signal.

7. A signal detection system based on pilot frequency derivation, characterized in that, include: The acquisition module is used to acquire the pilot residual matrix of the communication system, project the pilot residual matrix of the communication system into the principal subspace, and obtain the projection position of the pilot residual matrix of the communication system in the principal subspace. The projection module is used to construct a symmetric point set matrix based on the projection position of the pilot residual matrix of the communication system in the main subspace, and project the symmetric point set matrix onto the subspace to form a plurality of receiving pilot points in the subspace. The construction module is used to construct a full combination candidate point set and a random candidate point set based on each receiving pilot point in the subspace, and to construct a total candidate point set based on the full combination candidate point set and the random candidate point set; The calculation module is used to calculate the estimation result of the interference noise covariance matrix of the communication system based on the total candidate point set; The detection module is used to detect the received signal based on the estimation result of the interference noise covariance matrix; The process of projecting the pilot residual matrix of the communication system into the principal subspace is as follows: Singular value decomposition (SVD) is performed on the pilot residual matrix of the communication system to obtain the left singular vector matrix and the non-negative diagonal matrix of SVD. An energy sequence is constructed based on the diagonal elements of the non-negative diagonal matrix; Set an energy accumulation ratio threshold, and determine the minimum subspace dimension r based on the energy accumulation ratio threshold and the energy sequence; Obtain the first r columns of the left singular vector matrix of the SVD, and project the pilot residual matrix of the communication system into the principal subspace based on the first r columns of the left singular vector matrix of the SVD.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the pilot-derived signal detection method as described in any one of claims 1-6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the pilot-derived signal detection method as described in any one of claims 1-6.

Images