User activity detection and user channel estimation methods, electronic devices, and storage media

Abstract

The application provides a user activity detection and user channel estimation method, an electronic device and a storage medium, and belongs to the technical field of communication. The method comprises the following steps: establishing a transmission system model of mMTC uplink; constructing a prior probability model according to the transmission system model and user activity; establishing a joint variable according to the user activity and the angle domain of the system channel; inputting the joint variable into the prior probability model to obtain the logarithmic probability density of the joint variable; performing likelihood probability calculation according to the pilot signal and the joint variable to obtain the logarithmic likelihood probability of the pilot signal with respect to the joint variable; constructing a maximum a posteriori probability problem according to the logarithmic probability density and the logarithmic likelihood probability; and performing maximum minimum solving on the maximum a posteriori probability problem to obtain the user activity detection result and the user channel estimation result. The scheme of the application can improve the accuracy of user activity detection and user channel estimation, and has good universality.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of communication, in particular to a user activity detection and user channel estimation method, an electronic device and a storage medium. BACKGROUND

[0002] In a massive Machine Type Communication (mMTC) system, due to a large number of antennas, if a high-resolution analog-to-digital converter is used in the receiver of a base station, the cost will be very high, so in order to save cost, a low-resolution analog-to-digital converter is usually used in the receiver. In the related art, the influence of the low-resolution analog-to-digital converter on the accuracy of user activity detection and user channel estimation is reduced by introducing quantization effect, but this way has higher requirements for the pilot measurement matrix, so the universality for different application scenarios is poor. SUMMARY

[0003] The main purpose of the embodiments of the present application is to provide a user activity detection and user channel estimation method, which introduces hierarchical joint sparsity representing user activity and system channel to perform user activity detection and user channel estimation, can better simulate the real physical architecture performance of the massive Machine Type Communication (mMTC) system, thereby improving the accuracy of user activity detection and user channel estimation, and has good universality.

[0004] To achieve the above purpose, a first aspect of the embodiments of the present application provides a user activity detection and user channel estimation method, which comprises:

[0005] establishing a transmission system model of mMTC uplink; wherein the transmission system model comprises a base station side and a user side, the base station side and the user side communicate through a system channel, and the user side is used to send a pilot signal to the base station side according to user activity;

[0006] constructing a prior probability model according to the transmission system model and user activity;

[0007] establishing a joint variable according to user activity and an angle domain of the system channel; wherein the joint variable is used to represent hierarchical joint sparsity of user activity and the system channel;

[0008] inputting the joint variable into the prior probability model to obtain a log probability density of the joint variable;

[0009] performing likelihood probability calculation according to the pilot signal and the joint variable to obtain a log likelihood probability of the pilot signal with respect to the joint variable;

[0010] Constructing a posterior probability maximization problem based on the log probability density and the log likelihood probability;

[0011] The maximum a posteriori probability problem is solved by a maximum minimization method to obtain a user activity detection result and a user channel estimation result.

[0012] In some embodiments, in the transmission system model, the base station side includes a base station with M antennas, and the user side includes N single-antenna user terminal devices;

[0013] The user terminal device has the characteristic of being occasionally active, and the activity status of the nth user terminal device is expressed as:

[0014]

[0015] Wherein, n is a positive integer between [1, N], active indicates that the user terminal device is in an active state, and none indicates that the user terminal device is in an inactive state. When the user terminal device is in an active state, the pilot signal is sent to the base station side;

[0016] The pilot signal is a pilot sequence signal including T symbols, and the pilot signal is expressed as:

[0017]

[0018] Among them, s n Indicates the activity of the nth user terminal device, h n ∈C M×1 is the channel between the nth user terminal device and the base station side, is the power coefficient including the transmit power and channel path loss, represents the pilot sequence assigned to the nth user terminal device, V∈C M×T is a complex Gaussian noise matrix with zero mean and covariance matrix, H = [h1,Λ,h N ]∈C M×N is the channel matrix between the user side and the base station side, S=diag[s1,Λ,s N ] is a diagonal matrix describing the activity of the user terminal device, is the large-scale fading matrix between the user side and the base station side, is the pilot sequence matrix on the user side;

[0019] The angle domain channel matrix between the user side and the base station side is:

[0020]

[0021] in, H = [h1, Λ, hn] is an angle domain channel matrix between the user side and the base station side, N M×N H is a channel matrix between the user side and the base station side, H is an array response channel matrix between the user side and the base station side,

[0022] H is an angle domain channel matrix The element in the mth row and the nth column of H is subject to a Gaussian distribution as follows:

[0023]

[0024] wherein m is a positive integer between 1 and M, H is an angle domain channel matrix The element in the mth row and the nth column of H is denoted as γ (m,n) is a Gaussian distribution precision, and CN(*) represents an N-dimensional complex number vector space function;

[0025] The probability density of γ (m,n) is as follows:

[0026]

[0027] wherein Γ(a) is a gamma function, and a and b are hyperparameters greater than zero.

[0028] In some embodiments, the active probability density of the nth user terminal device in the prior probability model is as follows:

[0029]

[0030] wherein q s represents a prior probability of the user terminal device being active, q s is in the range of (0, 1), and s n represents the activity of the nth user terminal device.

[0031] In some embodiments, the joint variable is as follows: wherein H is an angle domain channel matrix between the user side and the base station side, and S = diag[s1, Λ, sn] is a diagonal matrix describing the activity of the user terminal device. N

[0032] When s n = 1, the nth user terminal device is in an active state, and the distribution function of the element in the mth row and the nth column of the joint variable X is as follows:

[0033]

[0034] wherein X​​(m,n) denotes the element in the mth row and nth column of the joint variable X, s n denotes the activity of the nth user equipment, denotes the element in the mth row and nth column of the angle domain channel matrix denotes the element in the mth row and nth column of the angle domain channel matrix (m,n) is the precision of the Gaussian distribution;

[0035] when s n = 0, the nth user equipment is in an inactive state, X (m,n) is zero, at which time the distribution function of X (m,n) is:

[0036]

[0037] wherein, δ(*) denotes an impact function, CN(*) denotes an N-dimensional complex vector space function, X (m,n) denotes the element in the mth row and nth column of the joint variable X, s n denotes the activity of the nth user equipment, denotes the element in the mth row and nth column of the angle domain channel matrix denotes the element in the mth row and nth column of the angle domain channel matrix

[0038] In some embodiments, the functional expression of the log probability density is:

[0039]

[0040] wherein, x is a vectorized real quantity of the joint variable X, is the ith r = (n-1)M+1 element of x, s n denotes the activity of the nth user equipment, q s denotes the prior probability of the user equipment activity, a, b and ε are all hyperparameters greater than zero, M is the number of antennas at the base station side, N is the number of user equipments at the user side, and n is a positive integer between 1 and N.

[0041] In some embodiments, the step of performing likelihood probability calculation according to the pilot signal and the joint variable to obtain the log likelihood probability of the pilot signal with respect to the joint variable specifically comprises:

[0042] quantizing the pilot signal by using a general B-bit scalar quantizer to obtain a quantized signal; wherein the quantized signal is expressed as:

[0043]

[0044] wherein, R denotes the quantized signal, and Y denotes the pilot signal;

[0045] vectorizing and real decomposition processing on the pilot signal, to obtain a pilot vector signal; wherein the pilot vector signal is expressed as:

[0046] y = R(Vec(Y)) = Φx + v

[0047] wherein y represents the pilot vector signal, Y represents the pilot signal, x is a vectorized real part of a joint variable X, v represents a vectorized real part of a complex Gaussian noise matrix V, and Φ is an observation matrix;

[0048] vectorizing the quantized signal, to obtain a quantized vector signal; wherein the quantized vector signal is expressed as:

[0049]

[0050] wherein r represents the quantized vector signal, R represents the quantized signal, x is a vectorized real part of a joint variable X, v represents a vectorized real part of a complex Gaussian noise matrix V, and Φ is an observation matrix, and the quantized vector signal r is linearly approximated as:

[0051]

[0052] wherein, is a decomposition matrix, R ry is a cross-correlation matrix of r and y, R yy is an autocorrelation matrix of y, n q is a residual noise, indicates that the first and second order statistical information on both sides of the equation is equal, and r is further simplified as:

[0053]

[0054] wherein A = KΦ is a linear parameter, and z = Kv + n q is a Gaussian noise, and a correlation matrix of the Gaussian noise z is

[0055]

[0056] wherein, is a Gaussian noise covariance, K T is a transpose matrix of the decomposition matrix K, K H is a conjugate matrix of the decomposition matrix K, is an autocorrelation matrix of n q , R rr is an autocorrelation matrix of r;

[0057] performing likelihood probability calculation on the pilot vector signal, to obtain the log-likelihood probability; wherein the log-likelihood probability is expressed as:

[0058]

[0059] wherein, r represents the quantized vector signal, A is a linear parameter, x is a vectorized real part of the joint variable X, T is a number of symbols contained in the pilot signal, ∑ is a correlation matrix of Gaussian noise z, and M is a number of antennas on the base station side.

[0060] In some embodiments, the maximization of the posterior probability problem is expressed as:

[0061]

[0062] wherein, logp(r|x) represents the log-likelihood probability, logp(x) represents the log-probability density, and f(x) is an objective function.

[0063] In some embodiments, the step of performing the maximization of the posterior probability problem by maximum-minimization to obtain the user activity detection result and the user channel estimation result specifically comprises:

[0064] performing maximum-minimization iteration on the variable of the objective function to construct an upper bound of the objective function, wherein the upper bound of the objective function is expressed as:

[0065]

[0066] wherein, j is the iteration number, x is a vectorized real part of the joint variable X, x (j) is the iteration result of x at the jth time, T is a number of symbols contained in the pilot signal, Ω2 is a quadratic coefficient matrix of the likelihood term, J is an approximation of the quadratic coefficient matrix of the likelihood term, f is a first residual of the likelihood term, Λ0 (j) , Λ1 (j) , W (j) are all prior quadratic term coefficients of the jth iteration, and g(x (j) ) is a residual function.

[0067] updating the iteration result of the variable of the objective function according to the upper bound of the objective function to obtain a target iteration result, wherein the target iteration result is:

[0068]

[0069] wherein, f(x|x (j) ) represents the upper bound of the objective function, x (j) is the iteration result of x at the jth time, Ω2 is a quadratic coefficient matrix of the likelihood term, J is an approximation of the quadratic coefficient matrix of the likelihood term, f is a first residual of the likelihood term, Λ0 (j) , Λ1 (j) , W (j)a prior quadratic coefficient of the jth iteration;

[0070] performing user activity deconstruction according to the target iteration result to obtain the user activity detection result, and performing user channel estimation according to the target iteration result to obtain the user channel estimation result; wherein a corresponding detection process of the user activity detection result is expressed as:

[0071]

[0072] wherein, is a conditional parameter, n is a positive integer between [1, N], is the ith element of r (n-1)M+1 elements, is the ith element of r +MN elements, q s represents a prior probability of user terminal device activity, a, b and ε are all hyperparameters greater than zero, M is the number of antennas on the base station side, and N is the number of user terminal devices on the user side;

[0073] The user channel estimation result is:

[0074]

[0075] wherein, U R represents an array response channel matrix between the user side and the base station side, is a corresponding target joint variable, is an iteration diagonal matrix corresponding to the description of the activity of the user terminal device.

[0076] To achieve the above purpose, a second aspect of the present application proposes an electronic device, comprising:

[0077] at least one memory;

[0078] at least one processor;

[0079] at least one program;

[0080] The program is stored in the memory, and the processor executes the at least one program to implement the method of the first aspect of the present application as described above.

[0081] To achieve the above purpose, a third aspect of the present application proposes a storage medium, which is a computer readable storage medium, and the computer readable storage medium stores computer executable instructions, and the computer executable instructions are used to make a computer execute:

[0082] The method according to the first aspect.

[0083] The user activity detection and user channel estimation method provided by the embodiments of the present application can better simulate the real physical architecture performance of a large-scale machine type communication system, thereby improving the accuracy of user activity detection and user channel estimation, and has good universality. BRIEF DESCRIPTION OF DRAWINGS

[0084] Figure 1 is a flowchart of the user activity detection and user channel estimation method provided by the embodiments of the present application;

[0085] Figure 2 is a flowchart of the step S500 shown in Figure 1

[0086] Figure 3 is a flowchart of the step S700 shown in Figure 1

[0087] is a schematic diagram of the hardware structure of the electronic device provided by the embodiments of the present application. Figure 4 DETAILED DESCRIPTION

[0088] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not intended to limit the present application.

[0089] It should be noted that the logical order is shown in the flowchart, but in some cases, the steps shown or described can be performed in an order different from that in the flowchart. The terms "first", "second", etc. in the specification and claims and the above drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence.

[0090] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application, and are not intended to limit the present application.

[0091] ​​Moreover, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the disclosure. One skilled in the relevant art will recognize, however, that the

[0092] The flowcharts shown in the drawings are only exemplary illustrations, and do not necessarily include all contents and operations / steps, nor are they necessarily executed in the order described. For example, some operations / steps can be further decomposed, and some operations / steps can be combined or partially combined, so the actual execution order can be changed according to actual conditions.

[0093] Please refer to Figure 1 , Figure 1 For an embodiment of the present application, an optional flowchart of a user activity detection and user channel estimation method, the user activity detection and user channel estimation method includes but is not limited to including steps S100 to S700:

[0094] Step S100, a transmission system model of mMTC uplink is established; wherein the transmission system model includes a base station side and a user side, the base station side and the user side communicate through a system channel, and the user side is used to send a pilot signal to the base station side according to a user activity;

[0095] Step S200, a prior probability model is constructed according to the transmission system model and the user activity;

[0096] Step S300, a joint variable is established according to the user activity and an angle domain of the system channel; wherein the joint variable is used to represent a hierarchical joint sparsity of the user activity and the system channel;

[0097] Step S400, the joint variable is input into the prior probability model to obtain a log probability density of the joint variable;

[0098] Step S500, a likelihood probability calculation is performed according to the pilot signal and the joint variable to obtain a log likelihood probability of the pilot signal with respect to the joint variable;

[0099] Step S600, a maximum a posteriori probability problem is constructed according to the log probability density and the log likelihood probability;

[0100] Step S700, the maximum a posteriori probability problem is solved by maximum minimization to obtain a user activity detection result and a user channel estimation result.

[0101] The steps S100 to S700 shown in the embodiments of the present application can better simulate the real physical architecture performance of a large-scale machine type communication system by introducing a hierarchical joint sparsity representing user activity and system channel, thereby improving the accuracy of user activity detection and user channel estimation, and because it conforms to the physical reality of the communication system, it can also maintain excellent detection and estimation performance in different application scenarios, thus having good universality. In addition, the embodiments of the present application do not require a high-frequency measurement matrix, and can save pilot resources in the case of a base station equipped with a low-precision analog-to-digital converter.

[0102] Some embodiments, in a transmission system model, the base station side includes a base station with M antennas, and the user side includes N single-antenna user equipment;

[0103] The user equipment has the characteristics of occasional activity, and the activity of the nth user equipment is expressed as:

[0104]

[0105] In formula (1), n is a positive integer between [1, N], active represents that the user equipment is in an active state, none represents that the user equipment is in an inactive state, and the user equipment sends a pilot signal to the base station side when it is in an active state;

[0106] The pilot signal is a pilot sequence signal containing T symbols, and the pilot signal is expressed as:

[0107]

[0108] In formula (2), s n represents the activity of the nth user equipment, h n ∈C M×1 is the channel of the nth user equipment and the base station side, is a power coefficient containing transmit power and channel path loss, represents the pilot sequence allocated to the nth user equipment, V∈C M×T is a complex Gaussian noise matrix with zero mean and covariance matrix, H=[h1,Λ,h N ]∈C M×N is a channel matrix between the user side and the base station side, S=diag[s1,Λ,s N is a diagonal matrix describing the activity of the user equipment, is a large-scale fading matrix between the user side and the base station side, is a pilot sequence matrix of the user side;

[0109] The angle domain channel matrix between the user side and the base station side is:

[0110]

[0111] In formula (3), is an angle domain channel matrix between the user side and the base station side, H = [h1, Λ, hn]∈C N ]∈C M ×N is a channel matrix between the user side and the base station side, denotes an array response channel matrix between the user side and the base station side;

[0112] The mth row, nth column element of the angle domain channel matrix obeys the following Gaussian distribution:

[0113]

[0114] In formula (4), m is a positive integer between [1, M], denotes the mth row, nth column element of the angle domain channel matrix , γ (m,n) is a Gaussian distribution precision, CN(*) denotes an N-dimensional complex number vector space function;

[0115] Specifically, the probability density of γ (m,n) is:

[0116]

[0117] In formula (5), Γ(a) is a gamma function, and a and b are both hyperparameters greater than zero. It should be noted that the values of a and b usually tend to zero.

[0118] Some embodiments, in the prior probability model, the active probability density of the nth user terminal device is:

[0119]

[0120] In formula (6), q s denotes the prior probability of the user terminal device being active, q s , the value range of s n denotes the activity of the nth user terminal device. The greater q s indicates the greater the possibility of the user terminal device being active. It should be noted that the specific value of q s is obtained by integrating historical data, which is not specifically limited here.

[0121] Some embodiments, the joint variable is:

[0122]

[0123] In formula (7), is an angle domain channel matrix between the user side and the base station side, S = diag[s1, Λ, sN] is a diagonal matrix describing the activity of the user terminal device. N The embodiment of the application combines the activity of the user and the sparsity of the angle domain of the system channel, and thus can better simulate the real physical architecture performance of a large-scale machine type communication system.

[0124] When s n = 1, the nth user terminal device is in an active state, and the distribution function of the element in the mth row and the nth column of the joint variable X is as follows:

[0125]

[0126] In formula (8), X (m,n) denotes the element in the mth row and the nth column of the joint variable X, s n denotes the activity of the nth user terminal device, denotes the element in the mth row and the nth column of the angle domain channel matrix , γ (m,n) is a Gaussian distribution precision;

[0127] When s n = 0, the nth user terminal device is in an inactive state, X (m,n) is zero, and the distribution function of X (m,n) is as follows:

[0128]

[0129] In formula (9), δ (*) denotes an impact function, CN (*) denotes an N-dimensional complex vector space function, X (m,n) denotes the element in the mth row and the nth column of the joint variable X, s n denotes the activity of the nth user terminal device, denotes the element in the mth row and the nth column of the angle domain channel matrix , and ε is a hyperparameter greater than zero. The embodiment of the application performs approximate processing on the impact function to facilitate calculation and save computing resources.

[0130] In combination with formula (8) and formula (9), the prior probability density of the joint variable X is as follows:

[0131]

[0132] In formula (10), p(s n = 1) is the probability of s n = 1, and p(s n = 0) is the probability of s n = 0.

[0133] Some embodiments, the joint variable X is vectorized and real-decomposed to obtain:

[0134] x = R(Vec(X)) (11)

[0135] In formula (11), x is the vectorized real part of the joint variable X.

[0136] In combination with formula (4) to (11), the functional formula of the logarithmic probability density is:

[0137]

[0138] In formula (12), x is the vectorized real part of the joint variable X, is the i r = (n-1)M+1 elements, s n represents the activity of the nth user terminal device, q s represents the prior probability of the user terminal device activity, a, b and ε are all hyperparameters greater than zero, M is the number of antennas on the base station side, N is the number of user terminal devices on the user side, and n is a positive integer between 1 and N.

[0139] Some embodiments, please refer to Figure 2 , step S500 specifically includes but is not limited to steps S501 to S504:

[0140] Step S501, using a general B-bit scalar quantizer to quantize the pilot signal to obtain a quantized signal; wherein the quantized signal is expressed as:

[0141]

[0142] In formula (13), R represents the quantized signal, and Y represents the pilot signal. The general B-bit scalar quantizer maps the value of the pilot signal to the quantized alphabet according to a certain quantization threshold, thereby completing the quantization.

[0143] Step S502, vectorizing and real-decomposing the pilot signal to obtain a pilot vector signal; wherein the pilot vector signal is expressed as:

[0144] y = R(Vec(Y)) = Φx + v (14)

[0145] In formula (14), y represents the pilot vector signal, Y represents the pilot signal, x is the vectorized real part of the joint variable X, and v represents the vectorized real part of the complex Gaussian noise matrix V, is the observation matrix;

[0146] Step S503, vectorizing the quantized signal to obtain a quantized vector signal; wherein the quantized vector signal is expressed as:

[0147]

[0148] In formula (15), r represents a quantized vector signal, R represents a quantized signal, x is a vectorized real part of a joint variable X, v represents a vectorized real part of a complex Gaussian noise matrix V, and Φ is an observation matrix.

[0149] The quantized vector signal r is linearly approximated as:

[0150]

[0151] In formula (16), r represents a quantized vector signal, R represents a quantized signal, x is a vectorized real part of a joint variable X, v represents a vectorized real part of a complex Gaussian noise matrix V, and Φ is an observation matrix. is a decomposition matrix, R ry is a cross-correlation matrix of r and y, R yy is a self-correlation matrix of y, n q is residual noise, indicates that the first and second order statistical information on both sides of the equation is equal.

[0152] In the embodiment of the application, r is further simplified as:

[0153]

[0154] In formula (17), A = KΦ is a linear parameter, and z = Kv + n q is a Gaussian noise, and the correlation matrix of the Gaussian noise z is

[0155]

[0156] In formula (18), r represents a quantized vector signal, R represents a quantized signal, x is a vectorized real part of a joint variable X, v represents a vectorized real part of a complex Gaussian noise matrix V, and Φ is an observation matrix. is a Gaussian noise covariance, K T is a transpose matrix of the decomposition matrix K, K H is a conjugate matrix of the decomposition matrix K, is a self-correlation matrix of n q , R rr is a self-correlation matrix of r.

[0157] In step S504, a likelihood probability calculation is performed on the pilot vector signal to obtain a log-likelihood probability, and the log-likelihood probability is expressed as:

[0158]

[0159] In formula (19), r represents a quantized vector signal, A is a linear parameter, x is a vectorized real part of a joint variable X, T is the number of symbols contained in a pilot signal, ∑ is a correlation matrix of a Gaussian noise z, and M is the number of antennas on the base station side.

[0160] The steps S501 to S504 shown in the embodiments of the present application linearly approximate the quantization of the nonlinearity, so that the log-likelihood probability function is simplified as a simple quadratic function, which greatly reduces the complexity of the calculation compared with the traditional integral method according to the quantization result.

[0161] In some embodiments, the maximization of the posterior probability problem is expressed as:

[0162]

[0163] In formula (20), logp(r|x) represents the log-likelihood probability, logp(x) represents the log-probability density, and f(x) is the objective function.

[0164] In some embodiments, with reference to Figure 3 , the step S700 specifically includes but is not limited to steps S701 to S703:

[0165] In step S701, the variable of the objective function is subjected to maximum-minimum iteration to construct an upper bound of the objective function; wherein the upper bound of the objective function is expressed as:

[0166]

[0167] In formula (21), j is the iteration number, x is the vectorized real part of the joint variable X, x (j) is the iteration result of x, T is the number of symbols contained in the pilot signal, Ω2 is the quadratic coefficient matrix of the likelihood term, J is the approximation of the quadratic coefficient matrix of the likelihood term, f is the first-order residual of the likelihood term, Λ0 (j) , Λ1 (j) , W (j) are the prior quadratic term coefficients of the jth iteration, and g(x (j) ) is the residual function.

[0168] In step S702, the iteration result of the variable of the objective function is updated according to the upper bound of the objective function, to obtain a target iteration result; wherein the target iteration result is:

[0169]

[0170] In formula (22), f(x|x (j) ) represents the upper bound of the objective function, x (j) is the iteration result of x, Ω2 is the quadratic coefficient matrix of the likelihood term, J is the approximation of the quadratic coefficient matrix of the likelihood term, f is the first-order residual of the likelihood term, Λ0 (j) , Λ1 (j) , W (j) are the prior quadratic term coefficients of the jth iteration;

[0171] In step S703, user activity is deconstructed based on the target iteration result to obtain a user activity detection result, and user channel estimation is performed based on the target iteration result to obtain a user channel estimation result. The corresponding detection process of the user activity detection result is expressed as follows:

[0172]

[0173] In formula (23), is the conditional parameter, n is a positive integer between [1, N], for The i r =(n-1)M+1 elements, for The i r +MN elements, q s represents the prior probability that the user terminal device is active, a, b, and ε are all hyperparameters greater than zero, M is the number of antennas on the base station side, and N is the number of user terminal devices on the user side;

[0174] The user channel estimation result is:

[0175]

[0176] In formula (24), U R represents the array response channel matrix between the user side and the base station side, for The corresponding target joint variable, for The corresponding iterative diagonal matrix describes the activity status of the user terminal device.

[0177] In steps S701 to S703 shown in the embodiment of the present application, the upper bound of the objective function is constructed in the result of the previous iteration through the maximum minimization method. By solving the upper bound, the target solution of this iteration is obtained, and the results of user activity detection and user channel estimation are obtained based on the target solution.

[0178] An embodiment of the present application further provides an electronic device, including:

[0179] at least one memory;

[0180] at least one processor;

[0181] at least one program;

[0182] The program is stored in the memory, and the processor executes the at least one program to implement the user activity detection and user channel estimation method implemented in the present application. The electronic device can be an intelligent terminal such as a computer.

[0183] See also Figure 4, Figure 4 The hardware structure of an electronic device is illustrated, and the electronic device includes:

[0184] The processor can be implemented in the form of a general-purpose CPU (Central Processing Unit), a microprocessor, an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits, and is configured to execute related programs to implement the technical solutions provided by the embodiments of the present application.

[0185] The memory can be implemented in the form of a ROM (Read Only Memory), a static storage device, a dynamic storage device, or a RAM (Random Access Memory). The memory can store an operating system and other application programs. When the technical solutions provided by the embodiments of the present application are implemented by software or firmware, the related program codes are stored in the memory and are called and executed by the processor to implement the user activity detection and user channel estimation method of the embodiments of the present application.

[0186] The input / output interface is configured to realize information input and output.

[0187] The input / communication interface is configured to realize communication interaction between the device and other devices. The communication can be realized by a wired manner (for example, a USB, a network cable, etc.) or a wireless manner (for example, a mobile network, WIFI, Bluetooth, etc.).

[0188] The bus is configured to transmit information between various components (for example, the processor, the memory, the input / output interface, and the input / communication interface) of the device.

[0189] The processor, the memory, the input / output interface, and the input / communication interface are connected to each other through the bus to realize communication connection within the device.

[0190] The embodiments of the present application also provide a storage medium, which is a computer-readable storage medium and stores computer executable instructions. The computer executable instructions are configured to enable a computer to execute the user activity detection and user channel estimation method described above.

[0191] The embodiments described in the embodiments of the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that, with the evolution of technology and the appearance of new application scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0192] The embodiments of the present application are described in detail above with reference to the drawings, but the present application is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the spirit of the present application. In addition, the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

Claims

1. A method for user activity detection and user channel estimation, characterized in that: The method comprises: Establishing an mMTC uplink transmission system model; wherein the transmission system model includes a base station side and a user side, wherein the base station side and the user side communicate via a system channel, and the user side is configured to send a pilot signal to the base station side based on user activity; Constructing a priori probability model based on the transmission system model and user activity; Establishing a joint variable based on the user activity and the angular domain of the system channel; wherein the joint variable is used to characterize the hierarchical joint sparsity of the user activity and the system channel; Inputting the joint variable into the prior probability model to obtain the logarithmic probability density of the joint variable; Performing likelihood probability calculation based on the pilot signal and the joint variable to obtain a log-likelihood probability of the pilot signal with respect to the joint variable; Constructing a posterior probability maximization problem based on the log probability density and the log likelihood probability; Performing a maximum minimization solution on the maximum a posteriori probability problem to obtain a user activity detection result and a user channel estimation result; The base station side includes a A base station with a root antenna, the user side includes A single-antenna user terminal device; The user terminal device has the characteristic of occasional activity. The activity status of each of the user terminal devices is expressed as follows: in, for A positive integer between Indicates that the user terminal device is in active state, Indicates that the user terminal device is in an inactive state, and the user terminal device sends the pilot signal to the base station side when it is in an active state; The pilot signal includes The pilot sequence signal of symbols is expressed as: in, Indicates the The activity status of the user terminal device, For the The channels between the user terminal equipment and the base station side, is the power coefficient including the transmit power and channel path loss, Indicates the assignment to The pilot sequence of the user terminal equipment, is a complex Gaussian noise matrix with zero mean and covariance matrix, is the channel matrix between the user side and the base station side, is a diagonal matrix describing the activity of the user terminal device, is the large-scale fading matrix between the user side and the base station side, is the pilot sequence matrix on the user side; The angle domain channel matrix between the user side and the base station side is: in, is the angle domain channel matrix between the user side and the base station side, is the channel matrix between the user side and the base station side, represents an array response channel matrix between the user side and the base station side; Angle domain channel matrix No. Row, No. The column elements follow the following Gaussian distribution: in, for A positive integer between represents the angle domain channel matrix No. Row, No. Elements of the column, is the Gaussian distribution accuracy, Represents an N-dimensional complex vector space function; The probability density of is: in, is the gamma function, and are all hyperparameters greater than zero.

2. The method according to claim 1, characterized in that In the prior probability model, The active probability density of the user terminal devices is: in, represents the prior probability that the user terminal device is active, The value range of is (0,1), Indicates the The activity status of the user terminal device.

3. The method according to claim 2, characterized in that The joint variables are: ,in, is the angle domain channel matrix between the user side and the base station side, is a diagonal matrix describing the activity of the user terminal device; when =1, The user terminal device is in an active state, the joint variable No. Row, No. The distribution function of the column elements is: in, Represents a joint variable No. Row, No. Column elements, Indicates the The activity status of the user terminal device, represents the angle domain channel matrix No. Row, No. Elements of the column, is the Gaussian distribution accuracy; when =0, the The user terminal device is in an inactive state, is zero, at this time The distribution function of is: in, represents the impulse function, represents the N-dimensional complex vector space function, Represents a joint variable No. Row, No. Column elements, Indicates the The activity status of the user terminal device, represents the angle domain channel matrix No. Row, No. Elements of the column, is a hyperparameter greater than zero.

4. The method according to claim 3, characterized in that The functional formula of the logarithmic probability density is: in, For joint variables The vectorized real component of for No. elements, Indicates the The activity status of the user terminal device, represents the prior probability that the user terminal device is active, 、 and are all hyperparameters greater than zero, is the number of antennas on the base station side, is the number of user-side devices on the user side, for A positive integer between .

5. The method according to claim 4, characterized in that The step of performing likelihood probability calculation based on the pilot signal and the joint variable to obtain the log-likelihood probability of the pilot signal with respect to the joint variable specifically includes: Use general The scalar quantizer quantizes the pilot signal to obtain a quantized signal; wherein the quantized signal is expressed as: in, represents the quantized signal, represents the pilot signal; Perform vectorization and real number decomposition processing on the pilot signal to obtain a pilot vector signal; wherein the pilot vector signal is expressed as: in, represents the pilot vector signal, represents the pilot signal, For joint variables The vectorized real component of represents the complex Gaussian noise matrix The vectorized real component of is the observation matrix; Vectorize the quantized signal to obtain a quantized vector signal; wherein the quantized vector signal is expressed as: in, represents the quantized vector signal, represents the quantized signal, For joint variables The vectorized real component of represents the complex Gaussian noise matrix The vectorized real component of is the observation matrix, quantized vector signal The linear approximation is: in, is the decomposition matrix, for and The cross-correlation matrix, for The autocorrelation matrix of is the residual noise, means that the first-order and second-order statistics on both sides of the equation are equal, Further simplified to: in, is a linear parameter, is Gaussian noise, Gaussian noise The correlation matrix is in, is the Gaussian noise covariance, Decomposition matrix The transposed matrix of Decomposition matrix The conjugate matrix of for The autocorrelation matrix of for The autocorrelation matrix of Performing a likelihood probability calculation on the pilot vector signal to obtain the log-likelihood probability; wherein the log-likelihood probability is expressed as: in, represents the quantized vector signal, is a linear parameter, For joint variables The vectorized real component of is the number of symbols contained in the pilot signal, Gaussian noise The correlation matrix of is the number of antennas on the base station side.

6. The method according to claim 5, characterized in that The problem of maximizing the posterior probability is expressed as: in, represents the log-likelihood probability, represents the logarithmic probability density, is the objective function.

7. The method according to claim 6, characterized in that The step of performing a maximum minimization solution on the maximum a posteriori probability problem to obtain a user activity detection result and a user channel estimation result specifically includes: Performing maximum minimization iterations on the variables of the objective function to construct an upper bound of the objective function; wherein the upper bound of the objective function is expressed as: in, is the number of iterations, For joint variables The vectorized real component of for No. The result of the iterations, is the number of symbols contained in the pilot signal, is the likelihood term quadratic coefficient matrix, is an approximation of the quadratic coefficient matrix of the likelihood term, is the first-order residual of the likelihood term, 、 、 All are The coefficient of the prior quadratic term for the iteration, is the residual function; The iteration result of the objective function variable is updated according to the upper bound of the objective function to obtain a target iteration result; wherein the target iteration result is: in, represents the upper bound of the objective function, for No. The result of the iterations, is the likelihood term quadratic coefficient matrix, is an approximation of the quadratic coefficient matrix of the likelihood term, is the first-order residual of the likelihood term, 、 、 All are The coefficient of the prior quadratic term at iterations; The user activity situation is deconstructed according to the target iteration result to obtain the user activity detection result, and the user channel is estimated according to the target iteration result to obtain the user channel estimation result. The corresponding detection process of the user activity detection result is expressed as follows: in, is the conditional parameter, for A positive integer between for No. elements, for No. elements, represents the prior probability that the user terminal device is active, 、 and are all hyperparameters greater than zero, is the number of antennas on the base station side, The number of user-side devices on the user side; The user channel estimation result is: in, represents the array response channel matrix between the user side and the base station side, for The corresponding target joint variable, for A corresponding iterative diagonal matrix describing the activity status of the user terminal device.

8. An electronic device, characterized in that: include: at least one memory; at least one processor; at least one program; The program is stored in the memory, and the processor executes at least one of the programs to implement: The method according to any one of claims 1 to 7.

9. A storage medium, wherein the storage medium is a computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to cause a computer to execute: The method according to any one of claims 1 to 7.

Images