Three-dimensional millimeter wave time-varying channel estimation method and device

Abstract

The invention discloses a three-dimensional millimeter wave time-varying channel estimation method and device, and belongs to the field of wireless communication, and the method comprises the steps: building a millimeter wave communication system containing array motion; a three-dimensional time-varying channel model considering the Doppler effect is established, composite signals, caused by array motion, in a three-dimensional time-varying channel are analyzed, each column of a direction matrix after synthesis is consistent with the array manifold of a planar array, and it is indicated that a moving linear array can form a virtual planar array; sparse representation of a three-dimensional time-varying channel is realized through linear array motion; a three-dimensional time-varying channel is recovered based on an iterative meshless weighting algorithm of array motion; comprising AoAs estimation and path gain estimation. According to the method, the negative influence of Doppler frequency shift in the time-varying channel on channel estimation is overcome, the virtual array element is constructed, channel estimation of the three-dimensional millimeter wave time-varying channel is realized through linear array motion, the channel estimation precision is improved, and the pilot frequency overhead and the algorithm complexity are greatly reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communications, and in particular relates to a three-dimensional millimeter wave time-varying channel estimation method and device. Background Art

[0002] Qibo Qin, Lin Gui, et al. published "Time-Varying Channel Estimation for Millimeter Wave Multiuser MIMO Systems" in IEEE Transactions on Vehicular Technology in 2018, focusing on the channel estimation problem of time-varying millimeter wave channels in mobile scenarios. Different from the channel estimation problem of traditional static channels, this paper takes into account the Doppler effect caused by user movement. By taking advantage of the fact that the angles of arrival (AoAs) and angles of departure (AoDs) on time-varying channels usually change slower than the path gain, a new transmission frame structure is proposed in which the channel estimation is decomposed into two independent stages. AoAs / AoDs estimation is first performed in the first stage, and then the path gain estimation is performed in the second stage. Specifically, in the AoAs / AoDs estimation stage, the discrete prolate spheroidal basis expansion model (DPS-BEM) is used to transform the AoAs / AoDs estimation into a block sparse recovery problem, and then an adaptive angle estimation algorithm (AAE) is proposed to estimate the AoAs / AoDs. In the path gain estimation stage, the AoAs / AoDs estimated in the previous stage are used to further estimate the path gain. The method proposed in this paper takes into account the influence of Doppler frequency shift under time-varying channels. The proposed scheme is suitable for channel estimation problems in non-static scenarios, providing a possibility for high-speed data transmission in high-speed mobile scenarios in the future. However, due to the limitation of the antenna aperture dimension, the method proposed in this paper can only use linear arrays to realize channel estimation of two-dimensional channels. If you want to perform channel estimation in a three-dimensional channel that is closer to reality, you need to use a more complex planar array, which undoubtedly increases the cost and computational complexity. In addition, although the DPS-BEM method can be used to capture the changes of the channel over time, there is still a certain error compared with the actual situation, and the relative channel estimation results have room for further improvement. Summary of the invention

[0003] The purpose of the present invention is to provide a three-dimensional millimeter wave time-varying channel estimation method and device. Compared with the prior art, the present invention realizes the channel estimation of the three-dimensional channel by the linear array by utilizing the improvement of the array aperture dimension brought about by the linear array movement, thereby reducing the calculation cost and complexity. The present invention also overcomes the negative impact of Doppler frequency shift on channel estimation, significantly improves the accuracy of channel estimation, and reduces pilot overhead.

[0004] The technical solution adopted by the present invention to solve the technical problem is as follows:

[0005] The present invention provides a three-dimensional millimeter wave time-varying channel estimation method, comprising the following steps:

[0006] Step 1: Establish a millimeter wave communication system with array motion;

[0007] Step 2: Establish a three-dimensional time-varying channel model considering the Doppler effect, and analyze the synthetic signal in the three-dimensional time-varying channel caused by the array motion. After synthesis, each column of the directional matrix is ​​consistent with the array manifold of the planar array, indicating that the moving linear array can form a virtual planar array.

[0008] Step 3: Achieving sparse representation of three-dimensional time-varying channels through linear array motion;

[0009] The set of quantized AoAs is defined as:

[0010]

[0011] in, and represents the quantized μ and γ, G μ and G γ They represent the number of grid points for quantization of μ and γ, μ=[μ1,...,μ L ] and γ=[γ1,...,γ L ] represent the transformation parameters μ of all paths respectively l and γ l The set of μ max and μ min Represent the maximum and minimum values ​​in μ, γ max and γ min Represent the maximum and minimum values ​​in γ respectively, and introduce deviation and To narrow the gap between the quantized AoAs set and the true angle;

[0012] The received signal after the synthesis of M consecutive pilot signals is expressed as:

[0013]

[0014] Where D(t) represents the synthesized channel combination matrix, represents the measurement matrix, M A Indicates the number of pilot signals in the AoAs estimation phase, N RF represents the number of array receiver RF chains, represents the direction matrix consisting of a set of quantized AoAs, represents the path gain corresponding to the quantized AoAs set;

[0015] Step 4: Using an iterative meshless weighted algorithm based on array motion to recover the three-dimensional time-varying channel;

[0016] S4.1: AoAs estimation;

[0017] Assume that the offset is 0, that is, Δμ = 0, Δγ = 0; use the orthogonal matching pursuit algorithm to solve the AoAs. After each iteration of the OMP algorithm, the selected index will be added to the support set S. The selected AoAs are S l represents the lth element in the support set S; then use the maximum likelihood method to estimate and The deviation from the true angle is Δμ S and Δγ S ; Set the iterative learning rate, calculate the likelihood function of each iteration, and get the deviation after n iterations and Final estimates of AoAs and It is expressed as:

[0018]

[0019] S4.2: Path gain estimation;

[0020] Use the least squares method to estimate the path gain:

[0021]

[0022] The estimated value of the final channel combining matrix is ​​expressed as:

[0023]

[0024] in, Represents the estimated value of the synthesized direction matrix, y syn (t) represents the received signal after M consecutive pilot signals are synthesized, N represents the number of array receiver antenna elements, L represents the number of multipath components, and m represents the sampling time.

[0025] Furthermore, in the millimeter wave communication system, the linear array is mounted on a moving object, the moving speed of the object is v, the angle between the array and the moving direction of the object is α, and the downlink signal from the single antenna base station of the roadside unit is received.

[0026] Furthermore, in the millimeter wave communication system, it is assumed that the array receiver contains N antenna elements and N RF RF link; the signal received at time t is expressed as:

[0027] y(t)=W H (t)h(t)x(t)+W H (t)n(t)

[0028] Wherein, the superscript H represents the conjugate transpose of the matrix; Represents the channel combination matrix, and the modulus value of each element in the matrix is ​​a constant represents the channel combination matrix; x(t) represents the pilot signal sent at time t; means that the mean is zero and the variance is σ 2 Additive white Gaussian noise vector.

[0029] Furthermore, in step 2, assuming that there are L multipath components in total, the channel combination matrix h(t) is expressed as:

[0030]

[0031] Among them, g l represents the channel gain of the lth path, ν l represents the Doppler frequency shift caused by relative motion, a(θ l ,φ l ) represents the steering vector of the receiving antenna, θ l and φ l Respectively represent the azimuth and elevation angles of the incident signal of the lth path; for a three-dimensional channel, the steering vector of the receiving antenna is expressed as:

[0032]

[0033] Among them, f l represents the frequency of the signal with Doppler shift in the first path, c represents the speed of light, d i (i=1, 2, ..., N-1) represents the distance between the (i+1)th antenna array element and the 1st antenna array element.

[0034] Furthermore, in step S4.1, the maximum likelihood method is used to estimate and The deviation from the true angle, Δμ S and Δγ S , expressed as:

[0035]

[0036] in, represents the estimated path gain, Represents the estimated value of the synthesized direction matrix.

[0037] Furthermore, in step S4.1, the likelihood function is expressed as:

[0038]

[0039] in, g represents the estimated value of the synthesized direction matrix, Tr represents the trace of the matrix, Indicates the synthesized received signal, with the superscript Represents the pseudoinverse of a matrix.

[0040] Furthermore, in step S4.1, during each iteration, it is assumed that the learning rates of the nth iteration are and Then the deviation obtained in the nth iteration is and They are:

[0041]

[0042] in, and They represent the likelihood function with respect to Δμ S and Δγ S Find the gradient.

[0043] The present invention also provides a three-dimensional millimeter wave time-varying channel estimation device, comprising: a memory and a processor; the memory stores executable instructions, and the processor is configured to execute the executable instructions in the memory to implement the steps of the three-dimensional millimeter wave time-varying channel estimation method.

[0044] The beneficial effects of the present invention are:

[0045] The present invention proposes a three-dimensional millimeter wave time-varying channel estimation method and device, which can use a linear array to achieve sparse representation and sparse recovery of a three-dimensional time-varying millimeter wave channel. Specifically, through an in-depth analysis of the synthetic signal in the time-varying channel caused by the array motion, it is found that a virtual plane array can be formed by a moving linear array. Based on this, the present invention proposes a sparse representation method of a three-dimensional channel achieved by linear array motion. In order to make full use of this sparsity, the present invention further proposes an efficient sparse channel recovery algorithm, which is divided into two stages: first, the angles of arrival (angles of arrival, AoAs) are estimated using a grid point and non-grid point based parameter estimation method, and then the path gain is estimated using the least squares method based on the estimated AoAs. The present invention constructs a virtual array element by utilizing the improvement of the array aperture dimension brought about by the linear array motion, and realizes the channel estimation of the three-dimensional channel by the linear array. The present invention also greatly improves the channel estimation accuracy by overcoming the negative impact of Doppler frequency shift on channel estimation in the time-varying channel; the simple array structure coupled with the staged channel estimation scheme greatly reduces the pilot overhead while also reducing the complexity of the algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 Schematic diagram of linear array motion (3D diagram).

[0047] Figure 2 Schematic diagram of linear array motion (top view).

[0048] Figure 3 The figure is a schematic diagram of a millimeter wave communication system with array motion (taking a vehicle motion scenario as an example).

[0049] Figure 4 The flowchart of the array motion-based iterative gridless weighted (AM-IGW) algorithm is shown in FIG. DETAILED DESCRIPTION

[0050] The technical problem to be solved by the present invention is how to use a linear array to achieve high-precision, low-complexity, and low-pilot-overhead three-dimensional channel estimation in a time-varying channel with rapid channel changes caused by array motion. The present invention first proposes a sparse representation method for a three-dimensional time-varying millimeter-wave channel through linear array motion, and further proposes an efficient sparse channel recovery algorithm for channel estimation, which mainly includes the following steps:

[0051] Step 1: Establish a millimeter wave communication system with array motion;

[0052] Step 2: Establish a three-dimensional time-varying channel model that takes into account the Doppler effect, and analyze the synthetic signal in the three-dimensional time-varying channel caused by the array motion. After synthesis, each column of the directional matrix is ​​consistent with the array manifold of the planar array, indicating that the moving linear array can form a virtual planar array.

[0053] Step 3: Achieving sparse representation of three-dimensional time-varying channels through linear array motion;

[0054] Step 4: Recover the three-dimensional time-varying channel using an iterative meshless weighted algorithm based on array motion.

[0055] The present invention also provides a three-dimensional millimeter wave time-varying channel estimation device realized by linear array movement, which mainly includes: a memory and a processor; wherein the memory stores executable instructions, and the processor is configured to execute the executable instructions in the memory to implement the steps of a three-dimensional millimeter wave time-varying channel estimation method of the present invention. Specific implementation method 1

[0057] In this embodiment, the movement mode of the array is specifically linear movement, so the scenario of the linear array performing linear movement is taken as a preferred embodiment of the present invention.

[0058] This embodiment provides a three-dimensional millimeter wave time-varying channel estimation method, and its specific implementation process is as follows:

[0059] (1) Establish a millimeter wave communication system with array motion;

[0060] like Figure 1 , Figure 2 and Figure 3 As shown in Figure 1, consider a millimeter wave communication system with array motion. The linear array is mounted on an object moving in a straight line, and the object's speed is v. The angle between the linear array and the object's moving direction is α, and it receives the downlink signal from the single-antenna base station of the roadside unit.

[0061] Assume that the linear array receiver has N antenna elements and N RF RF link. The signal received at time t can be expressed as:

[0062] y(t)=W H (t)h(t)x(t)+W H (t)n(t)

[0063] The superscript H represents the conjugate transpose of the matrix. Represents the channel combination matrix, and the modulus value of each element in the matrix is ​​a constant represents the channel combination matrix; x(t) represents the pilot signal sent at time t. For the convenience of subsequent discussion, it is assumed that x(t) = 1; means that the mean is zero and the variance is σ 2 Additive white Gaussian noise vector.

[0064] (2) Establish a three-dimensional time-varying channel model that takes into account the Doppler effect and conduct an in-depth analysis of the synthetic signal in the three-dimensional time-varying channel caused by array motion;

[0065] In order to better reflect the actual situation, the present invention adopts a three-dimensional time-varying channel model that takes into account the Doppler effect. Assuming that there are L multipath components in total, the channel combination matrix h(t) can be expressed as:

[0066]

[0067] Among them, g l represents the channel gain of the lth path, ν l represents the Doppler frequency shift caused by relative motion, a(θ l ,φ l ) represents the steering vector of the receiving antenna, θ l and φ l Respectively represent the azimuth and elevation angles of the incident signal of the lth path. For a three-dimensional channel, the steering vector of the receiving antenna can be expressed as:

[0068]

[0069] Among them, f l represents the frequency of the signal with Doppler shift in the first path, c represents the speed of light, d i (i=1, 2, ..., N-1) represents the distance between the (i+1)th antenna array element and the 1st antenna array element.

[0070] In order to simplify the expression, the transformation parameter μ is defined l =cos(θ l )sin(φ l ), γ l = sin(θ l )sin(φ l ), p l =μ l cos(α)+γ l sin(α). Under this definition, the signal frequency containing the Doppler shift in represents the carrier frequency, and λ represents the carrier wavelength. The nth element in the receiving antenna steering vector can be expressed as:

[0071]

[0072] Because, v<<c, |p l |≤2,|μl |≤1, m' represents a constant near 1, so The steering vector of the receiving antenna can be further expressed as:

[0073]

[0074] Consider the mth sampling moment after time t, and combine the Doppler frequency shift The channel matrix at this sampling time can be expressed as:

[0075]

[0076] Where τ represents the sampling period, d α =vτ represents the distance that the object carrying the linear array moves in one sampling period, represents the path gain including Doppler shift, represents the direction matrix at time m, μ=[μ1,...,μ L ] and γ=[γ1,...,γ L ] represent the transformation parameters μ of all paths respectively l and γ l A collection of .

[0077] Therefore, the signal received at time t+mτ can be expressed as:

[0078]

[0079] Considering that the received signal after the synthesis of M consecutive pilot signals can be expressed as:

[0080]

[0081] Among them, the synthesized direction matrix is ​​expressed as The synthesized channel combination matrix is ​​expressed as The synthesized noise vector is expressed as

[0082] At this time, the synthesized direction matrix A syn Any column in (μ,γ) can be expressed as:

[0083]

[0084] in, It can be seen from the above formula that each column of the synthesized direction matrix is ​​consistent with the array manifold of the planar array, which shows that the moving linear array forms a virtual planar array.

[0085] (3) Sparse representation of three-dimensional time-varying channels through linear array motion;

[0086] In order to demonstrate the sparsity of the above three-dimensional time-varying channel model, the quantized AoAs set is defined as:

[0087]

[0088] in, and represents the quantized μ and γ, μ max and μ min Represent the maximum and minimum values ​​in μ, γ max and γ min Represent the maximum and minimum values ​​in γ, G μ and G γ denote the number of grid points for quantization of μ and γ respectively. However, in practical applications, the real AoAs are continuous and cannot completely match the quantized AoAs set. Therefore, it is necessary to introduce a deviation and To narrow the gap between the quantized AoAs set and the true angles.

[0089] At this point, the synthesized received signal can be further expressed as:

[0090]

[0091] in, represents the measurement matrix, M A represents the number of pilot signals in the AoAs estimation phase, represents the direction matrix consisting of a set of quantized AoAs, represents the path gain corresponding to the quantized AoAs set. In this case, ||β(t)||0=L, L<<G μ G γ , which means that the sparsity of the above formula is L, and the sparse signal recovery method can be used to estimate the parameters and thus perform channel estimation.

[0092] (4) AoAs estimation and path gain estimation;

[0093] In order to obtain higher accuracy while reducing pilot overhead and computational complexity, the present invention divides channel estimation into two stages: AoAs estimation and path gain estimation. The two stages use M A and M P Based on this, the present invention proposes an array motion-based iterative gridless weighted (AM-IGW) algorithm to recover the three-dimensional time-varying channel. The specific implementation process is as follows: Figure 4 shown.

[0094] Considering that it is more complicated to estimate AoAs directly, we first assume that the offset is 0, that is, Δμ = 0, Δγ = 0. In this case, the AoAs estimation problem can be transformed into a standard grid-based sparse signal recovery problem, and the parameters are estimated using a grid-based parameter estimation method. Specifically, the orthogonal matching pursuit (OMP) algorithm can be used to solve it. After each iteration of the OMP algorithm, the selected index will be added to the support set S, and the selected AoAs will be Where S l Represents the lth element in the support set S.

[0095] However, in practice, the predefined quantization set does not match the actual angle. In order to further improve the accuracy of parameter estimation, it is necessary to use a non-grid parameter estimation method to estimate the deviation. Specifically, the maximum likelihood method can be used to estimate and The deviation from the true angle, Δμ S and Δγ S , which can be specifically expressed as:

[0096]

[0097] in, represents the estimated path gain, represents the estimated value of the synthesized direction matrix. After some processing, the likelihood function can be expressed as:

[0098]

[0099] Where Tr represents the trace of the matrix, Indicates the synthesized received signal, superscript) Represents the pseudoinverse of a matrix.

[0100] In each iteration, assuming that the learning rates of the nth iteration are and Then the deviation obtained in the nth iteration is and They are:

[0101]

[0102] in, and They represent the likelihood function with respect to Δμ S and Δγ STo ensure the stability and efficiency of the optimization process, the present invention also uses a backtracking algorithm to dynamically adjust the learning rate of each iteration. and

[0103] Ultimately, a more precise estimate of AoAs and It can be expressed as:

[0104]

[0105] In addition, the estimated values ​​of the heading and elevation angles can be expressed as:

[0106]

[0107] In the path gain estimation stage, the least squares method is used to estimate the path gain:

[0108]

[0109] The estimated value of the final channel combining matrix can be expressed as:

[0110]

[0111] in, Represents the estimated value of the synthesized direction matrix.

[0112] The present invention fully considers the improvement of array aperture dimension brought about by array motion and the influence of Doppler frequency shift on channel estimation in time-varying channels, and proposes a channel estimation method with high estimation accuracy, low pilot overhead and low complexity, which realizes channel estimation of three-dimensional channels using linear array motion under time-varying channels.

[0113] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A three-dimensional millimeter wave time-varying channel estimation method, characterized in that: The following steps are involved: Step 1: Establish a millimeter wave communication system with array motion; Step 2: Establish a three-dimensional time-varying channel model considering the Doppler effect, and analyze the synthetic signal in the three-dimensional time-varying channel caused by the array motion. After synthesis, each column of the directional matrix is ​​consistent with the array manifold of the planar array, indicating that the moving linear array can form a virtual planar array. Step 3: Achieving sparse representation of three-dimensional time-varying channels through linear array motion; The set of quantized AoAs is defined as: in, and represents the quantized μ and γ, G μ and G γ They represent the number of grid points for quantization of μ and γ, μ=[μ1,...,μ L ] and γ=[γ1,...,γ L ] represent the transformation parameters μ of all paths respectively l and γ l The set of μ max and μ min Represent the maximum and minimum values ​​in μ, γ max and γ min Represent the maximum and minimum values ​​in γ respectively; at the same time, introduce the deviation and To narrow the gap between the quantized AoAs set and the true angle; The received signal after the synthesis of M consecutive pilot signals is expressed as: Where D(t) represents the synthesized channel combination matrix, represents the measurement matrix, M A Indicates the number of pilot signals in the AoAs estimation phase, N RF represents the number of array receiver RF chains, represents the direction matrix consisting of a set of quantized AoAs, represents the path gain corresponding to the quantized AoAs set; Step 4: Using an iterative meshless weighted algorithm based on array motion to recover the three-dimensional time-varying channel; S4.1: AoAs estimation; Assume that the offset is 0, that is, Δμ = 0, Δγ = 0; use the orthogonal matching pursuit algorithm to solve the AoAs. After each iteration of the OMP algorithm, the selected index will be added to the support set S. The selected AoAs are S l represents the lth element in the support set S; then use the maximum likelihood method to estimate and The deviation from the true angle is Δμ S and Δγ S ; Set the iterative learning rate, calculate the likelihood function of each iteration, and get the deviation after n iterations and Final estimates of AoAs and It is expressed as: S4.2: Path gain estimation; Use the least squares method to estimate the path gain: The estimated value of the final channel combining matrix is ​​expressed as: in, Represents the estimated value of the synthesized direction matrix, y syn (t) represents the received signal after the synthesis of M consecutive pilot signals, N represents the number of array receiver antenna elements, L represents the number of multipath components, and m represents the sampling time.

2. A three-dimensional millimeter wave time-varying channel estimation method according to claim 1, characterized in that: In the millimeter wave communication system, the linear array is mounted on a moving object, the moving speed of the object is v, the angle between the array and the moving direction of the object is α, and the downlink signal from the single antenna base station of the roadside unit is received.

3. A three-dimensional millimeter wave time-varying channel estimation method according to claim 1, characterized in that: In the millimeter wave communication system, it is assumed that the array receiver contains N antenna elements and N RF RF link; the signal received at time t is expressed as: y(t)=W H (t)h(t)x(t)+W H (t)n(t) Wherein, the superscript H represents the conjugate transpose of the matrix; Represents the channel combination matrix, the modulus value of each element in the matrix is ​​a constant represents the channel combination matrix; x(t) represents the pilot signal sent at time t; means that the mean is zero and the variance is σ 2 Additive white Gaussian noise vector.

4. A three-dimensional millimeter wave time-varying channel estimation method according to claim 1, characterized in that: In step 2, assuming that there are L multipath components in total, the channel combination matrix h(t) is expressed as: Among them, g l represents the channel gain of the lth path, ν l represents the Doppler frequency shift caused by relative motion, a(θ l ,φ l ) represents the steering vector of the receiving antenna, θ l and φ l Respectively represent the azimuth and elevation angles of the incident signal of the lth path; for a three-dimensional channel, the steering vector of the receiving antenna is expressed as: Among them, f l represents the frequency of the signal with Doppler shift in the first path, c represents the speed of light, d i (i=1, 2, ..., N-1) represents the distance between the (i+1)th antenna array element and the 1st antenna array element.

5. The three-dimensional millimeter wave time-varying channel estimation method according to claim 1, characterized in that: In step S4.1, the maximum likelihood method is used to estimate and The deviation from the true angle, Δμ S and Δγ S , expressed as: in, represents the estimated path gain, Represents the estimated value of the synthesized direction matrix.

6. A three-dimensional millimeter wave time-varying channel estimation method according to claim 1, characterized in that: In step S4.1, the likelihood function is expressed as: in, g represents the estimated value of the synthesized direction matrix, Tr represents the trace of the matrix, Indicates the synthesized received signal, with the superscript Represents the pseudoinverse of a matrix.

7. The three-dimensional millimeter wave time-varying channel estimation method according to claim 1, characterized in that: In step S4.1, during each iteration, assume that the learning rates for the nth iteration are and Then the deviation obtained in the nth iteration is and They are: in, and They represent the likelihood function with respect to Δμ S and Δγ S Find the gradient.

8. A three-dimensional millimeter wave time-varying channel estimation device, characterized in that: include: Memory and processor; The memory stores executable instructions, and the processor is configured to execute the executable instructions in the memory to implement the steps of a three-dimensional millimeter wave time-varying channel estimation method as described in any one of claims 1 to 7.

Images