Dynamic target beam tracking method based on coverage error range and adaptive codebook

Abstract

The invention provides a dynamic target beam tracking method based on a coverage error range and an adaptive codebook. The method comprises the following steps: constructing a beam tracking scene for a dynamic target; the base station calculates the maximum error range of the dynamic target trajectory in the current time period; the base station adaptively adjusts the beam width and calculates the optimal beam width and the optimal beam orientation; the base station obtains the optimal code word; and the base station obtains a beam tracking result of the dynamic target. According to the invention, the base station calculates the beam orientation by using the trajectory prediction information, the influence of the trajectory prediction error on the beam tracking precision is considered, the generated beam width covers the error range, the influence on the tracking precision due to the increase of the training frequency is avoided, and since the base station adaptively adjusts the beam with the purpose of minimizing the beam switching overhead, the tracking precision is improved. The beam switching frequency is reduced, the problems of high system overhead and long time delay in the prior art are solved, and the real-time performance of beam tracking is further improved.

Description

Technical Field

[0001] The invention belongs to the technical field of beam tracking, and relates to a dynamic target beam tracking method based on a coverage error range and an adaptive codebook, which can be used for a base station to perform real-time beam alignment on a dynamic target and improve communication quality. Background Art

[0002] Beam tracking is a technology that adjusts the radiation direction and intensity of the directional beam generated by the antenna according to the position and direction of the target. It is often used for dynamic targets, such as drone communications. In traditional wireless communications, the position and direction of the moving target are constantly changing, which will cause signal attenuation and interference. In order to overcome this challenge, when transmitting signals, the transmitter and receiver use multiple antennas to work together, which can concentrate the energy of the transmitted signal in a certain direction to form a beam with a narrow beam width and a large gain. It can effectively compensate for the significant path loss of the beam generated by the base station and improve the transmission efficiency and quality of the signal. However, due to the use of narrow beams, the movement of the dynamic target causes its position to deviate from the direction of the beam generated by the initial base station, and the quality of the received wireless signal will be greatly reduced. Therefore, in practical applications, it is necessary to introduce beam tracking technology to track dynamic targets in real time to ensure that the transmitting beam is always pointing to the target user.

[0003] The realization of beam tracking is inseparable from beam training. Traditional fixed-resolution codebooks used for beam training in beam tracking, such as N-phase codebooks or discrete Fourier transform DFT codebooks, usually adopt an exhaustive method, which is implemented by using each codeword in the codebook to communicate with the target and selecting the codeword with the highest signal-to-noise ratio (SNR). The disadvantage is that the training overhead and time consumption are high. At the same time, when the speed of a dynamic target is too fast, the movement time of the target within a beam range is shortened, which means that the proportion of training time will increase. Narrower beam ranges under high gain requirements will also lead to the problem of further increase in the proportion of training time. In addition to using codebooks, a common method is to use technologies such as Kalman filters to predict the target trajectory and use the prediction information to directly point the beam to the target direction. However, when using this beam tracking technology, the beam direction emitted by the base station will be directly affected by the trajectory prediction error. The erroneous position information caused by the error will generate an offset beam, and the beam tracking process will eventually fail. In beam tracking technology, tracking accuracy can be used to judge the quality of tracking results. Usually, the communication rate when a dynamic target communicates with a base station can be used to measure the tracking accuracy. The higher the communication rate, the higher the tracking accuracy, and vice versa.

[0004] At present, there are many beam tracking methods for dynamic targets. For example, Jiawei Chen et al. proposed a beam tracking method for dynamic targets in “Hierarchical Codebook Design for Near-Field mmWave MIMO Communications Systems” (IEEE Wireless Communications Letters, VOL. 12, Issue: 11, November 2011). 2023), a beam tracking method based on hierarchical codebook training is proposed, and its main steps are: (1) the base station calculates the lower codebook according to a steering beam gain approximation method to maximize the gain of beam tracking; (2) the base station adopts the beam rotation and beam repositioning method to calculate the appropriate position of the beam covered by the upper codebook; (3) the base station sends a training codeword sequence for beam training, records the training delay, selects the codeword that makes the received signal power the highest, and obtains the optimal transmit beam. This method can obtain a higher average beam tracking gain and a higher minimum beam tracking gain. Compared with the exhaustive search, the hierarchical codebook can greatly reduce the steps of beam training. However, its disadvantage is that this method will shorten the beam coherence time and increase the training frequency when the target moves at a high speed, which will reduce the communication rate of dynamic target communication and thus affect the beam tracking accuracy. At the same time, this method needs to search and train the codewords in the codebook each time it is trained, which increases the system delay and training overhead, further affecting the tracking accuracy of beam tracking. Summary of the invention

[0005] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and propose a dynamic target beam tracking method based on coverage error range and adaptive codebook to solve the technical problem of low tracking accuracy in the prior art.

[0006] To achieve the above object, the technical solution adopted by the present invention includes the following steps:

[0007] (1) Construct a beam tracking scenario for dynamic targets;

[0008] Construct a beam tracking scenario including a uniform linear array base station with a total number of N antennas located at the origin of a two-dimensional coordinate system and a dynamic target distributed in a two-dimensional coordinate system for wireless communication, wherein N ≥ 2;

[0009] (2) The base station calculates the maximum error range of the dynamic target trajectory in the current time period;

[0010] The base station predicts the trajectory S in the current time period based on the positioning information of the dynamic target in the historical time period, and uses the angle θ between the points closer to and farther from the base station in the starting position and the end position in S to reach the base station normal1 ,θ 2 Calculate the maximum error range e of S max ;

[0011] (3) The base station adaptively adjusts the beam width and calculates the optimal beam width and optimal beam direction;

[0012] The base station adaptively adjusts the beam width to minimize the beam switching overhead and obtains the theoretical optimal beam width Φ -3dB , and through Φ -3dB and the maximum error range e max Calculate the optimal beam width φ′ -3dB and the optimal beam heading ψ;

[0013] (4) The base station obtains the optimal codeword:

[0014] The base station determines the communication rate R of the starting position of the predicted trajectory S St The communication rate threshold R Th Does it satisfy R St >R Th , if so, through the optimal beam width φ′ -3dB and the optimal beam direction ψ to calculate the optimal codeword W opt Otherwise, the optimal beam direction ψ is adaptively adjusted to obtain the optimal codeword W opt ;

[0015] (5) The base station obtains the beam tracking results of the dynamic target;

[0016] The base station uses the optimal codeword W opt Communicate with the dynamic target and determine the transmission rate R of the dynamic target communicating with the base station Rt Does it satisfy R Rt >R Th , if so, the dynamic target is within the beam range of the current time period, otherwise execute step (2).

[0017] Compared with the prior art, the present invention has the following advantages:

[0018] (1) In the present invention, the base station calculates the beam direction by using trajectory prediction information and takes into account the influence of trajectory prediction error on beam tracking accuracy. The generated beam width covers the error range, thereby maximizing the tracking accuracy in the current scenario. When the dynamic target moves at a high speed, there is no need to retrain each time, thereby avoiding the influence of a sharp increase in training frequency on tracking accuracy when the dynamic target moves at a high speed. The communication rate of the dynamic target can be increased, and the beam tracking accuracy is effectively improved.

[0019] (2) The present invention aims to minimize the beam switching overhead when the base station adaptively adjusts the beam, thereby reducing the frequency of beam switching and applying all the time consumed by the system to data transmission. This overcomes the problems of high system overhead and long delay in the prior art, thereby improving the real-time performance of beam tracking. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart for realizing the present invention;

[0021] Figure 2 The figure is a comparison chart of simulation results of tracking accuracy between the present invention and the prior art. DETAILED DESCRIPTION

[0022] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0023] Reference Figure 1 , the present invention comprises the following steps:

[0024] Step 1), constructing a beam tracking scenario for a dynamic target;

[0025] A beam tracking scenario is constructed, including a uniform linear array base station with a total number of N antennas located at the origin of the two-dimensional coordinate system and a dynamic target distributed in the two-dimensional coordinate system that realizes wireless communication with the base station. The dynamic target is distributed in the two-dimensional space and can communicate with the base station. The dynamic target moves irregularly in the two-dimensional space, and the base station adaptively adjusts the beam to track it to ensure the communication quality of the dynamic target. When the base station transmits a beam, the more the beam deviates from the dynamic target, the lower the communication rate between the base station and the dynamic target. Therefore, the communication rate can be used to judge the beam tracking accuracy of the base station for the dynamic target. In this embodiment, N=32.

[0026] Step 2), the base station calculates the maximum error range of the dynamic target trajectory in the current time period;

[0027] The base station predicts the trajectory S in the current time period based on the positioning information of the dynamic target in the historical time period. The trajectory prediction method includes a Kalman filter, a BP neural network or a least squares fitting method. The least squares method is used in this embodiment, and the trajectory prediction algorithm error can be controlled within 0.5 meters. The historical time period positioning information of the dynamic target is collected into two-dimensional discrete trajectory points, and the discrete trajectory points are converted into a Gaussian plane discrete coordinate set by Gaussian projection method, and the trajectory prediction result S is obtained by fitting.

[0028] The angle θ between the base station's normal line and the point closer to the base station and the point farther from the base station in S is 1 ,θ 2 Calculate the maximum error range e of S max , the calculation formula is:

[0029] e max =e 1 +e 2

[0030]

[0031] Among them, e 1 、e 2 They represent the maximum prediction errors of the starting position and the end position in the predicted trajectory S, respectively. Denote the arrival angle as θ 1 and θ 2 The standard deviation of the trajectory prediction error of the target, K represents the number of subcarriers that can be transmitted by each antenna of the base station, in this example K = 30, ρ represents the signal-to-noise ratio when the beam generated by the base station communicates with the dynamic target, F represents the number of samples of the dynamic target when the base station performs trajectory prediction, in this embodiment F = 1000;

[0032] Step 3), the base station adaptively adjusts the beam width and calculates the optimal beam width and optimal beam direction;

[0033] The base station adaptively adjusts the beam width to minimize the beam switching overhead and obtains the theoretical optimal beam width Φ -3dB , in order to minimize the overhead caused by beam switching, the beam width should be maximized. Considering the requirements for the communication link rate, the point closer to the base station should be at the edge of the beam. The beam drop of 3dB is regarded as the beam width. The gain at the point closer to the base station should be half of the maximum gain. The adjustment formula is:

[0034]

[0035] Among them, R Th Indicates the preset communication rate threshold, d 1 represents the distance between the point closer to the base station and the point before the base station in the starting position and the end position in S, ρ represents the signal-to-noise ratio when the beam generated by the base station communicates with the dynamic target, φ 1 represents the angle between the starting position in S and the direction of the beam center;

[0036] The base station passes Φ -3dB and the maximum error range e max Calculate the optimal beam width φ′ -3dB and the optimal beam direction ψ, the calculation formulas are:

[0037] φ′ -3dB =Φ -3dB +e max

[0038] ψ=θ1 +φ 1

[0039] Step 4), the base station obtains the optimal codeword:

[0040] The base station determines the communication rate R of the starting position of the predicted trajectory S St The communication rate threshold R Th Does it satisfy R St >R Th , if so, through the optimal beam width φ′ -3dB and the optimal beam direction ψ to calculate the optimal codeword W opt , the calculation formula is:

[0041] W opt =[W * (1,1),W * (2,1),...,W * (m,p),...,W * (N′,1)]

[0042] W * (m,p)=W(m,p)×X T (m)

[0043]

[0044] X(m)=exp(jπ(m-1)Δψ)

[0045]

[0046] Where exp(·) represents the exponential function with the natural constant e as the base, j represents the imaginary unit, Indicates the rounding up operation, W * (m,p) represents the weight value in the weight vector required for the beam in the pth direction generated by the mth antenna used for beam tracking in the base station, N′ represents the number of antennas used by the base station for beam tracking, W(m,p) represents the weight value in the weight vector required for the beam in the pth direction generated by the mth antenna of the discrete Fourier change codebook, X(m) represents the phase shift weight value of the mth antenna of the base station in the direction of Δψ, Δψ represents the interval between the direction of the beam generated by the first codeword in the generated codebook W and the optimal beam direction ψ, and D represents the number of bits of the digital phase shifter used by the base station;

[0047] If the base station determines the communication rate R at the starting position of the predicted trajectory S St The communication rate threshold R Th Does not meet R St >R Th, then it is necessary to adaptively adjust the optimal beam direction ψ, and the base station adjusts the optimal beam width φ′ -3dB and the theoretical optimal beam width Φ -3dB Calculate the refinement factor α, the calculation formula is:

[0048]

[0049] in Indicates a round-down operation;

[0050] Because the communication rate between the base station and the target is not up to standard and the tracking accuracy is insufficient, the gain of the beam generated by the base station should be increased. The base station calculates the weighted value W in the weighted vector required for the c-th direction beam generated by the b-th antenna in the a-th layer codebook for beam tracking in the training layered codebook according to α. Train (a, b, c), we get a set W of E×αN′×αN′ weighted values. Train , E represents the number of all prime factors of α, where W Train The calculation formula for (a,b,c) is:

[0051]

[0052] X′(b)=exp(jπ(b-1)Δψ)

[0053] Where N a represents the number of antennas used by the a-th layer codebook for beam tracking, and its value is the multiple of the a-th prime factor multiplied from the first prime factor of α in N′ from small to large. W′(a,b,c) represents the weighted value in the weighted vector required for the c-th direction beam generated by the b-th antenna used for beam tracking in the a-th layer codebook in the codebook with consistent beam width in each layer. X′(b) represents the weighted value of the phase shift of the b-th antenna of the base station in the direction of Δψ.

[0054] Finally, the base station passes W Train The codewords corresponding to each row in the upper layer communicate with the dynamic target in turn, and obtain the signal-to-noise ratio when communicating with the dynamic target. Since the beam width of the top layer codebook of the calculated layered codebook is adaptively adjusted to φ′ -3dB , the initial search range is limited to a smaller range, the total number of searches is only the sum of all prime factors of α, and then the codeword corresponding to the maximum signal-to-noise ratio is taken as the optimal codeword W opt ;

[0055] Step 5), the base station obtains the beam tracking result of the dynamic target;

[0056] The base station uses the optimal codeword W opt Communicate with the dynamic target and determine the transmission rate R of the dynamic target communicating with the base station Rt Does it satisfy RRt >R Th , if so, the dynamic target is within the beam range of the current time period, otherwise execute step (2).

[0057] The following is a further description of the technical effects of the present invention in conjunction with the simulation results:

[0058] 1. Experimental conditions and contents:

[0059] The application platform of the simulation experiment is: the processor is 14-core Intel i5 13600KF, the base frequency is 3.5GHz, the turbo frequency is 5.1GHz, and the memory is 32GB. The software platform is: MATLAB R2021B.

[0060] The parameters used in the simulation experiment are shown in Table 1.

[0061] Table 1

[0062] Parameter name Parameter Value Number of base station antennas 32 Horizontal distance (m) 100 Channel transmission signal-to-noise ratio (dB) 35 Positioning sampling number (times) 1000 Number of base station subcarriers 30

[0063] The tracking accuracy of the beam tracking method in the present invention is compared with that in the existing "Hierarchical Codebook Design for Near-Field mmWave MIMO Communications Systems". The results are as follows: Figure 2 As shown;

[0064] 2. Experimental results analysis:

[0065] like Figure 2 As shown, the horizontal axis represents the dynamic target at different speeds V, and the vertical axis represents the communication rate R between the dynamic target and the base station under the two methods. The curve of the communication rate of the present invention changing with speed is always above the existing training-based method, and the slope of the communication rate decrease is more gentle. This result shows that under the same moving speed conditions, the tracking accuracy of the present invention is higher, thereby providing better communication quality for the target. Specifically, in low-speed scenarios, the difference in tracking accuracy between the present invention and the training-based method is not significant; however, as the speed continues to increase, the gap between the two gradually widens. This phenomenon shows that the present invention shows higher applicability when dealing with high-speed dynamic targets. This finding is consistent with our expectations, because the present invention uses trajectory prediction information to calculate the beam direction to perform beam tracking on dynamic targets, which only involves matrix calculations, effectively avoiding the high time cost caused by frequent beam training in the training-based method, thereby achieving better tracking accuracy.

[0066] Depend on Figure 2It can be seen that the curve of the communication rate between the dynamic target and the base station of the present invention shows a downward trend as the speed changes, indicating that as the target moving speed continues to increase, the tracking accuracy of the beam tracking algorithm is also decreasing. This is consistent with the actual situation. The faster the moving speed of the dynamic target, the more accurate the base station's positioning of it will be, which will in turn affect the accuracy of the base station's trajectory prediction for the current period. This will cause the calculated maximum trajectory prediction error range to become larger, resulting in an increase in the width of the optimal beam, and at the same time, the antenna gain will be reduced. In addition, the base station needs to complete the entire calculation process in a shorter time.

[0067] The above simulation experiments show that a dynamic target beam tracking method based on an adaptive codebook with a coverage error range is suitable for scenarios with high dynamic target speeds. It has high tracking accuracy and can ensure the communication quality of dynamic targets when they are moving. It is an effective and stable beam tracking method.

Claims

1. A dynamic target beam tracking method based on coverage error range and adaptive codebook, characterized in that: The steps include: (1) Constructing a beam tracking scenario for a dynamic target; Construct a beam tracking scenario including a uniform linear array base station with a total number of N antennas located at the origin of a two-dimensional coordinate system and a dynamic target distributed in a two-dimensional coordinate system for wireless communication, wherein N ≥ 2; (2) The base station calculates the maximum error range of the dynamic target trajectory in the current time period; The base station predicts the trajectory S in the current time period based on the positioning information of the dynamic target in the historical time period, and calculates the maximum error range e of S by the angles θ1 and θ2 between the points closer to and farther from the base station in the starting position and the end position of S and the normal line of the base station max ; (3) The base station adaptively adjusts the beam width and calculates the optimal beam width and optimal beam direction; The base station adaptively adjusts the beam width to minimize the beam switching overhead and obtains the theoretical optimal beam width Φ -3dB , and through Φ -3dB and the maximum error range e max Calculate the optimal beam width φ′ -3dB and the optimal beam heading ψ; (4) The base station obtains the optimal codeword: The base station determines the communication rate R of the starting position of the predicted trajectory S St The communication rate threshold R Th Does it satisfy R St >R Th , if so, through the optimal beam width φ′ -3dB and the optimal beam direction ψ to calculate the optimal codeword W opt Otherwise, the optimal beam direction ψ is adaptively adjusted to obtain the optimal codeword W opt ; (5) The base station obtains the beam tracking results of the dynamic target; The base station uses the optimal codeword W opt Communicate with the dynamic target and determine the transmission rate R of the dynamic target communicating with the base station Rt Does it satisfy R Rt >R Th , if so, the dynamic target is within the beam range of the current time period, otherwise execute step (2).

2. The method according to claim 1, characterized in that The maximum error range of S described in step (2) is e max , the calculation formula is: the max =e1+e2 Among them, e1 and e2 represent the maximum prediction errors of the starting position and the end position in the predicted trajectory S, respectively. They represent the standard deviation of the trajectory prediction error of the target with arrival angles of θ1 and θ2 respectively, K represents the number of subcarriers that can be transmitted by each antenna of the base station, K ≥ 10, ρ represents the signal-to-noise ratio when the beam generated by the base station communicates with the dynamic target, and F represents the number of samples of the dynamic target when the base station performs trajectory prediction, F ≥ 100.

3. The method according to claim 2, characterized in that The base station in step (3) adaptively adjusts the beam width with the goal of minimizing the beam switching overhead, specifically: The base station adaptively adjusts the beam width with the goal of minimizing the beam switching overhead. The adjustment formula is: Among them, R Th represents the preset communication rate threshold, d1 represents the distance between the starting position and the end position in S that is closer to the base station and the base station, and φ1 represents the angle between the starting position in S and the center direction of the beam.

4. The method according to claim 3, characterized in that The optimal beam width φ′ described in step (3) -3dB and the optimal beam direction ψ, the calculation formulas are: f - ′ 3dB =Φ -3dB +e max ψ=θ1+φ1.

5. The method according to claim 1, characterized in that The optimal beam width φ′ described in step (4) -3dB and the optimal beam direction ψ to calculate the optimal codeword W opt , the calculation formula is: W opt =[W * (1,1),W * (2,1),...,W * (m,p),...,W * (N′,1)] W * (m,p)=W(m,p)×X T (m) X(m)=exp(jπ(m-1)Δψ) Where exp(·) represents the exponential function with the natural constant e as the base, j represents the imaginary unit, Indicates rounding up operation, W * (m,p) represents the weighted value in the weighted vector required for the beam in the p-th direction generated by the m-th antenna used for beam tracking in the base station, N′ represents the number of antennas used by the base station for beam tracking, W(m,p) represents the weighted value in the weighted vector required for the beam in the p-th direction generated by the m-th antenna of the discrete Fourier change codebook, X(m) represents the weighted value of the phase shift in the Δψ direction changed by the m-th antenna of the base station, Δψ represents the interval between the direction of the beam generated by the first codeword in the generated codebook W and the optimal beam direction ψ, and D represents the number of bits of the digital phase shifter used by the base station.

6. The method according to claim 5, characterized in that The adaptive adjustment of the optimal beam direction ψ described in step (4) is implemented by: (4a) The base station uses the optimal beam width φ′ -3dB and the theoretical optimal beam width Φ -3dB Calculate the refinement factor α: in Indicates a round-down operation; (4b) The base station calculates the weight value W in the weight vector required for the c-th direction beam generated by the b-th antenna in the a-th layer codebook for beam tracking in the training layered codebook according to α Train (a, b, c), we get a set W of E×αN′×αN′ weighted values. Train , E represents the number of all prime factors of α, where W Train The calculation formula for (a,b,c) is: X′(b)=exp(jπ(b-1)Δψ) Where N a represents the number of antennas used by the a-th layer codebook for beam tracking, and its value is the multiple of the a-th prime factor multiplied from the first prime factor of α in N′ from small to large. W′(a,b,c) represents the weighted value in the weighted vector required for the c-th direction beam generated by the b-th antenna used for beam tracking in the a-th layer codebook in the codebook with consistent beam width in each layer. X′(b) represents the weighted value of the phase shift of the b-th antenna of the base station in the direction of Δψ. (4c) The base station passes W Train The codewords corresponding to each row in the high-level layer communicate with the dynamic target in turn, and obtain the signal-to-noise ratio when communicating with the dynamic target. Then, the codeword corresponding to the maximum signal-to-noise ratio is used as the optimal codeword W opt .

Images