A Fast Beam Alignment Method Based on Particle Swarm Optimization Search
By employing a wide-beam-narrow-beam hierarchical search method and particle swarm optimization algorithm, the problem of high beam alignment complexity in large-scale MIMO systems is solved, achieving low-complexity beam alignment and improving the performance and access latency of the communication system.
Patent Information
- Application Number
- CN202311684083.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-05
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2043-12-05
AI Technical Summary
The beam alignment methods in existing communication standards are too complex in large-scale MIMO systems and cannot effectively handle the problem of high-precision narrow beam alignment generated by a large number of antenna arrays.
A particle swarm optimization (PSO)-based search method is adopted to simplify the beam alignment process of large-scale MIMO systems through wide-beam-narrow-beam hierarchical search and PSO algorithm. The optimal beam alignment is determined by utilizing the fast convergence and low complexity of the PSO algorithm.
It effectively reduces the beam alignment complexity of large-scale MIMO systems, improves the performance of communication systems, reduces the possibility of signal collisions and network congestion, and reduces initial access delay, making it suitable for high-speed communication scenarios.
Smart Images

Figure CN117639871B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of large-scale multiple-input multiple-output communication systems, and in particular to a fast beam alignment method based on particle swarm optimization search. Background Technology
[0002] Massive Multiple-Input Multiple-Output (MIMO) is a wireless communication technology based on antenna arrays. It improves the spectral efficiency, reliability, and energy efficiency of a system by using a large number of antennas and complex signal processing algorithms. Therefore, MIMO technology is widely used in fields such as communications and radar, and is one of the core enabling technologies of modern wireless communication systems. With its significant diversity and multiplexing advantages, MIMO systems are widely used in LTE-A, Wireless Local Area Networks (WLANs), and the World Interoperability for Microwave Access (WiMAX) standard. Furthermore, with the continuous improvement of hardware integration technology, MIMO technology has evolved further. MIMO technologies in 5G communication standards include Massive MIMO and beamforming. Among them, Massive MIMO technology, based on traditional MIMO technology, significantly increases the number of antenna elements at the transmitting and receiving ends, which can greatly improve the spatial gain of the Massive MIMO system while reducing noise, fading, and inter-cell interference. Compared to traditional MIMO technology, massive MIMO technology, relying on transceivers equipped with a large number of antennas, can significantly improve the channel capacity and spectral efficiency of communication links under the same bandwidth and video resources. Beamforming technology, based on the massive antenna arrays of massive MIMO communication systems, further expands the performance advantages of MIMO systems in terms of spectral efficiency and signal gain. Specifically, beamforming technology generates highly directional spatial beams by controlling the phase and power of the transceiver antennas, which can enhance the signal in the direction of a specific user or target, thereby improving the signal transmission rate and quality, and effectively suppressing interference.
[0003] Millimeter-wave technology and massive MIMO technology complement each other. First, the shorter wavelength of millimeter waves allows for smaller antenna sizes, facilitating the deployment of numerous antennas within a small footprint, enabling large-scale MIMO integrated communication systems. Second, millimeter-wave massive MIMO communication systems can provide sufficiently large beam gain through beamforming / precoding techniques to compensate for the rapid attenuation and loss of millimeter-wave signals in free space, establishing stable communication links. However, because the spatial beams generated by beamforming in massive antenna arrays are narrow, when the location of the other party is unknown, it is necessary to scan the beams in three-dimensional space and obtain the maximum received energy through feedback. The optimal beam alignment process is then determined based on this maximum received energy. However, existing beam alignment methods in communication standards, such as traversal scanning and hierarchical iterative search methods, are only suitable for traditional MIMO communication systems with a small number of beams. In massive MIMO communication systems, the introduction of numerous additional antennas and the generation of more high-precision narrow beams through beamforming make the complexity of traditional beam alignment methods unacceptable. Therefore, it is necessary to research fast beam alignment methods suitable for massive MIMO antenna systems. Summary of the Invention
[0004] To address the technical problems existing in the background art, this invention proposes a fast beam alignment method based on particle swarm optimization search.
[0005] This invention proposes a fast beam alignment method based on particle swarm optimization search, comprising:
[0006] S1. The transmitter and receiver each select M1 (M1 < M) antenna array elements to generate multiple sets of transmit-receive first beam pairs. The number of antenna arrays at both the transmitter and receiver is M, and the first beamwidth is... The number of first beams at both the transmitter and receiver is k1 and
[0007] S2. Traverse and scan multiple sets of transmit-receive first beam pairs to obtain the first receive energy matrix corresponding to the multiple sets of transmit-receive first beam pairs.
[0008] S3. Obtain the optimal first beam pointing in multiple sets of transmit-receive first beam pairs according to the first receive energy matrix E1, and calculate the corresponding first transmitter angle region and first receiver angle region based on the optimal first beam pointing.
[0009] S4. The transmitting and receiving ends each select M2 (M1 < M2 < M) antenna array elements to generate multiple sets of transmit-receive second beam pairs, and traverse and scan the multiple sets of transmit-receive second beam pairs based on the first transmitting end angle region and the first receiving end angle region to obtain the second receiving energy matrix corresponding to the multiple sets of transmit-receive second beam pairs; the second beamwidth is... The number of second beams at both the transmitter and receiver is k2 and
[0010] S5. Obtain the optimal second beam pointing in multiple sets of transmit-receive second beam pairs according to the second receive energy matrix E2, and calculate the corresponding second transmitter angle region and second receiver angle region based on the optimal second beam pointing.
[0011] S6. The transmitter and receiver each select M antenna array elements to generate multiple sets of transmit-receive third beam pairs, and use a particle swarm optimization search strategy to determine the optimal third receive energy in the solution space formed by the second transmitter angle region and the second receiver angle region, so as to obtain the optimal transmit-receive third beam pair.
[0012] Preferably, the beamwidth relationship among the first beam, the second beam, and the third beam is as follows: first beam > second beam > third beam.
[0013] Preferably, step S1 specifically includes:
[0014] Control the transmitter with the first beamwidth Generate the first transmission beam
[0015] Control the receiver to use the first beamwidth Generate and receive the first beam
[0016] The number of first beams at both the transmitter and receiver is k1 and
[0017] Preferably, step S2 specifically includes:
[0018] The system iterates through multiple transmit-receive first beam pairs to obtain the first received energy corresponding to each transmit-receive first beam pair, and constructs the first received energy matrix.
[0019] First receiving energy matrix The calculation process is as follows:
[0020] E1(θ i ,φ j )=∫ θ ∫ φ P T GLOSS T1(θ i )R1(φ j )dθdφ
[0021] Where, T1(θ) i ) indicates that the beam pointing at the transmitting end is θ i At that time, the wide beam gain of the transmitter, R1(φ) j ) indicates that the receiver beam direction is φ j The receiver's wide beam gain at that time, P T G represents the transmit power of the transmitter. LOSS This refers to link attenuation during signal transmission.
[0022] Preferably, the calculation process for the optimal first beam pointing is as follows:
[0023] (θ best ,φ best ) = argmaxE1(θ i ,φ j )
[0024] Where, θ i The beam direction of the transmitting end; φ j θ is the beam pointing at the receiving end. best ,φ best The optimal first beam direction;
[0025] The specific angle region of the first transmitting end is as follows:
[0026] The specific angle region of the first receiving end is as follows:
[0027] Preferably, the transmitting end and the receiving end each select M2 (M1 < M2 < M) antenna array elements to generate multiple sets of transmit-receive second beam pairs, specifically:
[0028] Control the transmitter with the second beamwidth Generate and transmit the second beam
[0029] Control the receiver to use the second beamwidth Generate and receive the second beam
[0030] The number of second beams at both the transmitter and receiver is k2 and
[0031] Preferably, the step of traversing and scanning multiple sets of transmit-receive second beam pairs based on the first transmitter angle region and the first receiver angle region to obtain the second receive energy matrix corresponding to the multiple sets of transmit-receive second beam pairs is carried out. Specifically:
[0032] The system iterates through multiple transmit-receive second beam pairs to obtain the second received energy corresponding to each transmit-receive second beam pair, and then constructs the second received energy matrix.
[0033] Second receiving energy matrix The calculation process is as follows:
[0034] E2(θ i ,φ j )=∫ θ ∫ φ P T G LOSS T2(θ i )R2(φ j )dθdφ
[0035] Where T2(θ) i ) indicates that the beam pointing at the transmitting end is θ i Narrow beam gain at the transmitter, R2(φ) j ) indicates that the receiver beam direction is φ j Narrow beam gain at the receiver, P T G represents the transmit power of the transmitter. LOSS This refers to link attenuation during signal transmission.
[0036] Preferably, the calculation process for the optimal second beam pointing is as follows:
[0037] (θ best2 ,φ best2 )=argmaxE2(θ i ,φ j )
[0038] The specific angle region of the second transmitting end is as follows:
[0039] The specific angle region of the second receiving end is as follows:
[0040] Preferably, step S6 specifically includes:
[0041] The transmitting and receiving ends each select M antenna array elements to generate multiple sets of transmit-receive third beam pairs;
[0042] The solution space is constructed based on the angle regions of the second transmitting end and the second receiving end.
[0043] Within the solution space, initialize the particle swarm parameters and randomly initialize the initial exploration positions of n particles and the velocity of each particle;
[0044] Calculate the objective function value of each particle at present. Based on the objective function values of all particles calculated at present and in the past, determine the historical optimal energy value and the corresponding optimal individual exploration point of each particle.
[0045] The optimal swarm exploration point is determined based on the historical best energy value explored by the entire particle swarm.
[0046] The velocity of each particle is updated based on the optimal individual exploration point and the optimal group exploration point to obtain the first exploration position of each particle in the next round;
[0047] Calculate the updated objective function value for each particle based on its first exploration position in the next round;
[0048] The exploration position and objective function value of each particle are iteratively updated in each round until the convergence condition is met or the maximum number of iterations is reached;
[0049] When the number of iterations is reached or the convergence condition is met, the transmit-receive third beam pair corresponding to the currently searched optimal objective function value is taken as the optimal transmit-receive third beam pair and output.
[0050] Preferably, the convergence condition is as follows: the difference between the optimal energy values obtained from two adjacent explorations is less than the convergence threshold.
[0051] This invention proposes a fast beam alignment method based on particle swarm optimization (PSO) search. Through a wide-beam-narrow-beam search approach, it can obtain a good solution space with low complexity. Based on the solution space obtained from two beam scans, a narrower working beam can be used for alignment, and the PSO algorithm effectively simplifies this narrow-beam alignment process. After these three stages of processing, the optimal transmit / receive beam pair pointing at the working beam accuracy can be obtained. Introducing the PSO algorithm into the framework of traditional beam alignment algorithms effectively solves the problem of high complexity in large-scale antenna systems by leveraging its simplicity, low complexity, and fast convergence speed. By introducing a wide-beam-narrow-beam hierarchical search method, an initial solution space without local optima is determined with low complexity, thus effectively ensuring the success rate of the PSO algorithm. Attached Figure Description
[0052] Figure 1 This is a schematic diagram illustrating the workflow of a fast beam alignment method based on particle swarm optimization search proposed in this invention.
[0053] Figure 2 This is a schematic diagram of the beam alignment structure of a fast beam alignment method based on particle swarm optimization search proposed in this invention.
[0054] Figure 3 This is a schematic diagram of the received energy structure under the beam pointing direction of different receivers and transmitters in a fast beam alignment method based on particle swarm optimization search proposed in this invention.
[0055] Figure 4 This is a schematic diagram of the wide-beam-narrow-beam scanning search structure of a fast beam alignment method based on particle swarm optimization search proposed in this invention. Detailed Implementation
[0056] Reference Figure 1-4 The present invention proposes a fast beam alignment method based on particle swarm optimization search, comprising:
[0057] S1. The transmitter and receiver each select M1 (M1 < M) antenna array elements to generate multiple sets of transmit-receive first beam pairs. The number of antenna arrays at both the transmitter and receiver is M, and the first beamwidth is... The number of first beams at both the transmitter and receiver is k1 and
[0058] In this embodiment, step S1 specifically includes:
[0059] Control the transmitter with the first beamwidth Generate the first transmission beam
[0060] Control the receiver to use the first beamwidth Generate and receive the first beam
[0061] The number of first beams at both the transmitter and receiver is k1 and
[0062] S2. Traverse and scan multiple sets of transmit-receive first beam pairs to obtain the first receive energy matrix corresponding to the multiple sets of transmit-receive first beam pairs.
[0063] In this embodiment, step S2 specifically includes:
[0064] The system iterates through multiple transmit-receive first beam pairs to obtain the first received energy corresponding to each transmit-receive first beam pair, and constructs the first received energy matrix.
[0065] First receiving energy matrix The calculation process is as follows:
[0066] E1(θ i ,φ j )=∫ θ ∫ φ P T GLOSS T1(θ i )R1(φ j )dθdφ
[0067] Where, T1(θ) i ) indicates that the beam pointing at the transmitting end is θ i At that time, the wide beam gain of the transmitter, R1(φ) j ) indicates that the receiver beam direction is φ j The receiver's wide beam gain at that time, P T G represents the transmit power of the transmitter. LOSS This refers to link attenuation during signal transmission.
[0068] S3. Obtain the optimal first beam pointing among multiple sets of transmit-receive first beam pairs according to the first receive energy matrix E1, and calculate the corresponding first transmitter angle region and first receiver angle region based on the optimal first beam pointing.
[0069] In this embodiment, the calculation process for the optimal first beam pointing is as follows:
[0070] (θ best ,φ best ) = argmaxE1(θ i ,φ j )
[0071] Where, θ i The beam direction of the transmitting end; φ j θ is the beam pointing at the receiving end. best ,φ best The optimal first beam direction;
[0072] The specific angle region of the first transmitting end is as follows:
[0073] The specific angle region of the first receiving end is as follows:
[0074] like Figure 4 As shown in (a), in this embodiment, the approximate region where the optimal beam pair is located is first determined by scanning the first beam, i.e., the wide beam. The specific steps are as follows:
[0075] 1) Set the number of antenna elements M1 (M1 < M) to generate the initial scanning wide beam, and obtain the beamwidth of the beam. And based on the current beamwidth Set the transmitter beam With receiver beam To effectively cover the entire angular space.
[0076] 2) Scan all current transmitter beams and all current receiver beams By traversing all beam pairs, the received energy matrix determined by the transmit and receive beam pairs can be obtained. as follows:
[0077] E1(θ i ,φ j )=∫ θ ∫ φ P T G LOSS T1(θ i )R1(φ j )dθdφ
[0078] Where T1(θ) i ) indicates that the beam pointing at the transmitting end is θ i At that time, the wide beam gain of the transmitter, R1(φ) j ) indicates that the receiver beam direction is φ j The receiver's wide beam gain at that time, P T G represents the transmit power of the transmitter. LOSS This refers to link attenuation during signal transmission. Based on the energy matrix E1, the optimal beam pointing under the current wide beam can be obtained as follows:
[0079] (θ best ,φ best ) = argmaxE1(θ i ,φ j )
[0080] 3) This yields the optimal transmitter angle region under wide beam conditions. Optimal receiver angle region Used for further beam searching.
[0081] S4. The transmitter and receiver each select M2 (M1 < M2 < M) antenna array elements to generate multiple sets of transmit-receive second beam pairs, and traverse and scan the multiple sets of transmit-receive second beam pairs based on the first transmitter angle region and the first receiver angle region to obtain the second receive energy matrix corresponding to the multiple sets of transmit-receive second beam pairs. The second beamwidth is The number of second beams at both the transmitter and receiver is k2 and
[0082] In this embodiment, the transmitting end and the receiving end each select M2 (M1 < M2 < M) antenna array elements to generate multiple sets of transmit-receive second beam pairs, specifically including:
[0083] Control the transmitter with the second beamwidth Generate and transmit the second beam
[0084] Control the receiver to use the second beamwidth Generate and receive the second beam
[0085] The number of second beams at both the transmitter and receiver is k2 and
[0086] In this embodiment, multiple sets of transmit-receive second beam pairs are traversed and scanned based on the first transmitter angle region and the first receiver angle region to obtain the second receive energy matrix corresponding to the multiple sets of transmit-receive second beam pairs. Specifically, it includes:
[0087] The system iterates through multiple transmit-receive second beam pairs to obtain the second received energy corresponding to each transmit-receive second beam pair, and then constructs the second received energy matrix.
[0088] Second receiving energy matrix The calculation process is as follows:
[0089] E2(θ i ,φ j )=∫ θ ∫ φ P T G LOSS T2(θ i )R2(φ j )dθdφ
[0090] Where T2(θ) i ) indicates that the beam pointing at the transmitting end is θ i Narrow beam gain at the transmitter, R2(φ) j ) indicates that the receiver beam direction is φ j Narrow beam gain at the receiver, P T G represents the transmit power of the transmitter. LOSS This refers to link attenuation during signal transmission.
[0091] S5. Obtain the optimal second beam pointing among multiple sets of transmit-receive second beam pairs based on the second receive energy matrix E2, and calculate the corresponding second transmitter angle region and second receiver angle region based on the optimal second beam pointing.
[0092] In this embodiment, the calculation process for the optimal second beam pointing is as follows:
[0093] (θ best2 ,φ best2 )=argmaxE2(θ i ,φ j )
[0094] The specific angle region of the second transmitting end is as follows:
[0095] The specific angle region of the second receiving end is as follows:
[0096] like Figure 4 As shown in (b), in this embodiment, a narrow beam search is performed within the first angle region between the transmitting end and the receiving end obtained in the previous stage. The specific steps are as follows:
[0097] 1) Set the number of antenna elements M2 (M1 < M2 < M) for narrow beam scanning to obtain the beamwidth of the beam. And based on the current beamwidth Set the transmit beam and receiving beam To effectively cover the optimal angle region determined by wide beam search. and
[0098] 2) Scan all transmitting beams within the scanning angle region. and all current receiver beams By traversing all beam pairs, the received energy matrix determined by the transmit and receive beam pairs can be obtained. as follows:
[0099] E2(θ i ,φ j )=∫ θ ∫ φ P T G LOSS T2(θ i )R2(φ j )dθdφ
[0100] Where T2(θ) i ) indicates that the beam pointing at the transmitting end is θ i Narrow beam gain at the transmitter, R2(φ) j ) indicates that the receiver beam direction is φ j Narrow beam gain at the receiver, P T G represents the transmit power of the transmitter. LOSS This refers to link attenuation during signal transmission. Based on the energy matrix E2, the optimal beam pointing under the current narrow beam can be obtained as follows:
[0101] (θ best2 ,φ best2 )=argmaxE2(θ i ,φ j )
[0102] 3) Transmitter angle domain
[0103] Receiver angle domain
[0104] This can then be used as the solution space for the particle swarm optimization algorithm to perform the next beam search.
[0105] S6. The transmitter and receiver each select M antenna array elements to generate multiple sets of transmit-receive third beam pairs, and use a particle swarm optimization search strategy to determine the optimal third receive energy in the solution space formed by the second transmitter angle region and the second receiver angle region, so as to obtain the optimal transmit-receive third beam pair.
[0106] Specifically, step S6 includes:
[0107] The transmitter and receiver each select M antenna array elements to generate multiple sets of transmit-receive third beam pairs. A solution space is constructed based on the second transmitter angle region and the second receiver angle region. Within the solution space, particle swarm parameters are initialized, and the initial exploration positions and velocities of n particles are randomly initialized. The objective function value of each particle is calculated. Based on the objective function values of all particles calculated currently and historically, the historical optimal energy value and corresponding optimal individual exploration point of each particle are determined. The optimal group exploration point is determined based on the historical optimal energy value explored by the particle swarm as a whole. The velocity of each particle is updated based on the optimal individual exploration point and the optimal group exploration point to obtain the first exploration position of each particle in the next round. The updated objective function value of each particle is calculated based on the first exploration position of each particle in the next round. The exploration position and objective function value of each particle in each round are iteratively updated until the convergence condition is met or the maximum number of iterations is reached. When the number of iterations is reached or the convergence condition is met, the transmit-receive third beam pair corresponding to the currently searched optimal objective function value is taken as the optimal transmit-receive third beam pair and output.
[0108] Specifically, the convergence condition is: the difference between the optimal energy values obtained from two consecutive explorations is less than the convergence threshold.
[0109] Specifically, the beamwidth relationship between the first beam, the second beam, and the third beam is as follows: first beam > second beam > third beam.
[0110] In this embodiment, adjusting the number of antenna elements to M during the particle swarm optimization search phase yields a beam with stronger directivity and higher directional accuracy than in the wide-beam search and narrow-beam search phases. The solution space of the particle swarm optimization search is the angular domain between the transmitter and receiver determined in the narrow-beam search phase. and
[0111] In this embodiment, the solution space is specifically:
[0112]
[0113] Find the optimal transmit-receive beam pair (θ) in the solution space. best ,φ best The objective function is the received signal energy E3(θ,φ) under the transmit-receive beam pair. The optimization objective of particle swarm optimization search is to maximize the received energy E3, and the specific steps are as follows:
[0114] 1) Initialize the particle swarm parameters in the solution space, including the particle swarm size n, the inertia weight w, the number of iterations and the convergence threshold η, where the particle swarm size depends on the size of the solution space, and the number of iterations and the convergence threshold depend on the desired alignment accuracy.
[0115] 2) Randomly initialize the initial exploration positions (θ) of n particles in the solution space. i ,φ i ), i∈{1,2,…,n}, and randomly initialize the velocity v of each particle. i Let i ∈ {1,2,…,n}, and calculate the objective function value E3 for each particle as follows:
[0116] E3(i)=∫ θ ∫ φ P T G LOSS T3(θ i )R3(φ i )dθdφ
[0117] Where T3(θ) i ) indicates that the beam pointing at the transmitting end is θ i The transmitter beam gain at that time, R3(φ) i ) indicates that the receiver beam direction is φ i Narrow beam gain at the receiver, P T G represents the transmit power of the transmitter. LOSS This refers to link attenuation during signal transmission.
[0118] 3) Based on the objective function values of all particles calculated currently and historically, the individual optimal exploration point (θ) corresponding to the historical optimal energy value of each particle can be determined. i ,φ i ) pbest The optimal exploration point (θ,φ) for the swarm is determined based on the historical optimal energy information obtained by the entire particle swarm. gbest The velocity of each particle is updated based on the two optimal exploration points as follows:
[0119] v i,next=wv i +c1r1((θ i ,φ i ) pbest -(θ i ,φ i ))+c2r2((θ,φ) gbest -(θ i ,φ i ))
[0120] Where c1 and c2 represent individual learning factors and group learning factors, indicating the degree to which the particle's next exploration strategy depends on individual experience and the experience of other particles, and r1 and r2 are random numbers used to increase the randomness of the exploration.
[0121] 4) Based on the updated velocity of each particle, the exploration position of each particle in the next round can be calculated as follows:
[0122] (θ i ,φ i ) next =(θ i ,φ i )+v i,next
[0123] And calculate the updated objective function value E3 corresponding to the particle based on the particle update position in the next round.
[0124] 5) If the number of iterations is reached or the convergence condition is met, output the transmit / receive beam pair corresponding to the currently searched optimal objective function as the optimal solution. The convergence condition is that the difference between the optimal energy values obtained in the two consecutive explorations is less than η; otherwise, return to step 3.
[0125] The fast beam alignment method based on particle swarm optimization search in this embodiment mainly includes three stages: (1) wide beam search stage; (2) narrow beam search stage; and (3) particle swarm optimization search stage. For the received energy solution space constructed by the transmitter beam and the receiver beam, a wide beam generated by a small number of antennas is first used to scan the spatial angle region. Based on the optimal received energy obtained from the scan, the approximate angle region where the optimal beam is located is determined. Then, within the angle region determined by the wide beam scan, a narrower spatial beam than in the first stage is generated by increasing the number of antennas, and a narrow beam is used to scan within this determined angle region to further determine the region where the optimal beam is located. The wide beam-narrow beam hierarchical search stage is designed because the beam generated by beamforming contains sidelobes. The alignment of the sidelobes will cause the solution space composed of received energy to contain many local optima. Therefore, wide beam-narrow beam search is used to refine the search region and avoid local optima. Through the wide beam-narrow beam search method, a good solution space can be obtained with low complexity. Based on the solution space obtained from two beam scans, a narrower working beam can be used for alignment, and the alignment process for this narrow beam can be effectively simplified using a particle swarm optimization algorithm. After the above three stages of processing, the optimal transceiver beam pair pointing at the working beam accuracy can be obtained.
[0126] To overcome the high beam alignment complexity in large-scale MIMO systems caused by the introduction of large-scale antenna arrays, this invention improves upon traditional beam alignment algorithms for multi-antenna systems and introduces Particle Swarm Optimization (PSO), a numerical optimization algorithm. Leveraging PSO's fast convergence speed, high robustness, and ease of implementation, a novel fast beam alignment algorithm for large-scale antenna arrays is proposed. This method effectively reduces the complexity of the beam alignment process in large-scale antenna systems, thereby reducing the likelihood of signal collisions and network congestion, and improving the performance of the communication system. Furthermore, the low-complexity beam alignment algorithm results in lower initial signal access delay, which is of great significance for the application of large-scale MIMO communication systems in high-speed communication scenarios.
[0127] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.
Claims
1. A fast beam alignment method based on particle swarm optimization search, characterized in that, include: S1, the transmitting end and the receiving end respectively select a plurality of groups of sending-receiving first beam pairs, the number of the antenna array of the transmitting end and the receiving end is , the first beam width is , the number of the first beam of the transmitting end and the receiving end is and , wherein ; S2. Traverse and scan multiple sets of transmit-receive first beam pairs to obtain the first receive energy matrix corresponding to the multiple sets of transmit-receive first beam pairs. ; S3, According to the first received energy matrix To obtain the optimal first beam pointing among multiple sets of transmit-receive first beam pairs, and to calculate the corresponding first transmitter angle region and first receiver angle region based on the optimal first beam pointing; S4, the transmitter and receiver are selected respectively. Each antenna element generates multiple sets of transmit-receive second beam pairs, and traverses and scans these multiple sets of transmit-receive second beam pairs based on the first transmitter angle region and the first receiver angle region to obtain the second receive energy matrix corresponding to each set of transmit-receive second beam pairs. The second beamwidth is The number of second beams at both the transmitting and receiving ends is [number missing]. and ,in, ; S5, According to the second received energy matrix To obtain the optimal second beam pointing among multiple sets of transmit-receive second beam pairs, and to calculate the corresponding second transmitter angle region and second receiver angle region based on the optimal second beam pointing; S6, the transmitter and receiver are selected respectively. Each antenna array element generates multiple sets of transmit-receive third beam pairs, and uses a particle swarm optimization search strategy to determine the optimal third receive energy in the solution space formed by the second transmit end angle region and the second receive end angle region, so as to obtain the optimal transmit-receive third beam pair. The calculation process for the optimal first beam pointing is as follows: in, The direction of the transmitting beam; The beam direction at the receiving end; The optimal first beam direction; The specific angle region of the first transmitting end is as follows: ; The specific angle region of the first receiving end is as follows: ; The calculation process for the optimal second beam pointing is as follows: The specific angle region of the second transmitting end is as follows: ; The specific angle region of the second receiving end is as follows: ; Step S6 specifically includes: Select the transmitter and receiver respectively Each antenna array element generates multiple sets of transmit-receive third beam pairs; The solution space is constructed based on the angle regions of the second transmitting end and the second receiving end. Within the solution space, initialize the particle swarm parameters and randomly initialize the initial exploration positions of n particles and the velocity of each particle; Calculate the objective function value of each particle at present. Based on the objective function values of all particles calculated at present and in the past, determine the historical optimal energy value and the corresponding optimal individual exploration point of each particle. The optimal swarm exploration point is determined based on the historical best energy value explored by the entire particle swarm. The velocity of each particle is updated based on the optimal individual exploration point and the optimal group exploration point to obtain the first exploration position of each particle in the next round; Calculate the updated objective function value for each particle based on its first exploration position in the next round; The exploration position and objective function value of each particle are iteratively updated in each round until the convergence condition is met or the maximum number of iterations is reached; When the number of iterations is reached or the convergence condition is met, the transmit-receive third beam pair corresponding to the currently searched optimal objective function value is taken as the optimal transmit-receive third beam pair and output.
2. The fast beam alignment method based on particle swarm optimization search according to claim 1, characterized in that, The specific beamwidth relationship between the first beam, the second beam, and the third beam is: first beam > second beam > third beam.
3. The fast beam alignment method based on particle swarm optimization search according to claim 1, characterized in that, Step S1 specifically includes: Control the transmitter with the first beamwidth Generate the first transmission beam ; Control the receiver to use the first beamwidth Generate and receive the first beam ; The number of first beams at both the transmitter and receiver is and .
4. The fast beam alignment method based on particle swarm optimization search according to claim 1, characterized in that, Step S2 specifically includes: The system iterates through multiple transmit-receive first beam pairs to obtain the first received energy corresponding to each transmit-receive first beam pair, and constructs the first received energy matrix. ; First receiving energy matrix The calculation process is as follows: in, Indicates the beam direction of the transmitting end is At that time, the wide beam gain of the transmitter, This indicates that the receiver beam direction is... Wide beam gain at the receiver during the time This indicates the transmission power of the transmitting end. This refers to link attenuation during signal transmission.
5. The fast beam alignment method based on particle swarm optimization search according to claim 1, characterized in that, The transmitter and receiver are respectively selected Each antenna element generates multiple sets of transmit-receive second beam pairs, specifically: Control the transmitter with the second beamwidth Generate and transmit the second beam ; Control the receiver to use the second beamwidth Generate and receive the second beam ; The number of second beams at both the transmitter and receiver is and .
6. The fast beam alignment method based on particle swarm optimization search according to claim 1, characterized in that, The first transmitter angle region and the first receiver angle region are used to traverse and scan multiple sets of transmit-receive second beam pairs to obtain the second receive energy matrix corresponding to the multiple sets of transmit-receive second beam pairs. Specifically: The system iterates through multiple transmit-receive second beam pairs to obtain the second received energy corresponding to each transmit-receive second beam pair, and then constructs the second received energy matrix. ; Second receiving energy matrix The calculation process is as follows: in Indicates the beam direction of the transmitting end is Narrow beam gain at the transmitter This indicates that the receiver beam direction is... Narrow beam gain at the receiver during the time This indicates the transmission power of the transmitting end. This refers to link attenuation during signal transmission.
7. The fast beam alignment method based on particle swarm optimization search according to claim 1, characterized in that, The convergence condition is as follows: the difference between the optimal energy values obtained from two consecutive explorations is less than the convergence threshold.
Citation Information
Patent Citations
Beam searching method used in multibeam switching antenna system and apparatus thereof
CN102394680A
ultra-dense millimeter wave D2D communication interference management method
CN109756874A
