A multi-arm equally-spaced fast beam scanning method

Abstract

The present invention discloses a multi-arm equally-spaced fast beam scanning method. First, multi-arm equally-spaced beam generation and codebook generation are performed at the transmitter / receiver. Secondly, in each round of beam scanning, for each transmit-receive multi-arm beam direction pair, the transmitter / receiver transmits / receives training symbols within one training symbol duration based on the generated equally-spaced multi-arm beam codebook. Finally, a decision is made according to the power of the received signal. After making decisions and performing set operations on the received signal power values of all rounds of scanning, the best alignment direction of the transmitter / receiver is obtained. The multi-arm equally-spaced beam scanning method can ensure high accuracy, significantly reduce the scanning overhead of the traditional exhaustive search scanning method, and at the same time, the multi-arm equally-spaced beam scanning method can exhibit better performance than other scanning methods.

Description

Technical Field

[0001] The present invention relates to the technical field of beam scanning, and particularly relates to a multi-arm equally spaced fast beam scanning method. Background Art

[0002] For millimeter-wave communication with high operating frequencies, the direct channel between an access point (AP) and the users it serves is vulnerable to severe blockage and propagation loss. To obtain a large passive beamforming gain, before implementing effective channel estimation and data transmission, the receiving end must coordinate with the transmitting beam training of the transmitting end for passive / reflection beam training in order to establish a high signal-to-noise ratio (SNR) link. Intelligent Reflecting Surfaces (IRS) or Reconfigurable Intelligent Surfaces (RIS) have become a promising low-cost technology for improving the spectral and energy efficiency of future wireless networks. In particular, by intelligently controlling signal reflection through a large number of low-cost passive reflection elements, IRS / RIS can dynamically program the radio propagation environment to achieve signal enhancement and / or interference suppression. Compared with traditional active relays, IRS / RIS has much lower hardware costs and energy consumption due to its passive reflection. IRS / RIS can be appropriately deployed to provide virtual line-of-sight (LoS) AP-IRS / RIS-user links, thus significantly improving their communication performance. However, this is challenging in practice because a large number of IRS / RIS reflection elements generate pencil-shaped sharp beams, so a large number of beam directions are required in the training codebook to cover the space of interest.

[0003] In the existing technology, the exhaustive search method selects the best beam by exhausting all possible beam directions, which requires scanning the entire beam space and will result in a long delay. The hierarchical search method performs spatial scanning according to the corresponding codebook at each stage. The scanning process is similar to the exhaustive search method, and then the best beam is found based on the received signal power, and a sub-codebook with higher resolution is subdivided in the next stage until the required spatial resolution is satisfied. The main disadvantage of hierarchical beam search is that using wide beams in the early stage will result in a reduction in beamforming gain, so spatial scanning may not be able to identify the correct wide beam at low SNR, and ultimately the best beam direction cannot be detected. Due to the severe path loss of the cascaded channel, this situation becomes worse in IRS / RIS-assisted millimeter-wave systems. In addition, since hierarchical search requires comparing and judging the scanning results of each stage before determining the scanning sub-codebook of the next stage, this will result in additional delay. Summary of the Invention

[0004] The object of the present invention is to address the problems of high training overhead and high time delay in the exhaustive beam scanning technology. Through an equispaced multi-arm beam composition method, a fast beam scanning technology that is applicable to both direct paths and IRS / RIS reflection paths is proposed. By this method, the accuracy of beam scanning is ensured, and the scanning overhead is significantly reduced, showing excellent performance.

[0005] The specific technical solution adopted by the present invention to solve its technical problems is as follows: A multi-arm equispaced fast beam scanning method, and the method comprises the following steps:

[0006] Step 1: Generation of multi-arm equispaced beams and codebook at the transmitter / receiver, and the specific steps are as follows:

[0007] Discretize the beam space into N directions and perform 1 + log2M rounds of beam scanning. In the first round, evenly divide these N directions into L multi-arm beams, and each multi-arm beam contains M = N / L sub-beam directions; starting from the second round, group the multi-arm beams in pairs, continuously divide the sub-beams contained in the multi-arm beams into two subsets, and select the subset at the odd positions of one multi-arm beam and the subset at the even positions of another multi-arm beam in the same group to form a new multi-arm beam.

[0008] Each column of the multi-arm equispaced beam codebook represents a multi-arm beam, and each column is evenly divided into M segments; each segment represents the codeword of a sub-beam within the multi-arm beam.

[0009] Step 2: Training phase. In each round of beam scanning, for each transmit-receive multi-arm beam direction pair, the transmitter / receiver transmits / receives training symbols based on the generated equispaced multi-arm beam codebook within one training symbol duration.

[0010] Step 3: Signal decision, which is only based on the power of the received signal for decision; for the received signal power in the first round of scanning, select the multi-arm beam direction pair with the maximum power value as the candidate set of the best transmit / receive direction; in the subsequent rounds of scanning, set a threshold and perform hard decision. If the power value of the multi-arm beam is greater than the threshold, it is considered to contain a useful signal, and take the intersection with the candidate set obtained in the previous round to form a new candidate set; if the power value of the multi-arm beam is less than the threshold, it is considered not to contain a useful signal, and exclude these directions from the candidate set. After making decisions and set operations on the received signal power values of all 1 + log2M rounds of scanning, obtain the best alignment direction of the transmitter / receiver.

[0011] Furthermore, in the generation of multi-arm equispaced beams, the sub-beam directions contained in each multi-arm beam are equally spaced, all being L.

[0012] Further, in the training phase, in each round of beam scanning, the number of multi-arm beams at the transmitter in this round is denoted as B1, and the number of multi-arm beams at the receiver in this round is denoted as B2. Then, there are B1·B2 transmit-receive multi-arm beam direction pairs.

[0013] Further, after 1 + log2M rounds, a total of received signals are obtained, where L1 and L2 are the numbers of multi-arm beams at the transmitter / receiver in the first round of scanning, respectively. If the transmitter / receiver is a single antenna, the transmitter / receiver does not generate multi-arm equispaced beams, and the number of multi-arm beam direction pairs corresponds to B2 / B1, and the number of received signals

[0014] Further, in the subsequent rounds of scanning for signal decision, only the multi-arm beams with the same directions as those in the candidate set obtained in the first round of scanning are subjected to decision.

[0015] Further, in the signal decision, the threshold T is set as the average of the theoretical power values of the received signal containing the useful signal and not containing the useful signal. The formula is as follows:

[0016]

[0017] where P(1, b * ) is the maximum received signal power value in the first round of scanning.

[0018] The beneficial effects of the present invention are as follows: The search method proposed by the present invention has high accuracy and low complexity.

[0019] (1) The multi-arm equispaced beam scanning method can effectively ensure the successful recognition rate, without performance loss compared with the exhaustive search method. The set number of sub-beams has an impact on the successful recognition rate, and the scanning parameters can be set according to the actual scenario requirements.

[0020] (2) The scanning complexity of the multi-arm equispaced beam scanning method is at the logarithmic level of the number of beam space directions. Compared with the traditional exhaustive search scanning method, it can significantly reduce the scanning complexity and training overhead, and effectively reduce the delay caused by beam scanning. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 is a schematic diagram of the composition of multi-arm equispaced beams;

[0022] Figure 2 is the subset combination method of multi-arm equispaced beams;

[0023] Figure 3 is the beam pattern generated by the codebook of multi-arm equispaced beams;

[0024] Figure 4 is the hierarchical search scanning beam;

[0025] Figure 5 is the instance simulation scenario diagram;

[0026] Figure 6 is the simulation result of the successful recognition rate varying with the signal-to-noise ratio;

[0027] Figure 7 is the simulation result of the scanning time varying with the number of beamspace directions. Detailed implementation manners

[0028] The following further elaborates on the detailed implementation manners of the present invention in conjunction with the accompanying drawings.

[0029] As Figure 1 shown, a multi-arm equally-spaced fast beam scanning method provided by the present invention. If the transmitting end / receiving end has an antenna array, then multi-arm equally-spaced beam generation and codebook generation are performed on the transmitting end / receiving end. If the transmitting end / receiving end is a single antenna, it is difficult for the transmitting end / receiving end to steer the antenna, and there is no need to perform multi-arm equally-spaced beam generation, codebook generation, and beam scanning on the transmitting end / receiving end. In this embodiment, the number of transmitting antennas is 1, and the number of receiving antennas is N r . The specific implementation steps are as follows:

[0030] Multi-arm equally-spaced beam generation and codebook generation: For the receiving end, the beamspace is discretized into N directions, denoted as 1, 2,..., N. In the first round, r = 1, these N directions are evenly divided into L multi-arm beams, and each multi-arm beam contains M = N / L sub-beam directions. At the same time, the sub-beams of each multi-arm beam are separated as much as possible (with an interval of L). Therefore, the sub-beam direction group of the b-th multi-arm beam in the first round is

[0031] Next, it is also necessary to scan log2M rounds. Starting from the second round, r = {2, 3,..., 1 + log2M}, and each multi-arm beam in the first round is evenly divided into 2 r-1 subsets [B(1, b)] 1：u(r) , [B(1, b)] u(r)+1∶2u(r) ,..., [B(1, b)] M-u(r)+1∶M , and each subset contains adjacent directions; the multi-arm beams in the first round are divided into combinations, with a spacing of multi-arm beams, that is select the subsets at odd positions in B(1, l) and the subsets at even positions in

[0032]

[0033] As Figure 1 , when N = 16, in the first round, it is evenly divided into L = 4 multi-arm beams, each multi-arm beam contains M = 4 sub-beams, and the interval between sub-beams is 4. In the beam generation from the second round to the 1 + log2M = 3rd round, the multi-arm beams are divided into combinations, with an interval of 2 multi-arm beams, that is, the 1st and 3rd multi-arm beams are in a group, and the 2nd and 4th multi-arm beams are in a group. As Figure 2 , for the combination of the 1st and 3rd multi-arm beams, in the second round, each multi-arm beam is evenly divided into 2 subsets, and the subset at the odd position (1) of the 1st multi-arm beam and the subset at the even position (2) of the 3rd multi-arm beam are taken to form the multi-arm beam in the second round; in the third round, each multi-arm beam is evenly divided into 4 subsets, and the subsets at the odd positions (1, 3) of the 1st multi-arm beam and the subsets at the even positions (2, 4) of the 3rd multi-arm beam are taken to form the multi-arm beam in the third round.

[0034] According to the codebook of N directions where represents the codeword in the nth direction, the corresponding multi-arm equally spaced beam codebook can be constructed where the multi-arm beam codebook each column represents a multi-arm beam, denoted as

[0035]

[0036] The sub-beams in each multi-arm beam correspond to the codewords intercepted from L = N r / M adjacent elements from the column corresponding to the direction in the codebook r . The codeword of the lth multi-arm beam in the r' = {1, 2, 3,..., 1 + log2M}th round can be expressed as where η m represents the discrete direction value corresponding to the mth sub-beam contained in the multi-arm beam B(r', l). As Figure 3 is the beam pattern generated according to the first-round multi-arm equally spaced beam codebook.

[0037] In the training stage, the receiving end receives training symbols within one training symbol duration based on the generated equally spaced multi-arm beam codebook. After 1 + log2M rounds, a total of received signals are obtained.

[0038] Signal decision: Only make a decision based on the power P(r, b) of the received signal. For the power of the received signal scanned in the first round, select the multi-arm beam with the largest power value as the candidate set of the best direction, that is where In the subsequent rounds of scanning, only the multi-arm beams containing the same direction need to be judged, denoted as B(r, l(r)), where Set the threshold Perform a hard decision. If the power value of the multi-arm beam is greater than the threshold, it is considered to contain a useful signal, and the intersection is taken with the candidate set obtained in the previous round to get a new candidate set; if the power value of the multi-arm beam is less than the threshold, it is considered not to contain a useful signal, and these directions are excluded from the candidate set , that is

[0039]

[0040] After judging the received signal power values for all 1 + log2M rounds of scanning, the best alignment direction at the receiving end can be obtained.

[0041] This embodiment also provides a hierarchical search scanning method as a comparison, and its specific implementation steps are as follows: Figure 4 , discretize the beam space into N directions with a resolution of In the first layer, construct two wide beams. The first wide beam contains the direction The second wide beam contains the direction The transmitter sends training symbols to the receiver. The receiver receives with one wide beam during each training symbol duration. According to the power of the received signal, select the wide beam with a larger power value. In this example, it is the first wide beam

[0042] In the second layer, continue to divide the wide beam selected in the first round into two beams, which respectively contain the directions and The transmitter sends training symbols to the receiver. The receiver receives with one beam during each training symbol duration. According to the power of the received signal, select the beam with a larger power value. In this example, it is the first beam In each subsequent round, divide the beam obtained in the previous round into two for scanning until the beam can no longer be subdivided, and then the best alignment direction is obtained.

[0043] The functions and effects of the present invention are further illustrated and demonstrated through the following simulation experiments:

[0044] Simulation conditions

[0045] As Figure 5 , set the number of users to K = 3, and there is a communication with an IRS / RIS-assisted system. For simplicity, we consider a group of IRS / RIS deployed in the y-z plane and arranged in parallel on the y-axis. The IRS / RIS is equipped with N v = 32, N h= 32 reflection units, the IRS / RIS is located at the center of the coordinate axes, i.e., (0, 0, 0) meters (m). The users are distributed on the x-y plane, so the elevation angle from the IRS / RIS to each user Since the beamforming in the vertical direction of the IRS / RIS does not depend on the user's position, only the beam alignment between the horizontal array of the IRS / RIS and the users needs to be considered. The base station centered at (16, 16, 0) meters consists of N A = 64 antennas, and the antenna spacing

[0046] Simulation results

[0047] Figure 6 shows the relationship between the beam successful identification rate and the signal-to-noise ratio. The accuracy of the multi-arm equidistant beam scanning method has a great relationship with the number of sub-beams of the multi-arm beam. When the number of sub-beams M is larger, the leakage interference between sub-beams is larger, the interference between different multi-arm beam subsets is larger, and the probability of successful identification is lower. When the number of sub-beams M = 2, compared with the traditional exhaustive search scanning method, the performance loss is smaller, especially when the signal-to-noise ratio is greater than 0 dB. At the same time, the hierarchical search scanning can also guarantee the accuracy to a certain extent. Figure 7 shows the relationship between the beam scanning time and the number of directions of beam space discretization. Both the multi-arm equidistant beam scanning method and the hierarchical search scanning method are at the logarithmic level of the number of directions, which greatly reduces the scanning time and training overhead compared with the traditional exhaustive search scanning method. At the same time, since the hierarchical search in this example adopts the method of sequential scanning for each receiver in this example, the scanning time is proportional to the number of users, while the multi-arm equidistant beam scanning method allows simultaneous scanning of multiple users. Therefore, the multi-arm equidistant beam scanning method can achieve the lowest scanning complexity.

[0048] It should be noted that the above embodiments are only used to explain the present invention, rather than to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A multi-arm equally-spaced fast beam scanning method, characterized in that, The method includes the following steps: Step 1: Generation of multi-arm equally-spaced beams and codebook at the transmitter / receiver. The specific steps are as follows: Discretize the beam space into N directions and perform 1 + log2M rounds of beam scanning. In the first round, evenly divide these N directions into L multi-arm beams, and each multi-arm beam contains M = N / L sub-beam directions. Starting from the second round, group the multi-arm beams in pairs. Continuously divide the sub-beams contained in the multi-arm beams into two subsets evenly. Select the subset of odd positions of one multi-arm beam and the subset of even positions of another multi-arm beam in the same group to form a new multi-arm beam. Each column of the multi-arm equally-spaced beam codebook represents a multi-arm beam, and each column is evenly divided into M segments; each segment represents the codeword of a sub-beam within the multi-arm beam. Step 2: Training phase. In each round of beam scanning, for each transmit-receive multi-arm beam direction pair, the transmitter / receiver sends / receives training symbols based on the generated equally-spaced multi-arm beam codebook within one training symbol duration. Step 3: Signal decision, which is made only based on the power of the received signal. For the received signal power in the first round of scanning, select the multi-arm beam direction pair with the maximum power value as the candidate set for the best transmit / receive direction. In the subsequent rounds of scanning, set a threshold and perform hard decision. If the power value of the multi-arm beam is greater than the threshold, it is considered to contain useful signals, and take the intersection with the candidate set obtained in the previous round to form a new candidate set; if the power value of the multi-arm beam is less than the threshold, it is considered not to contain useful signals, and exclude these directions from the candidate set. After making decisions and set operations on the received signal power values of all 1 + log2M rounds of scanning, obtain the best alignment direction of the transmitter / receiver.

2. The multi-arm equally-spaced fast beam scanning method according to claim 1, characterized in that In the generation of multi-arm equally-spaced beams, the sub-beam directions contained in each multi-arm beam are equally spaced, all being L.

3. A multi-arm equally-spaced fast beam scanning method according to claim 1, characterized in that, In the training phase, in each round of beam scanning, denote the number of multi-arm beams of the transmitter in this round as B1, and the number of multi-arm beams of the receiver in this round as B2. Then there are B1·B2 transmit-receive multi-arm beam direction pairs.

4. A multi-arm equally-spaced fast beam scanning method according to claim 3, characterized in that, After 1 + log2M rounds, a total of received signals are obtained, where L1 and L2 are the number of multi-arm beam directions scanned by the transmitter / receiver in the first round; if the transmitter / receiver is a single antenna, the transmitter / receiver does not generate multi-arm equally spaced beams, and the number of multi-arm beam direction pairs corresponds to B2 / B1, and the number of received signals 5. A multi-arm equally spaced fast beam scanning method according to claim 1, characterized in that, In the subsequent rounds of scanning for signal decision, only make decisions on the multi-arm beams that contain the same directions as the candidate set obtained in the first round of scanning.

6. A multi-arm equally spaced fast beam scanning method according to claim 1, characterized in that, In the signal decision, the threshold T is set as the average of the theoretical power values when the received signal contains useful signals and does not contain useful signals. The formula is as follows: Among them, P(1, b * ) is the maximum received signal power value in the first round of scanning.

Images