Search

CN-122026969-A - Ultra-large-scale MIMO system beam focusing method based on self-adaptive time delay-phase structure

CN122026969ACN 122026969 ACN122026969 ACN 122026969ACN-122026969-A

Abstract

The invention provides a beam focusing method of a super-large-scale MIMO system based on a self-adaptive time delay-phase structure, which relates to the technical field of radio communication and comprises the following steps of constructing the self-adaptive time delay-phase structure suitable for the super-large-scale MIMO system, establishing a near-field beam focusing joint optimization mathematical model, solving a full-digital optimal beam focusing matrix through maximum ratio transmission, converting a switch matrix optimization problem into a bipartite graph matching problem, solving the optimal matching matrix through a Hungary algorithm, solving a phase shift matrix through a Li Manpu conjugate gradient method, optimizing a time delay matrix through a gradient descent method, optimizing a digital precoding matrix through a least square method, and giving a base station beam focusing matrix through alternate iteration to complete the beam focusing of the super-large-scale MIMO system. The structure provided by the invention can realize dynamic sharing with the antenna array element by utilizing the switch network, thereby effectively enhancing the self-adaptive capacity of the mixed precoding to the channel environment.

Inventors

  • LIU FULAI
  • Suo Luyao

Assignees

  • 东北大学

Dates

Publication Date
20260512
Application Date
20260414

Claims (7)

  1. 1. The ultra-large-scale MIMO system beam focusing method based on the self-adaptive time delay-phase structure is characterized by comprising the following steps of: Step 1, constructing a self-adaptive time delay-phase structure suitable for a super-large-scale MIMO system, and establishing a near-field beam focusing joint optimization mathematical model; step 2, solving a full digital optimal beam focusing matrix through maximum ratio transmission; Step 3, converting the switch matrix optimization problem into a bipartite graph matching problem, and solving an optimal matching matrix by using a Hungary algorithm; Step 4, solving a phase shift matrix by Li Manpu conjugate gradient method; Step 5, optimizing a time delay matrix by using a gradient descent method; Step 6, optimizing the digital precoding matrix by using a least square method; and 7, alternately iterating the steps 3-6 to give a base station beam focusing matrix, and finishing the beam focusing of the ultra-large-scale MIMO system.
  2. 2. The method for focusing a beam of a super-massive MIMO system based on an adaptive delay-phase structure according to claim 1, wherein in step 1, the adaptive delay-phase structure is A root radio frequency chain, Individual delays By means of phase shifters Is connected to A root antenna, wherein, The number of the time delays connected with each radio frequency chain is represented; representing the number of switches; Representing the near field beam focusing joint optimization mathematical model by a minimized residual criterion as: Wherein, the An all-digital beam focusing matrix is shown, Representing the maximum transmit power; , representing the analog phase-shifting matrix, A phase shift vector representing the phase shifter; representing a time-delay matrix, A delay vector representing the delay of the delayer, Represent the first The time delay of the time delay unit, Representing the maximum time delay that the delayer can compensate; Representing a matrix of switches, Representing the baseband digital precoding vector, Represent the first The elements.
  3. 3. The method for focusing a beam of a super-massive MIMO system based on an adaptive delay-phase structure according to claim 1, wherein in step 2, an optimal beam focusing matrix is obtained by maximum ratio transmission using a near-field channel matrix Wherein Representing the channel matrix between the base station and the user.
  4. 4. The method for focusing a beam of a super-massive MIMO system based on the adaptive delay-phase structure according to claim 1, wherein said step 3 comprises the steps of: step 31, the switch matrix optimization problem is converted into: ; ; step 32, developing the objective function to minimize the problem Will (i) be Defining as a weight matrix; Step 33, when Is that Is provided with an integer multiple of Extension of Is that , wherein, The weight matrix is represented by a matrix of weights, Representing the expanded weight matrix, converting the problem into Solving the minimum weight matching on the cost matrix; Step 34, distributing by using Hungary algorithm to construct a selection matrix 。
  5. 5. The method for focusing a beam of a super-massive MIMO system based on the adaptive delay-phase structure according to claim 1, wherein said step 4 comprises the steps of: Step 41, phase shift matrix parameters form a complex round popularity The optimization problem is translated into: Step 42, calculating European gradient Will (i) be Orthographic projection onto manifold Obtaining the corresponding Riemann gradient The method comprises the following steps: ; step 43, updating the search direction: ; Wherein, the Representing a vector transfer function; the parameters of Polak-Ribiere are indicated, Representing Li Manpu parameters; Step 44, executing a pullback operator To update the updated point Mapping to the sum origin In the same manifold space: ; ; Wherein, the Representing manifold Midpoint (midpoint) Is a tangential space to the plane of the lens; Representing the Armijo backtracking line searching step length; And Is to satisfy inequality Is a minimum non-negative integer of (2).
  6. 6. The method for focusing a beam of a super-massive MIMO system based on the adaptive delay-phase structure according to claim 1, wherein said step 5 comprises the steps of: step 51, converting the time delay matrix optimization problem into: ; step 52, calculating the derivative with respect to the delay parameter: ; Wherein, the ; Step 53, updating the delay parameter ; Representing the learning rate.
  7. 7. The method for focusing a beam of a super-massive MIMO system based on the adaptive delay-phase structure according to claim 1, wherein in step 6, a least squares solution is used to solve the digital precoding vector 。

Description

Ultra-large-scale MIMO system beam focusing method based on self-adaptive time delay-phase structure Technical Field The invention relates to the technical field of radio communication, in particular to a beam focusing method of a super-large-scale MIMO system based on a self-adaptive time delay-phase structure. Background With the demands of the sixth generation mobile communication (6 th Generation Mobile Networks, 6G) for spectrum efficiency and energy efficiency in the future, ultra-large-scale MIMO (Multiple Input Multiple Output, MIMO) is one of the key technologies for future wireless communication as a further evolution of massive MIMO technology. By deploying the antenna of the ultra-large scale array, the beam space resolution capability is improved while higher beam gain is obtained, so that higher spectrum efficiency is realized. However, the large array aperture increases the rayleigh distance to several hundred meters, and future 6G wireless communication may occur in a near field region except a far field, and on the other hand, signal delay between array elements when a wireless electromagnetic wave signal reaches a base station antenna array becomes non-negligible compared with a communication symbol period, so that a beam splitting phenomenon may occur, and communication rate and quality may be reduced. For this reason, beam focusing is considered as one of key technologies for future wireless communication because it can focus signals on specific positions, effectively reduce interference and attenuation, and improve wireless transmission performance. In addition, in order to eliminate the beam splitting phenomenon, a True delay (True TIME DELAY, TTD) device compensates for the delay by generating a phase shift proportional to the product of the time delay and the subcarrier frequency to improve signal transmission performance. Therefore, the research on the beam focusing method of the ultra-large-scale MIMO system based on TTD plays a vital role in the development of future 6G wireless communication. The time delay-phase structure based on TTD can compensate time delay, eliminate signal interference and improve signal transmission gain by optimizing a time delay matrix, a phase shift matrix and a digital precoding matrix. The patent application number 202510522582.8 discloses a terahertz antenna structure based on a mixed single-double layer delay line and a precoding method, wherein the method adopts a single-layer delay line structure in the middle part of an antenna needing to provide small time delay, and adopts a double-layer delay line structure in the two end parts needing to provide large time delay. The patent document with application number of 202311044911.X discloses a terahertz beam forming structure and method based on dynamic grouping of subarrays and true delay, the method is divided into antennas of subarrays and radio frequency links connected with the subarrays, wherein the radio frequency links and the subarrays are grouped, and the radio frequency links and the subarrays are fully connected through the true delay and phase shifters inside the grouping. At present, the related structure mainly comprises time delay-phase and antenna fixed connection, and the connection mode is difficult to be adaptively adjusted according to the channel state change so as to meet the diversified communication requirements of users. The beam focusing optimization method based on manifold optimization and graph matching theory becomes a feasible means for solving the problems of multi-parameter coupling characteristics, constant mode constraint of a time delay matrix and a phase shift matrix and 0-1 integer programming of a switch matrix of the ultra-large-scale MIMO system. On one hand, the current manifold optimization algorithm mainly calculates the Euclidean gradient to obtain an accurate Riemann gradient, and then adopts a conjugate gradient descent algorithm and the like to solve the asymptotic optimal solution of the optimization problem. The patent with application number '202310309860.2' discloses a 'large-scale MIMO downlink pre-coding manifold optimization method', which converts the constrained optimization problem in Euclidean space into unconstrained optimization problem in manifold space based on the precoder set meeting total power constraint, each user power constraint and antenna-by-antenna power constraint on different Li Manzi manifolds. Based on this, a Riemann conjugate gradient method is provided to design precoders on manifolds that meet different constraints. The patent with application number 202310933943.9 discloses a conjugate gradient beam forming generation method based on manifold optimization, which converts a problem model for minimizing interference signal energy into a problem model on a complex manifold, and solves the problem model on the complex manifold by a gradient descent method. On the other hand, the current graph matching theory is considered