A RIS assisted near field wireless localization method based on GABP neural network
Reconfigurable intelligent surface is a potentially key technology in 6G,which can be deployed in wireless communication systems to assist base stations in localizing users and enhancing localization performance.In the millimeter wave frequency band,due to the higher frequency,the array aperture and number of reflection units of RIS panels are larger,expanding the near-field range,and users are likely to be in the near-field region of RIS.Therefore,the RIS-assisted wireless communication system was considered to localize the user under near-field conditions,and the user localization problem was modeled as a parameter estimation problem,and the maximum likelihood estimation was used to achieve the estimation of the user's 3D coordinate parameters.To solve the established maximum likelihood estimation problem,a GABP algorithm,which combined the genetic algorithm and back propagation neural network,was proposed.Simulation results demonstrated that the proposed GABP algorithm was more computationally efficient than traditional genetic algorithm.
reconfigurable intelligent surfacenear field positioningmaximum likelihood estimationGABP algorithm
NETL
NSTL
万方数据
