Abstract
Aiming at the problem that the optimal parameters of the least square support vector regression (LSSVR) localization model in Internet of Things (IoT) are difficult to determine, a positioning method based on the improved particle swarm optimization (PSO) algorithm is proposed. First, the positioning model is constructed by LSSVR, then the PSO algorithm is improved by adaptively adjusting the inertia weights and the learning factors, and finally the improved PSO algorithm is used to search the optimal parameters of the LSSVR positioning model, which avoids the blindness of the parameter search. The simulation results show that the positioning accuracy of the proposed algorithm is improved by 25.9% and 19.7%, respectively, compared with the LSSVR and PSO‐LSSVR algorithms, and has better positioning stability and real‐time performance.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call