Optimization of WSN localization algorithm based on improved multi-strategy seagull algorithm
Xiuwu Yu,Yinhao Liu,Yong Liu,Liu, Yinhao
DOI: https://doi.org/10.1007/s11235-024-01137-2
2024-04-12
Telecommunication Systems
Abstract:This paper proposes a multi-strategy modified seagull algorithm to optimize DV-Hop localization algorithm (DISO) to improve the precision of non-range-ranging localization algorithm in wireless sensor networks. Firstly, the algorithm analyzes the causes of errors in the positioning of the traditional non-ranging location algorithm DV-Hop, and improves these steps. Among them, the communication area of anchor nodes is divided by different radii, so as to reduce the influence of distance on hop number. The node distribution is stochastic, so the mean square error is used instead of the unbiased estimation, and the weight is introduced to calculate the average jump distance, which reduces the error caused by the random distribution of nodes. Secondly, the objective function optimization method is used to replace the trilateral measurement, and the improved seagull optimization algorithm is used for iterative optimization. Finally, the seagull optimization algorithm is modified in view of its shortcomings. The chaotic mapping was used to initialize the seagull population and increase its diversity. The flight parameters of seagull and the position update methods of the worst and best seagull are improved, and the optimization ability of the algorithm is improved by combining levy flight mechanism and T distribution variation strategy. The simulation results show that the initial population distribution of DISO algorithm is more uniform, which establishes a basic advantage for the subsequent optimization. Keeping the other parameters consistent, DISO algorithm has higher positioning accuracy than other comparison algorithms, no matter changing the number of anchor nodes or the total number of nodes or changing the communication radius. The positioning errors of DISO algorithm are reduced by 45.63%, 17.17%, 22.61% and 11.68% compared with DV-Hop algorithm and other comparison algorithms under different number of anchor nodes. Under different total number of nodes, the positioning error is reduced by 49.91%, 20.81%, 35.80% and 9.20%. Under different communication radius, the positioning error is reduced by 55.47%, 21.07%, 24.84% and 13.11%. It is proved that DISO algorithm has more accurate localization results.
telecommunications