In classical three-dimensional WSN localization algorithms,the Maximum Likelihood Estimation(MLE)method encounters issues with non-invertible matrices.To address this concern,a three-dimensional WSN localization algorithm based on multi-strategy optimization is proposed.Firstly,a Sine mapping is introduced to apply chaotic mapping to the initial population of the fish swarm,enhancing the diversity and uniformity of the initial population.Secondly,a combination of non-linear convergence factor and adaptive weight strategy is employed to optimize the iterative updates of fish positions,preventing the algorithm from converging to local optima and further improving search speed and optimization accuracy.Finally,a multi-strategy enhanced fish swarm optimization algorithm is employed in three-dimensional space to calculate the positions of each unknown node,resolving the issue of non-invertible matrices and effectively reducing the computational errors of the target node positions.Experimental results demonstrate that the newly proposed localization algorithm,classical three-dimensional localization algorithm,three-dimensional weighted DV-Hop localization algorithm,and grey wolf optimization-based three-dimensional localization algorithm exhibit average localization errors of 13%,73%,30%and 17%,respectively.
关键词
三维DV-Hop/金枪鱼群算法/Sine混沌映射/非线性收敛因子/自适应权重
Key words
3D DV-Hop/tuna swarm optimization/sine chaotic mapping/nonlinear convergence factor/adaptive weight
维普
万方数据