Abstract:Wireless sensor networks have the characteristics of large scale, self-organization, reliability, data-centricity, integration, etc., and are widely used in military, medical, mine monitoring, safety production and other fields. However, the existing non-ranging positioning algorithms for wireless sensor networks suffer from large positioning errors. Aiming at this problem, a DV-Hop (distance vector-hop) localization algorithm based on the global artificial fish swarm algorithm optimization was proposed, namely DEWF-D localization algorithm. In the algorithm, the error-prone steps of the DV-Hop algorithm were optimized in the non-ranging positioning algorithm, and finally more accurate positioning coordinates were obtained by reducing the errors in the algorithm process. First, messages were transmitted by the beacon nodes with two different communication radii, and the number of hops was precisely processed to reduce the error caused by the number of hops; then, the average distance per hop was calculated using the minimum mean square error criterion and the error weighting method; Finally, the global artificial fish swarm algorithm was used to replace the trilateration method for coordinate calculation. Simulation results showed that, compared with the DV-Hop algorithm and other algorithms, the positioning accuracy of the proposed algorithm is respectively improved by 28.3%, 6.9%, and 12.5% under different beacon node densities; and under different communication radii, the positioning accuracy is improved by 24.4%, 7.6%, 14.8%, respectively. It was demonstrated that the DEWF-D algorithm can effectively improve the positioning accuracy and solve the problem of large positioning errors in previous positioning algorithms.