Objectives  At present, in the research on the optimization of wireless sensor network (WSN) deployment under the application of indoor positioning, most of studies are generally carried out under the ideal conditions without obstacles, and the impact of obstacles on WSN is rarely considered. To consider the influence of obstacles for WSN deployment, we propose a multi-objective optimization method of WSN topology considering obstacles under the application of indoor positioning using received signal strength indication sensors.

  Methods  Firstly, a concept of effective coverage and beacon node deployment model is proposed based on the indoor positioning algorithm and sensor coverage model. Secondly, after the analysis of obstacle perception model and beacon node deployment strategy, a multi-objective optimization model of sensor deployment considering obstacles is proposed.Finally, the non-dominated sorting genetic algorithm Ⅲ is introduced to solve this optimization model. Three optimization objectives of all individuals, effective coverage, number of sensors, and convex hull area, as well as constraints considering obstacles and WSN topology rationality are calculated.

  Results  The optimization method without considering obstacles is used as the comparison method, and three uniform sensor deployments are compared with the numerical simulation results of the proposed method.Regular triangle, square and hexagon are the mosaic shapes for the uniform sensor deployment.In the 1 000 generation, the effective coverage rate of the proposed method increased by 9.62% compared with the comparison method and increased by 52.7%, 112.1% and 16.6% respectively compared with the uniform deployment of regular triangle, square and regular hexagon.

  Conclusions  The method proposed in this paper not only has advantages in improving the effective coverage rate and accelerating convergence of the optimization algorithm, but also increases the effective coverage rate compared with the traditional uniform deployment of sensors. Therefore, the proposed method will be helpful to improve the accuracy of indoor positioning.