Research Of Three-Dimensional Localization Algorithm For Wireless Sensor Networks

Wireless sensor networks (WSN) is a self-organizing system that usually comprises large numbers of wirelessly connected sensor nodes, and is one of the current research focuses. For the large scale sensor networks, it is usually not feasible for the manual measurement or configuration to get the node coordinates. Therefore, there is a necessity to study on the self-positioning technology for the sensor nodes.So far, according to various application requirements, a variety of positioning methods for WSN have been proposed. These methods are mainly divided into two categories:range-based and range-free. However, most of these methods are for the two-dimensional environment, but few are for the three-dimensional environment.Firstly, this paper introduces the basic concepts of WSN localization, and discusses the factors and evaluation criteria of three-dimensional localization, and then deeply analyzes two classical range-based three-dimensional localization algorithms for WSN:the multilateration measurement localization algorithm and the least square localization algorithm. On the base of these researches, the paper proposes two novel localization algorithms for WSN:the three-dimensional localization algorithm for WSN based on the BFGS optimization and the three-dimensional localization algorithm for WSN based on the particle swarm optimization (PSO). The aim is to improve the positioning accuracy.By defining a novel objective function for three-dimensional localization, the three-dimensional localization algorithm for WSN based on the BFGS optimization transforms the localization work into an unconstrained optimization problem. The BFGS optimization algorithm, which is a traditional optimization method, is successfully and effectively adopted for the three-dimensional localization. The experimental results show that the positioning accuracy of the proposed algorithm is higher than the one of the least square localization algorithm, and the positioning time of the proposed algorithm is less than the one of the three-dimensional localization algorithm for WSN based on PSO, but is more than the one of the least square localization algorithm.According to a direct proportion relationship of the measured distance with the measuring errors, an improved three-dimensional localization objective function with weighted the measured distance is defined in the three-dimensional localization algorithm for WSN based on PSO, and PSO method is adopted to solve the optimal problem of the objective function. In the process of solving the measured distance equations, the minimum distance equation is chosen as the common subtraction equation for reducing the order of the equations. The simulation results show that the process can reduce localization errors. And the final experimental results indicate that the proposed algorithm has a higher accuracy than the multilateration measurement localization algorithm, the least square localization algorithm, the three-dimensional localization algorithm for WSN based on the BFGS optimization and the classical PSO localization algorithm, but it is at cost of more localization time.In addition, the simulation results also indicate the effects of the amount and the distribution of the beacon nodes:as the number of the beacon nodes increases, the positioning accuracy increases gradually, until the number of the beacon nodes is large enough, the positioning accuracy curve flattens out; Compared with the random distribution of the beacon nodes, the performance under the marginal distribution is better.