The Study On Relative Localization Algorithm Of Wireless Sensor Networks

Wireless sensor networks have great prospect in the applications of military affairs, circumstance observation, disaster relief operation, dangerous area domination, etc. One of the key qualifications for these applications is localization of the targets needing monitoring and the network node. And some routing scheme, information of network coverage, load balance, topology control, and so on also depend on the location information of the nodes in networks. All these indicate that localization is a fundamental and critical issue for the application, operation and management of wireless sensor networks, and the study for self-positioning of wireless sensor networks possesses great significance in the theory and practice, will become an important part in the research area.The localization of wireless sensor networks can be generalized in two parts that are in correlation, they are localization of the nodes in the networks and that of the targets needing monitoring, the later relies on the first. The relative localization algorithm——clustering-based SPA algorithm(scalable and distributed GPS free positioning for sensor networks), is a classical algorithm that was presented for the localization of nodes in wireless sensor networks with no beacon nodes and advance configurated infrastructure. It is well scalable. But the disadvantages are heavy communication cost and corresponding influence on enegy cost, network communication load and positioning time.Two amelioration methods were presented in this paper――optimized schemes for the choice of assistant nodes in the construction of the Local Coordinate System(LCS) and for the combination of LCSs in the construction of Network Coordinate System (NCS). It can effectively reduce the communication cost in the construction courses of LCS and NCS with no influence on the localization coverage. The simulation based on J-Sim showed the improved algorithm exhibits lower rate of localization error and communication cost comparing with the original, and the localization coverage remains the same. It's proved that the presented amelioration schemes are feasible and the improved algorithm is more applicable.