An improved bearing-only based Semidefinite Programming algorithm for large scale WSN node localization

Tianxu Du,Zhi Wang
DOI: https://doi.org/10.1109/ICIEA.2014.6931216
2014-01-01
Abstract:Node localization is one of the most challenging issues in large scale sensor networks. Few bearing-only based methods are used to locate nodes in a large scale system. Existing bearing-only based Semidefinite Programming (BOSDP) method consumes a lot of energy and it may not yield good results if the beacons are placed randomly. Moreover, the BOSDP approach cannot estimate the locations and orientations of unknown nodes simultaneously. In this paper, we propose an improved Bearing-Only Semidefinite Programming (IBOSDP) method to estimate the positions and orientations of unknown nodes in large scale WSNs. By introducing a new triangular linear equation constraint, the proposed IBOSDP algorithm could solve the node localization problem effectively. Different from the prior work, only bearing information is required to estimate the unknown nodes, which could minimize the node's power consumption. Moreover, the localization accuracy of the new IBOSDP is unrelated to the deployment of beacons in the network. Importantly, the orientation information can also be calculated by the proposed algorithm. Extensive simulation results demonstrate that the new algorithm could achieve higher localization accuracy than the BOSDP when the beacons are randomly placed.
What problem does this paper attempt to address?