Majorization–Minimization-Based Target Localization Problem from Range Measurements

Kaifeng Gao,Jiang Zhu,Zhiwei Xu
DOI: https://doi.org/10.1109/lcomm.2019.2963834
IF: 3.5529
2020-01-01
IEEE Communications Letters
Abstract:In this letter, received signal strength (RSS) based target localization is studied. First we start with the case of known emitted power of target. The least squares (LS) approach is adopted, and the object function is shown to be decomposed as a convex quadratic plus a concave term. Consequently, a Majorization-Minimization (MM) based approach from range measurements termed R-MM is proposed, which iteratively finds the local optimum through simple operations. To further improve the performance, unconstrained squared-range (USR)-based LS estimation is used as an initial point for R-MM, named R-MM-USR. Then R-MM-USR is extended to solve the target localization under unknown emitted power through alternating minimization. For both cases Cramer Rao bound (CRB) is derived. In addition, the computation complexity of R-MM-USR is lowest among the high accuracy approaches. Besides, numerical simulations are conducted to demonstrate the near optimal performance, compared with CRB and other state-of-art methods.
What problem does this paper attempt to address?