RSS-Based Localization of Multiple Directional Sources With Unknown Transmit Powers and Orientations

• Main Authors: Peiliang Zuo, Tao Peng, Kangyong You, Wenbin Guo, Wenbo Wang
• Format: Article
• Language: English
• Published: IEEE 2019-01-01
• Series: IEEE Access
• Subjects: Multiple directional sources; localization; CRLB; unknown orientation; maximum likelihood
• Online Access: https://ieeexplore.ieee.org/document/8753566/

Received-signal-strength (RSS)-based localization has received widespread attention recently. Due to the simple acquisition of the RSS measurements, the adequate inexpensive sensors in sensor networks are capable of providing the information needed for the positioning of multiple target sources. However, few studies have focused on the RSS-based localization of multiple directional sources that are common in reality. Based on the deduced parametric Optimal Maximum Likelihood (OML) solution, this paper proposes three new grid search-based algorithms, namely Alternating Projection (i.e., OMLAP) algorithm, Expectation-Maximization like (i.e., OMLEM) algorithm, and Particle Swarm Optimization (i.e., OMLPSO) algorithm. They can be utilized for estimating the transmit powers, locations, and orientations of multiple directional sources. Combining the interpolation process and proposed power threshold setting method, the search space is obviously reduced. Moreover, the corresponding Cramer-Rao lower bounds (CRLB) are also derived to characterize the estimation accuracy of the algorithms. Both the scenarios with different Signal-to-Noise Ratios (SNRs) and the scenarios with different sensor quantities are considered in the simulation, and the results demonstrate the effectiveness of the proposed algorithms and indicate that they are suitable for the parameter estimation of multiple directional sources.