MM-APS-SCENT: A Robust Parameter Estimation Enabled Target Transmitter Localization Framework

To achieve better localization accuracy in a complex electromagnetic environment with multiple modulated signals, in this paper, an end-to-end joint analysis framework is presented. This framework is target-transmitter-localization-oriented and apriori-signal-parameter-enabled. Specifically, on the one hand, given that the localization accuracy of existing algorithms is generally influenced by multiple factors such as calculated complexity, channel fading, and unavailable signal parameters in a complex electromagnetic environment, signal parameter estimation is considered and employed to overcome above difficulties for enhancement of the localization accuracy. To facilitate signal parameter estimation, mixed signals are transformed into a time-frequency image using smoothed-pseudo-Wigner-Vile distribution and image processing. On the other hand, based on the Min-Max algorithm and area partition strategy, an improved weighted centroid localization algorithm, MM-APS-SCENT, is proposed. For the MM-APS-SCENT algorithm, we further devise an area partition criterion to fulfill the superior localization accuracy. Finally, numerical results prove the correctness of theoretical analysis and show effectiveness, reliability, and robustness of the proposed MM-APS-SCENT algorithm compared with Min-Max, E-Min-Max, and weighted-cent algorithms.