A SUPPORT VECTOR APPROACH IN SEGMENTED REGRESSION FOR MAP-ASSISTED NON-COOPERATIVE SOURCE LOCALIZATION
Hao Sun, Weiming Huang, Xianghao Yu, Junting Chen
Abstract
This paper presents a non-cooperative source localization approach based on received signal strength (RSS) and 2D environment map, considering both line-of-sight (LOS) and non-line-of-sight (NLOS) conditions. Conventional localization methods, e.g., weighted centroid localization (WCL), may perform bad. This paper proposes a segmented regression approach using 2D maps to estimate source location and propagation environment jointly. By leveraging topological information from the 2D maps, a support vector-assisted algorithm is developed to solve the segmented regression problem, separate the LOS and NLOS measurements, and estimate the location of source. The proposed method demonstrates a good localization performance with an improvement of over 30% in localization rooted mean squared error (RMSE) compared to the baseline methods.
BibTeX
@inproceedings{icassp2026_asupportvectorap,
title = {A SUPPORT VECTOR APPROACH IN SEGMENTED REGRESSION FOR MAP-ASSISTED NON-COOPERATIVE SOURCE LOCALIZATION},
author = {Hao Sun and Weiming Huang and Xianghao Yu and Junting Chen},
booktitle = {ICASSP 2026},
year = {2026}
}