OFDM Source Localization Algorithm for Multiple Arrays in Multipath Environments via Combining Enhanced Spatial Smoothing and Clustering

When addressing the problem of signal source localization in multipath environments, traditional subspace‑based algorithms often suffer from rank deficiency. This issue severely degrades the performance of angle-of-arrival (AOA) estimation. This paper focuses on the localization of orthogonal frequency division multiplexing (OFDM) sources in multipath environments. A low-complexity two-step localization method based on spatial smoothing is put forward. First, enhanced spatial smoothing preprocessing on the signal subspace (ESSP-SS) is performed to overcome the covariance matrix's rank deficiency. Next, the accurate estimation of AOAs is realized by using the multi-carrier property. Then, the line-of-sight (LOS) is identified by exploiting the power differences between LOS and non-line-of-sight (NLOS) paths. Finally, the clustering algorithm fuses LOS angle estimates from multiple antenna arrays to determine the source position. The simulation outcomes indicate that the proposed method significantly improves positioning precision in multipath scenarios.