Unlocking the dynamic potential: Next-gen DOA estimation for moving signals via BSCS with adaptive weighted Kalman filter in 6G networks

In the pursuit of enabling the unprecedented capabilities of the sixth-generation (6G) technology, this paper endeavours to advance the state-of-the-art in the direction of arrival (DOA) estimation techniques for dynamic scenarios. This work introduces an innovative adaptive compressive sensing (CS) technique, termed the BS weighted-CSKF algorithm. This approach integrates CS principles with a CS-oriented Kalman filter (KF), providing enhanced adaptability to fluctuating and moving source signals. Comparative analysis against existing CS-based DOA estimation methods demonstrates the superior performance of the proposed algorithm, particularly in low signal-to-noise ratio (SNR) environments. Notably, the BS weighted-CSKF algorithm operates effectively even in unknown noise field scenarios, eliminating the requirement for orthogonality between the signal and subspace noise or singular value decomposition. This capability enables accurate DOA estimation without prior knowledge of the number of signal sources. Additionally, investigations into rank-one updates of the covariance matrix highlight the algorithm’s ability to estimate a higher number of sources than sensors employed without imposing constraints on source properties. The algorithm’s versatility extends to coherent and spatially correlated sources, further enhancing its applicability in diverse scenarios. Moreover, employing BS CS-based DOA estimation techniques yields a significant computational load reduction, exceeding 35% compared to the conventional element-space (ES) CS-based approach. Leveraging the proposed technique, fluctuating moving source signals can be efficiently detected and tracked using fewer snapshots, facilitating real-time monitoring and analysis in dynamic environments.