Narrowband Interference Mitigation in SC-FDMA Using Bayesian Sparse Recovery

TL;DR

This paper introduces a Bayesian sparse recovery-based method for mitigating narrowband interference in SC-FDMA systems, utilizing sparse signal properties, grid mismatch modeling, and data-aided techniques to enhance spectral efficiency and system robustness.

Contribution

It proposes a novel NBI mitigation scheme that accounts for grid mismatch and employs a sparsifying transform, improving interference cancellation in SC-FDMA systems.

Findings

Effective NBI mitigation demonstrated through numerical results.

Enhanced spectral efficiency via data-aided recovery.

Extension to MIMO systems with collaborative support search.

Abstract

This paper presents a novel narrowband interference (NBI) mitigation scheme for SC-FDMA systems. The proposed NBI cancellation scheme exploits the frequency domain sparsity of the unknown signal and adopts a low complexity Bayesian sparse recovery procedure. At the transmitter, a few randomly chosen sub-carriers are kept data free to sense the NBI signal at the receiver. Further, it is noted that in practice, the sparsity of the NBI signal is destroyed by a grid mismatch between NBI sources and the system under consideration. Towards this end, first an accurate grid mismatch model is presented that is capable of assuming independent offsets for multiple NBI sources. Secondly, prior to NBI reconstruction, the sparsity of the unknown signal is restored by employing a sparsifying transform. To improve the spectral efficiency of the proposed scheme, a data-aided NBI recovery procedure is outlined that relies on adaptively selecting a subset of data carriers and uses them as additional measurements to enhance the NBI estimation. Finally, the proposed scheme is extended to single-input multi-output systems by performing a collaborative NBI support search over all antennas. Numerical results are presented that depict the suitability of the proposed scheme for NBI mitigation.