Adaptive Self-Interference Cancellation for Full-Duplex Wireless Communication Systems

TL;DR

This paper proposes an adaptive LMS-based self-interference cancellation method for full-duplex wireless systems, aiming to improve signal quality by reducing interference before ADC in single-user, single-cell scenarios.

Contribution

It introduces an adaptive LMS algorithm for real-time SI channel estimation and subtraction, enhancing full-duplex communication performance.

Findings

LMS-based cancellation reduces self-interference significantly.

Improves bit error rate and spectral efficiency.

Demonstrates robustness in near-far interference conditions.

Abstract

In this letter, we consider single-cell, single-user systems wherein uplink and downlink user equipment communicate with a full-duplex relay. Due to the near-far problem, the self-interference (SI) can be 100-1000x the received signal power. In this context, we consider the adaptive Least Mean Squares (LMS) algorithm to estimate the SI channel and then subtract the SI from the desired received signal before the analog-to-digital converter (ADC). We measure the robustness of this technique in terms of bit error rate (BER) and spectral efficiency.