Detection for 5G-NOMA: An Online Adaptive Machine Learning Approach

TL;DR

This paper introduces an online adaptive machine learning detection method for 5G-NOMA uplink that improves performance over traditional MMSE-SIC detection, especially with larger cluster sizes, by using a robust kernel-based filter.

Contribution

It proposes a novel online learning detection algorithm using a sum space of linear and Gaussian RKHSs for 5G-NOMA uplink, enhancing robustness and performance.

Findings

Outperforms MMSE-SIC detection for large clusters

Robust against network variations due to sum space kernel design

Improves spectrum efficiency and connectivity in 5G-NOMA

Abstract

Non-orthogonal multiple access (NOMA) has emerged as a promising radio access technique for enabling the performance enhancements promised by the fifth-generation (5G) networks in terms of connectivity, low latency, and high spectrum efficiency. In the NOMA uplink, successive interference cancellation (SIC) based detection with device clustering has been suggested. In the case of multiple receive antennas, SIC can be combined with the minimum mean-squared error (MMSE) beamforming. However, there exists a tradeoff between the NOMA cluster size and the incurred SIC error. Larger clusters lead to larger errors but they are desirable from the spectrum efficiency and connectivity point of view. We propose a novel online learning based detection for the NOMA uplink. In particular, we design an online adaptive filter in the sum space of linear and Gaussian reproducing kernel Hilbert spaces (RKHSs). Such a sum space design is robust against variations of a dynamic wireless network that can deteriorate the performance of a purely nonlinear adaptive filter. We demonstrate by simulations that the proposed method outperforms the MMSE-SIC based detection for large cluster sizes.