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

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, latency, and 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.

Efficient User Clustering Using a Low-Complexity Artificial Neural Network (ANN) for 5G NOMA Systems

“…Machine learning can be defined as the ability to infer knowledge from user clustering and subsequently to use the knowledge to adapt the behavior of an ML algorithm based on the acquired knowledge. Recent research contributions have shown that the ML provides an effective solution to fast data clustering since various information features of higher dimensionality can be used flexibly [20]. In addition, from a knowledge discovery point of view, it is insightful to design effective algorithms by utilizing the underlying structures of the clustering information.…”

Adaptive Spatial Modulation MIMO Based on Machine Learning

“…As shown in [24], the DNN is capable of learning the statistics of the channel model and capturing its sparsity features. In [28], an online adaptive machine learning approach was proposed for striking an elegant trade-off between the bit error ratio (BER) and complexity in 5G-NOMA systems. In [29], a deep auto-encoder was proposed for channel estimation in wireless energy transfer systems, where an autonomous channel learning scheme was utilized to tackle the phase ambiguity issue.…”

Set-Theoretic Learning for Detection in Cell-Less C-Ran Systems

“…Many mobile communication systems perform channel estimation or learning of other parameters before the actual data communication takes place [11,13,14]. Under the as-sumption of Rayleigh block fading [15], learning (through training) and data communication is performed within each coherence block which is defined as a block of channel symbols over which the channel is assumed to be constant.…”

“…In this section, we compare the performance of D&F strategy with centralized Q&F strategy for QPSK modulation for limited fronthaul capacity and also with that of the method in [8] that can also be applied to our problem. To perform filtering at each RRH, we use the learning-based method in [11] that has been shown to out perform the conventional MMSE-SIC based systems. Following the approach of nonorthogonal multiple access systems, devices are assumed to be allocated to clusters that are assigned disjoint resource blocks (RBs) of the system spectrum (no inter-cluster interference).…”