Interference Mitigation in 5G Network Densification Technologies: Algorithms and Performance Limits

Abstract

The advent of fifth generation (5G) wireless technology is expected to unleash an unprecedented boost in network capacity, spectral and energy efficiencies, and peak data rates, accompanied by a significant increase in the number of connected devices via ultra-low latency connections. To achieve these aggressive goals, network densification has emerged as a mainstream technology in 5G in various manifestations to improve the capacity and spectral efficiency: increasing the number of base stations, increasing the number of antennas per site (a.k.a. massive multiple-input multiple-output (MIMO)), deploying distributed cell-free massive MIMO, employing distributed device-to-device (D2D) communications, and applying non-orthogonal multiple access (NOMA) communications, among many others. However, interference, whether in the form of inter-user or inter-cell, remains the major bottleneck as we densify the networks and reuse the spectral resources, and cannot be eliminated if we rely on network-centric topologies. While the spatial dimensions available at the centralized and distributed massive MIMO base stations (BSs) can be leveraged to suppress interference at the user equipment (UE), new approaches for interference mitigation that take into consideration the underlying hardware constraints and impairments, as well as signaling overhead are needed. In addition, by exploiting the physical proximity of communicating devices, offloading traffic from network-centric entities to distribu-ted D2D networks and increasing resource utilization via NOMA communications, adequate user pairing criteria and power allocation policies become attractive efficient interference mitigation schemes with affordable complexity and signaling overhead.

In this dissertation, we investigate the problem of efficient interference mitigation schemes for emerging network densification technologies in 5G communications from four different perspectives. First, we propose and analyze channel allocation (CA) and power control (PC) schemes to mitigate interference in a D2D underlaid cellular system modeled as a random network using stochastic geometry. Second, we extend the proposed interference mitigation techniques to consider NOMA MIMO systems. Third, in the context of massive MIMO systems, we propose and analyze the performance of various baseband processing schemes under low resolution analog-to-digital converters (ADCs). We analyze the uplink achievable rate by a massive MIMO system when the base station is equipped with a large number of low-resolution ADCs. We propose new techniques that account for the severe non-linearity effects of the coarse quantization and incorporate a pilot-based channel estimation error. Fourth, in the context of distributed massive MIMO systems, we study angle-domain processing techniques targeted for suppressing interference in frequency-division duplexing (FDD) based cell-free massive MIMO systems. Most prior work on cell-free (distributed) massive MIMO systems assume time-division duplexing mode, although FDD systems dominate current wireless standards. Efficient power control schemes are investigated for cell-free massive MIMO systems while considering the effect of backhaul power consumption. This dissertation describes the research scope, presents the completed work, and draws the future work.