Performance Analysis and Optimization of Interference Limited Multiantenna Bidirectional Relay System

Abstract

Over the past decade, cooperative communication has emerged as an attractive technique for overcoming the shortcomings of point-to-point wireless communication systems. Cooperative relaying improves the performance of wireless networks by creating an array of multiple independent virtual sources transmitting the same information as the source node. Also, when relays are deployed near the edge of the cellular network, they can provide additional coverage in network blind spots. But due to dense-frequency reuse, they are typically exposed to co-channel interference (CCI). Therefore, CCI often dominates AWGN in high frequency and degrades the performance of wireless systems. Moreover, in wireless sensor networks, the interferers transmit power level similar to the source and therefore limit the signal-to-interference-plus-noise ratio (SINR). This thesis investigates the impact of co-channel interference on multi-antenna cooperative two-way relay networks and underlay relaying systems.
First, we study the performance analysis of multi-antenna amplify-and-forward (AF) two-way relay network (TWRN) in the presence of CCI. To fully exploit multiple antenna diversity by avoiding high feedback overhead, we employ transmit and receive antenna selection at the user nodes. Specifically, we formulate the exact expression of the generalized signal-to-interference-plus-noise ratio (SINR). Based on the derived SINR, we further derived the upper-bounded cumulative distribution function. Subsequently, we evaluate the expression of the symbol error rate and outage probability (OP). We determined the overall outage probability (OOP) and the overall symbol error rate to assess the overall performance of the system. With the aim to gain more insights, high SNR analysis of OP and OOP are analyzed, which provides the information about coding gain and diversity order of the system. Based on Jensen’s inequality, we derive upper bounds on the achievable rate. Furthermore, we formulate three optimization problems to minimize outage viz., optimization of relay location, optimization of power allocation and joint optimization of relay location and power allocation.
Next, we extend the analysis of a multi-antenna based two-way relay system with CCI under Nakagami-m fading. Herein, we employ maximum-ratio-transmission (MRT)/maximum-ratio-combining (a.k.a. beamforming) at user nodes in the presence of CCI at all the terminals. With such a complicated but practical set-up, we first deduce the instantaneous end-to-end SINRs after performing partial and self-interference cancellation. Vii Based on SINRs, we obtained the CDF and probability density function (PDF). We further derive a tight lower bound on OP that has a simple and compact closed-form representation. To acquire the overall performance, we obtained an upper bound expression of sum symbol error rate (SSER). Furthermore, to render insights into the performance degradation due to the effect of interference, asymptotic expression of OP and SSER are obtained, which easily enables us to evaluate diversity order and coding gain. We further investigate the joint optimization problem of relay location and power allocation to minimize the OP and SSER.
We further extend to present the comparative analysis of beamforming and antenna selection in the presence of interferers with different powers. Herein, a single-antenna relay terminal helps the users with an arbitrary number of antenna to exchange information bi-directionally. Under such a scenario, we evaluate and compare the performance of multi-antenna transmission strategies viz., antenna selection and beamforming. We derive the generalized tight upper bound expressions of OP for both the strategies over Rayleigh fading channel. To gain more insights, we determine the achievable diversity order of the considered system through asymptotic analysis. We further analyze the power optimization problem to minimize the OP for both scenarios.
The last part of the thesis focuses on the investigation of a practical scenario of interference-limited spectrum sharing systems. It consists of multi-antenna primary user (PU) network and multi-antenna multiple secondary user (SU) network with a single antenna relay. The distinct fading parameters, as well as unequal average fading powers between the interference and relaying links, are assumed under Nakagami-m fading. Using the underlay cognitive approach with the cooperation process among the SU nodes, an approximate closed-form expression for the OOP of the secondary network is derived. In this analysis, an opportunistic scheduling and antenna selection algorithm in the secondary network has been employed. Considering the interference constraints due to the presence of PU receiver, power allocation of the secondary user is done. We examined the effects of PU & SU transmit antenna, number of SUs, fading severity parameter and secondary network relay placement on the system performance. We further investigate the relay location optimization to minimize the secondary network OOP.
The tightness of our analysis is attested through Monte Carlo simulation and highlights the impact of interference under various antenna configuration on the overall system performance.