Energy-efficient power and feedback bit allocation method in wireless coordinated communication systems

Abstract

In this thesis, the resource allocation schemes are investigated to improve energy efficiency of wireless communication systems such as coordinated multi-point transmission systems, and two-way relay systems. This thesis consists of three parts, those are as follows. In the first part of the thesis, the power allocation problem is solved in the context of maximizing the energy efficiency (EE) of orthogonal frequency division multiplexing (OFDM) signaling over a two-way amplify-and-forward (AF) relay network with two sources and one relay. The EE is defined as the ratio of the spectral efficiency (SE) over the total power consumption incorporating the transmission power, the fixed circuit power and the dynamic circuit power which is rate- dependent. In the EE maximization (EEM) problem, the efficiencies of the power amplifiers (PAs) of the three terminals can differ from one other. It is shown that EE is a quasi-concave function of the transmission power. Based on this result, we use Dinkelbach`s method for fractional programming and propose a two-step allocation (TSA) algorithm to solve the inner loop of the Dinkelbach`s method. Simulation results demonstrate that the proposed scheme can efficiently maximize the EE. In the second part of the thesis, the bit allocation for channel state information (CSI) feedback in coordinated beamforming (CB) systems is proposed. The optimization problem under consideration is to minimize the total transmission power subject to the target signal to interference and noise ratio (SINR) constraints. A suboptimal solution to the problem is developed that provides closed-form expressions for the total feedback rate and the feedback rates of each user and its channels. The advantage of the proposed scheme over the methods with equally distributed feedback rates among the users and/or channels is demonstrated by computer simulation: the CB system employing the proposed scheme causes less outage with less transmission power. In the third part, the bit allocation algorithm is extended to the maximum-minimum weighted SINR with quality of service (QoS). The optimization problem with the probabilistic QoS constraints are intractable, we apply the Bernstein approximation to convert the probabilistic constraints to deterministic convex sub-problems. The original optimization problem becomes convex, but the standard convex optimization method is not applicable. We develop a novel bit allocation algorithm which is based on bi-section and cutting plane methods. The advantage of the proposed scheme over the methods with equally distributed feedback rates among the users and/or channels is demonstrated by computer simulation.