Logo image
Multi-objective H2/H∞ Power Tracking Control in Communication System : Pareto Optimal Approach
Thesis

Multi-objective H2/H∞ Power Tracking Control in Communication System : Pareto Optimal Approach

Huang, Kuo-Chan
Masters, 國立清華大學, 通訊工程研究所
2011

Abstract

多目標最佳化 線性矩陣不等式 功率控制 直序展頻分碼多重存取 柏拉圖最佳解 Multi-objective optimization problem (MOP) linear matrix inequality (LMI) power control DS-CDMA Pareto optimal solution
In order to track a desired signal-to-interference-plus-noise-ratio (SINR) to achieve high system capacity and better communication link quality for cellular communication systems, a feedback power controller is designed for the optimal SINR tracking control. The optimal SINR tracking problem can be regarded as the single-objective (SO) power control problem. However, in order to overcome round-trip delay, channel fading and noises, the SO power control is used to efficiently attenuate the effect of channel delay and these interferences to achieve robust SINR tracking. Hence, these two objectives are needed in power control design. Therefore, we propose the multi-objective (MO)H2/H-infinity power control for DS-CDMA cellular system in this paper. Therefore, the considered multi-objective H2/H-infinity power tracking control is not easy to solve directly. Hence, we propose to minimize the upper bounds of both objectives to solve the multi-objective H2/H-infinity power control problem from the suboptimal viewpoint. Then the MO H2/H-infinity power control problem is transformed to minimizing two upper bounds under the constraint of three linear matrix inequalities (LMIs), i.e. a LMIs-constrained MO problem (MOP). By combining the LMI toolbox in MATLAB with evolutionary searching algorithm, we could easily obtain a set of H2/H-infinity solutions called Pareto optimal solutions for designer selection. Finally, a numerical simulation is given to illustrate the design procedure and to confirm the performance of the proposed MO H2/H-infinity power control for DS-CDMA cellular system. Keywords: Multi-objective optimization problem (MOP), linear matrix inequality (LMI), power control, DS-CDMA, Pareto optimal solution.

Metrics

1 Record Views

Details

Logo image