Robust Ergodic Uplink Resource Allocation in Underlay OFDMA Cognitive Radio Networks
Authors:
- 16 March 2015
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 15 (2) , 419-431
Abstract
The ergodic resource allocation (ERA) problem for uplink transmission in underlay cognitive radio networks (CRNs) is investigated. The objective is to maximize the ergodic sum-rate of secondary users (SUs) considering the unavailability of perfect channel state information (CSI), and subject to transmit power limitations of SUs, and the interference threshold constraint to guarantee the quality of service of primary users. Since with average-based formulation of ERA, the interference threshold constraint and transmit power limitations of SUs do not hold instantaneously, one can replace the average-based constraints in ERA with their outage-based counterparts. For the uncertainty on the CSI values, we utilize the robust optimization theory where the uncertain parameters are modeled as a sum of the estimated value and error which is assumed to be bounded. We then map the considered ERA problems to their robust counterparts. Generally, the robust approaches degrade the performance (e.g., sum rate of SU), as they conservatively consider the error to be in the maximum extent and try to preserve the constrains under any condition of error (worst-case scenario). We aim to moderate this effect by using appropriate models for uncertain parameters, relaxing the worst-case scenario, and stochastically preserving the constraints. Moreover, robust problems are in general non-convex and suffer from high computational complexity due to the existence of uncertain system parameters. Therefore, we use effective suboptimal approaches to solve them with a reasonable complexity. This includes methods based on chance constraint approach as well as an iterative scheme. The proposed solutions provide a trade-off between robustness, performance, and complexity. Simulation results reveal that by using the proposed schemes, stable sum-rate of SUs in the presence of CSI uncertainties can be achieved while the instantaneous power and interference constraints are met with a desired probability.Keywords
This publication has 50 references indexed in Scilit:
- Power Allocation Over Fading Cognitive MIMO Channels: An Ergodic Capacity PerspectiveIEEE Transactions on Vehicular Technology, 2012
- Optimizing Orthogonal Multiple Access Based on Quantized Channel State InformationIEEE Transactions on Signal Processing, 2011
- Outage Behavior of Discrete Memoryless Channels Under Channel Estimation ErrorsIEEE Transactions on Information Theory, 2009
- Robust IWFA for Open-Spectrum CommunicationsIEEE Transactions on Signal Processing, 2009
- On the Design of Linear Transceivers for Multiuser Systems with Channel UncertaintyIEEE Journal on Selected Areas in Communications, 2008
- Convex Approximations of Chance Constrained ProgramsSIAM Journal on Optimization, 2007
- A robust maximin approach for MIMO communications with imperfect channel state information based on convex optimizationIEEE Transactions on Signal Processing, 2005
- Adaptive resource allocation in multiuser OFDM systems with proportional rate constraintsIEEE Transactions on Wireless Communications, 2005
- Space‐Time Processing for MIMO Communications2005
- Convex Optimization2004