Energy‐efficient cross‐layer design of dynamic rate and power allocation techniques for cognitive green radio networks

Abstract

ABSTRACTIn this paper, we investigate cross‐layer adaptive rate scheduling techniques for cognitive green radio networks, where a secondary base station is communicating with secondary users (SUs). The base station is equipped with individual finite size buffer for each SU. The activity statistics of the primary users (PUs) are independent and identically distributed. The SUs detect their states and select free channels of the PUs. We study two different methods for PU (or channel) selection. The power minimization problem is formulated as an infinite‐horizon partially observable Markov decision process. The adaptation policy is obtained using maximum likelihood heuristic policy (MLHP) technique because optimal policy for partially observable Markov decision process is intractable. We assume that transition probabilities of the PUs and fading channel between the SU's transmitter and receiver are known. By tracking beliefs of the PUs' hidden states, the SU takes decision on the transmission rate to minimise energy consumption along with delay for a given bit error rate of the communications. Simulation results are given to show the performance of the proposed MLHP. We find that MLHP performs very close to fully observable optimal policy. We provide pointers to choose design parameters (such as delay, number of antennas and channels) for the cognitive green radio network so that the scheduler becomes the most energy‐efficient for a given quality of service requirements of the handled application. Copyright © 2013 John Wiley & Sons, Ltd.