Multi-objective optimization of advanced sleep mode for energy saving in cognitive radio network

The Advanced Sleep Modes (ASM) concept corresponds to entering the Base Station (BS) progressively deeper and less energy-intensive states to reduce energy consumption. Introducing the ASM can mitigate energy wastage during low-traffic periods in the Cognitive Radio Network (CRN). In this study, we propose a strategy for integrating ASM within the CRN architecture to effectively handle primary and secondary traffic across varying ASM sleep states. Additionally, we study the general scenario of CRN with heterogeneous secondary users, imperfect sensing, and unreliable BS due to the arrival of negative packets (virus attack). By modeling the entire system as a three-dimensional discrete-time Markov chain, we conduct the transient analysis of the proposed model. Through numerical demonstrations involving reliability and queueing analyses, we substantiate the validity of the proposed model and examine the impact of reliability on its performance. Then, we showcased the effectiveness of the ASM strategy by comparing it with the Sleep Mode (SM) strategy in terms of the waiting time and blocking probability of the secondary user and the degree of energy savings. Also, simulation experiments are conducted to corroborate the accuracy and validity of the numerical results. Finally, we formulate the Cost Benefit Function (CBF), which depends on both the successful transmission and waiting time of secondary packets. Subsequently, we obtain the Pareto optimal solution for CBF and the degree of energy saving using the Non-dominated Sorting Genetic Algorithm II (NSGA-II) and Multi-Objective Particle Swarm Optimization (MOPSO) techniques for multi-objective optimization.