Resource management for traffic imbalance problem in green cognitive radio networks

The main goal of the cognitive radio network (CRN) is to provide high throughput and ubiquitous data access, but now it is focusing on energy-efficient green CRN. The energy consumption of CRN depends upon the total network time of cognitive users (CUs) and the unbalanced traffic load of CUs (when multiple CUs try to approach the same channel and their time and energy are wasted due to congestion). Hence, we design the channel decision models based on queuing priority for two different channel selection strategies, first probability-based channel selection strategy (PCSS) and second sensing-based channel selection strategy (SCSS). This model helps to calculate the optimum channel selection probability in PCSS and the optimum number of channels in the SCSS. With the help of the above-mentioned parameters, the total network time of CUs can be minimized, and their traffic load can be distributed among multiple applicant channels. We have considered the effects of PU's interruption and sensing errors (missed detection and false alarm) in our analysis. The result shows that, for heavy traffic load of CUs, the sensing-based strategy provides lesser network time, while for low traffic load, the probability-based strategy performs better. Next, we analyze the energy consumption at various operational modes in CRN and propose a channel sensing strategy that represents the energy-saving percentage for different conditions. The proposed strategies can minimize the total network time over 60% and energy consumption over 75% compared to the non-load balancing strategy. (C) 2021 Elsevier B.V. All rights reserved.