Efficient data transmission in multi hop cognitive radio networks using artificial intelligence based expert systems

Abstract

Cognitive Radio (CR) is an emergent communication platform that offers solutions for spectrum scarcity issues. Cognitive Radio Networks (CRNs) offer increased bandwidth to mobile consumers through wireless heterogeneous architectures as well as dynamic spectrum acquisition mechanisms. Secondary Users (SUs) can opportunistically explore and employ the blank spaces present in the channels that are licensed. This makes the SU evacuate the licensed channel and then switch to a vacant channel when an incumbent Primary User (PU) interferes with the channel. When a Primary User (PU) interferes with the channel, it causes degradation of SUs because of frequent switching of channels.CRNs enforce challenges such as spectrum sensing, spectrum allocation, routing, etc. The functions of spectrum management can resolve those challenges to realize a new paradigm of the network. The traditional methods of spectrum management appear to provide unreliable performance in the case of an actual environment where noise is predominant. The current research in spectrum management is being explored using Computational Intelligence (CI) and Artificial Intelligence (AI). Hence, this research work proposes two frameworks based on (i) Genetic Algorithm Optimized Fuzzy Decision System (GA optimized FDS) which employs CI, and (ii) Deep Recurrent Reinforced Learning-based Q-Routing (DRRL based Q-routing) which employs AI for performing efficient spectrum management. newlineThe first framework consists of GA-optimized FDS for channel selection, channel switching, and spectrum allocation in a multi-channel multi-hop CRN. newline