Scalable Cognitive Routing Protocol for Mobile Ad-Hoc Networks

This paper introduces a novel scalable cognitive routing protocol (SCRP) for mobile ad-hoc networks. It is an on-demand cognitive routing protocol. It employs an intelligent flooding protocol, a novel approach for scalable flooding. A neural network machine learning method is adopted to make nodes aware of history. The intelligent flooding protocol saves routing overhead because nodes selectively flood RREQ packets along predicted strong links and over predicted good frequencies. The intelligent flooding protocol can be divided into two parts, the space flooding protocol and the spectrum flooding protocol. Simulation results show that SCRP scales well by network size, network dynamics and network spectrum.