Performance Optimization through Channel Profiling in Cognitive Sensor Networks

Cognitive Radio (CR) is an important future wireless radio access technology having the potential to solve the spectrum scarcity and under-utilization problems. Using CR however, poses new challenges for network nodes and architectures due to the inherent complexity of cognitive radio technology. In this paper, we propose a unique database-driven “Channel Profiler Framework” for cognitive radio-driven wireless sensor networks often referred to as cognitive sensor networks, and analyze its impact on network performance optimization. The channel profiling architecture is used to circumvent the unpredictability of channels’ availability in the presence of primary users of the licensed spectrum in the context of cognitive radio-driven networks. We take cognitive sensor networks as candidate network type and gradient-based routing protocols for analysis of our proposed framework. Simulation results are given which show the benefit of the proposed channel profiling framework for effective channel and path selection. The presented architecture integrates very well with the architecture of a cognitive sensor node based on specific task-driven components.