Spectrum access in cognitive smart-grid communication system with prioritized traffic

Smart grid (SG) comprises heterogeneous characteristics of traffic such as control commands (safety, sensing, smart-meter readings) among power generation, transmission and end users. An effective and reliable two-way communication infrastructure between users and utility providers is important for improving the performance of smart grid. To support such traffic types in the smart grid, we have studied cognitive radio communication infrastructure-based smart grid. In this paper, we proposed a prioritized spectrum access scheme for cognitive smart grid communication systems to address the quality of service (QoS) requirements of heterogeneous traffic. SG traffic is categorized into two priority classes (high and low based on service requirements). We model the proposed spectrum access scheme (by considering the spectrum sensing error) using a multidimensional Markov chain. We have measured the performance based on the blocking probability, the dropping probability, the interference probability and the call completion rate for both the traffic types (high and the low class smart grid traffic). By Considering an environment where the traffic is heavy tailed, we have modeled the arrivals to follow a Pareto distribution, as the Poisson process model may under estimate the traffic especially when traffic is heavy tailed. The results illustrate that the proposed priority based scheme is able to significantly improve the QoS of high-priority traffic. It is also observed that the prioritized system is preferable over the non-prioritized system where all traffic types are treated the same in terms of SG data delivery.