Higher demand for good quality, more reliable and cleaner electrical power has driven the focus towards more robust generation, transmission and distribution framework. It has been envisioned that, a more intelligent grid capable of taking fast and reliable decisions is the need of the hour The Smart Grid. Through intercommunicating devices at every level, the smart grid is expected to take power management to a new level of sophistication and efficiency. The communication network for connecting all these devices is of paramount importance to enable two way information exchange, and reliable and fast automated control. It needs to comply with critical demands of throughput and delay. Use of wireless links is set to be handicapped by ever increasing spectrum scarcity. In order to build up the reliable network, use of Cognitive Radio (CR) provides a feasible solution, by accessing a wider range of spectra through Dynamic Spectrum Access (DSA). In this paper we present a CR based Wireless Sensor and Actuator Network (WSAN) model for distribution grid automation. Random appearances of Primary User (PU) have been modeled for DSA implementation. The performance of CR model is compared with legacy Wi-Fi model to evaluate its feasibility for adaptation to the demanding smart grid communication requirements. © 2013 IEEE.