Statistical Modeling for Spectrum Usage Characterizing Wireless Fading Channels and Mobile Service Dynamics

Cognitive radio (CR) is considered a promising technology for addressing the issue of spectrum scarcity and has been receiving increasing attention in recent years. One of its “cognitive” capabilities is environment awareness, which gathers useful information from the radio environment, such that CR devices can accordingly change the operating (e.g., spectrum sensing) strategies. Spectrum measurement is an effective means for radio environment awareness since ground-truth observation of spectrum usage is more convincing than any theoretical assumption. In this paper, we perform in-depth statistical analysis on wireless signal strength in mobile service (MS) bands and propose a novel statistical model based on characteristic function (CF), which characterizes both wireless fading channels and MS dynamics. We validate the proposed model through empirical studies on the spectrum measurement data previously collected in Guangdong Province of China. We statistically fit the CF of the overall signal strength in MS (GSM uplink) band and estimate both Nakagami fading parameters of wireless channels and the arrival rate of MS. Then, we further utilize the estimated information to optimize the spectrum sensing strategy, which aims at maximizing the secondary user´s (SU´s) expected achievable throughput. The simulation results show that the optimal sensing strategy that we propose significantly improves the long-term average achievable throughput since it fully utilizes the estimated information on wireless fading channels and MS dynamics.