Inter-cell collaborative spectrum monitoring for cognitive cellular networks in fading environment

We propose a novel inter-cell power allocation for multi-carrier cognitive cellular networks. The proposed scheme incorporates the network-wide primary service communication activity into sub-channel power allocation. To model the primary service activity we define sub-channel activity index (SAI). SAI is then evaluated through a simple yet efficient collaborative spectrum monitoring scheme with very low signaling overhead. Corresponding to a secondary user transmission over a sub-channel, a utility function is defined which is a decreasing function of SAI, and an increasing function of the sub-channel achievable rate. Optimal power allocation is then formulated to maximize the total secondary base station (SBS) utility, subject to SBS transmit power, and primary system collision probability constraints. The sub-optimal solutions to the non-convex optimization are then obtained utilizing dual decomposition method. Comparing with a cognitive cellular network with no signalling among the SBSs, where SBS adopts equal sub-channel power allocation, simulation results indicate a significant gain on the achievable rate. We further compare the rate performance with an ideal system in which perfect interference channel state, and spectrum sensing information are available at the SBS and a combination of underlay and overlay access techniques are adopted. Comparing to the ideal system, the proposed method requires significantly lower signaling overhead while its rate performance closely follows the ideal access.