Joint Optimization of Collaborative Sensing and Radio Resource Allocation in Small-Cell Networks

The current trend towards heterogeneous networks requires some sort of self-organization capability, in terms of self-optimization and self-configuration. The basic steps enabling self-optimization typically require listening to the environment, learning and adapting resource allocation consequently. In this work, we propose a joint optimization of sensing parameters and radio resource allocation in order to maximize the opportunistic throughput, under the constraint of limiting undue interference towards primary users. The method applies to wireless networks where the sensing nodes are allowed to cooperate to improve their local decision capabilities using a fully decentralized approach based on distributed consensus. The proposed optimization maximizes the opportunistic throughput, taking into account decision errors, sensing time and time necessary to achieve a consensus over the sensed variables. Because of the lack of knowledge of the channel between the cognitive users and the macro-users, the interference constraint is formulated in probabilistic form, depending on the spatial distribution of the nodes and on the channel fading model. Our formulation allows us to find the optimal false alarm rate and, consequently, the optimal decision thresholds jointly with the optimal power/bit allocation. Finally, we show that, given a total power budget constraint, there is an optimal way of distributing the available power between the power spent to achieve consensus and the power spent for transmitting the data payload.