Distributed Energy-Efficient Cross-Layer Optimization for Multihop MIMO Cognitive Radio Networks With Primary User Rate Protection

Due to the unique physical-layer characteristics associated with multiple-input multiple-output (MIMO) and cognitive radio (CR), the network performance is tightly coupled with mech-anisms at the physical, link, network, and transport layers. In this paper, we consider an energy-efficient cross-layer optimization problem in multihop MIMO CR networks (CRNs) and provide a new formulation to balance the weighted network utility and the weighted power consumption of secondary users (SUs), with the minimum transmission rate constraint of primary users (PUs) and the SU power consumption constraint. However, this formulation is highly challenging due to the nonconvexity of the PU rate constraint. We propose a solution that features a linearization-based alternative optimization method and a heuristic primal recovery method. We further develop a distributed algorithm to jointly optimize the covariance matrix of the transmitted signal vector at each SU node, bandwidth allocation at each SU link, rate control at each session source, and multihop/multipath routing. Extensive simulation results demonstrate that the performance of the proposed distributed algorithm is very close to that of the centralized algorithm, and the proposed formulation provides an efficient way to save power consumption significantly, while achieving the network utility very close to that achieved with full power consumption.