A joint design for distributed stable routing and channel assignment over multihop and multiflow mobile ad hoc cognitive networks

Data communication in mobile ad hoc cognitive networks (MACNets) significantly suffers from link instability and channel interference. The availability and stability of each link in MACNets highly depends on not only the relative movement of neighbor nodes but also the adjacent communication among primary nodes and among cognitive nodes. In multihop and multiflow MACNets, this problem becomes even worse because multiple links potentially interfere with each other. In this paper, we propose a cross-layer distributed approach, called mobility-prediction-based joint stable routing and channel assignment (MP-JSRCA), to maximize the network throughput by jointly selecting stable routes and assigning channels avoiding inter- and intra-flow interferences based on mobility prediction. To quantitatively measure the communication quality of links, we propose a new metric data transmission cost (DTC) that captures node mobility, impact to primary nodes, and channel conflict among cognitive nodes. In our MP-JSRCA, each relay node selects the best link with the smallest DTC as the next hop, within a specified sector region towards the destination. NS2-based simulation results demonstrate that our MP-JSRCA algorithm significantly improves network throughput, and the higher degree of interference MACNets experience, the more improvement can be achieved.