Re-establishing Network Connectivity in Post-Disaster Scenarios Through Mobile Cognitive Radio Networks

Network interoperability and self-organization constitute important requirements of wireless systems for disaster recovery operations. In these scenarios, the original communication infrastructure might be damaged as a consequences of natural calamities, and the whole network might be partitioned into multiple partitions (called islands in the following) that operate on different frequencies/wireless technologies. In this paper, we investigate techniques to maximally re-establish the connectivity among heterogeneous islands through the utilization of specialized repairing units called Stem Nodes (SNs). A SN combines spectrum reconfigurability (offered by Software Defined Radio technology) with self-positioning and dynamic routing functionalities, and thus it is able to replace damaged components of the original infrastructure. Moreover, sets of SNs can self-organize into multi-hop mesh structures connecting heterogeneous islands. Here, we study the problem of determining the optimal configuration of each SN (in terms of location and frequency) so that the number of connected devices of the original network is maximized. Given the NP-hardness of the optimal solution, we propose approximated solutions that reduces the computational complexity over specific topologies. We then compare the centralized algorithms with distributed solution (based on virtual spring approach) that enable SNs to explore the environment (in both space/frequency domain) and to self-organize into virtual mesh structures. Simulation results confirm the effectiveness of distributed algorithm to maximally re-establish the network connectivity on large-scale scenarios.