Wireless testbed implementation of cognitive topology control for power efficient connected networks

摘要

It is of paramount importance to Army to understand if cognitive algorithms can enhance network survivability and robustness. An essential function towards achieving survivability and robustness is topology control for forming power efficient connected topologies, so that network lifetime is extended without affecting the performance and functionality of the network. We have proposed a fully distributed game theoretic (GT) topology control (TC) approach, which aims on the assignments of transmission power to nodes so that a power efficient connected topology is formed. We have modeled and analyzed GT TC under CERDEC's C-NEDAT tool, where we discover its very promising performance characteristics. In this paper we present our effort to implement the GT TC approach on a realistic testbed, for verifying and validating the C-NEDAT performance characteristics and for identifying and understanding the key challenges on implementing such an approach on a realistic setting. The details of the implementation and the key findings and observations from the implementation of GT TC on the wireless testbed are highlighted. Furthermore, the validity of the expected theoretical performance and its comparison with the actual measured performance are indicated through the use case scenarios.