Autonomous reconfiguration of backbones in free space optical networks

Abstract

This research focuses on algorithms for autonomous reconfiguration of backbone networks utilizing the free space optical (FSO) technology. Such networks consist of high data rate and narrow beam FSO links between fixed and/or mobile nodes. The main challenge is to maintain the required quality and connectivity of such networks amidst changing atmospheric (obscuration/loss of sight/scintillation etc), platform (mobility/jitter) and traffic conditions, which critically affect their performance. This paper focuses on the algorithms and heuristics for the dynamic and autonomous reconfiguration processes of free space optical (FSO) sensor networks with the goal of optimizing performance based on network layer parameters. The objective is to form a bi-connected topology with minimum congestion, given the traffic patterns in the network in the form of a traffic matrix. In current research, we discuss heuristics for forming 3-degree networks to solve the congestion minimization problem. A 3-degree network is the one in which all nodes have three optical transceivers on it. We present low computation, high performance algorithms for 3-degree networks, which find low congestion bi-connected topologies quickly. Based on our preliminary data of efficiency versus complexity, we discuss the possible applications of such heuristics in future dynamic, high bandwidth and autonomously reconfigurable tactical networks.