A Game Theory Based Congestion Control Protocol for Wireless Personal Area Networks
In Wireless Sensor Networks (WSNs), congestion can cause a plethora of malfunctions such as packet loss, lower throughput and energy inefficiency, potentially resulting in reduced deployment lifetime and under-performing applications. This has led to several proposals describing congestion control (CC) mechanisms for sensor networks. Furthermore, the WSN research community has made significant efforts towards power saving MAC protocols with Radio Duty Cycling (RDC). However, careful study of previous work reveals that RDC schemes are often neglected during the design and evaluation of congestion control algorithms. In this paper, we argue that the presence (or lack) of RDC can drastically influence the performance of congestion detection. In addition, most WSN CC mechanisms are evaluated under traditional sensor network topologies and protocols (e.g. trickle data dissemination, tree data collection). The emerging IPv6 over Low power Wireless Personal Area Networks (6LoWPAN) and related standards pose a new requirement: we now need to investigate if previous findings regarding congestion control are still applicable. In this context, this paper contributes a comprehensive evaluation of existing congestion detection mechanisms in a simulated, multi-node 6LoWPAN sensor network. We present results from two sets of experiments, differentiated by the presence or lack of RDC.