Dynamic channel selection with reinforcement learning for cognitive WLAN over fiber
International Journal of Communication Systems
| Hi-index | 0.00 |
Congestion in the ubiquitous sensor network occurs when traffic load exceeds available capacity at any point in a network. Congestion causes overall channel quality to degrade and loss rates to rise, leads to buffer drops and increases latencies. Therefore, congestion and latency have effects on energy efficiency, memory size, buffer size, and throughput. This paper proposes a new method, called TCCP (Tournament-based Congestion Control Protocol), to minimize congestion in the ubiquitous sensor network. In this paper, the winner is the stream with the highest importance level in the competition of ubiquitous sensor network. In general, there are a lot of congestion and delay in the continuous streaming of multimedia streams with lower importance level. The proposed method consists of three parts: tournament decision, best-fit control strategy and service differentiation. The final winner in the tournament controls congestion effectively, minimizes packet loss due to congestion, decreases energy consumption, and improves QoS. The simulation result shows that the proposed method is more effective and has better performance compared with those of CCF (congestion control and fairness), PCCP (priority based congestion control protocol), and CODA (congestion detection and avoidance). Copyright © 2010 John Wiley & Sons, Ltd.