Realistic and Efficient Multi-Channel Communications in Wireless Sensor Networks

This paper demonstrates how to efficiently use multiple channels in Wireless Sensor Networks (WSNs) to improve communication performance. At first an empirical study of multi-channel realities is conducted through intensive experiments. This is further analyzed to understand the practical issues of current multi-channel protocols. The study observes that current multi-channel protocols are not suitable for WSNs, due to the small number of available channels and unavoidable time errors found in real networks. With these observations, a novel tree-based multichannel scheme is proposed for data collection applications, which allocates channels to disjoint trees and exploits parallel transmissions among trees. This kind of arrangement works well with a small number of channels and has a very simple transmission scheme without the need for synchronization at nodes, which makes it suitable for practical WSNs. Next focus is to define and solve the issue of partitioning networks into subtrees with minimizing the intra-tree interferences. The complexity of the problem is analyzed and thus a greedy solution algorithm is proposed. Evaluation results show that it reduces interference in dense networks over other schemes. Finally, the Tree-based Multi-Channel Protocol (TMCP) is implemented in a real testbed. The performance is evaluated through both simulation and real experiments. The experiment conducted demonstrates that the TMCP better accommodates multi-channel realities found in WSNs than other multi-channel protocols.