The Effect of Information on Scheduling Performance in Multi-Hop Wireless Networks

Previous research has estimated the performance of wireless networks by assuming that nodes in the network can obtain precise network information. However, in reality, available network information is mostly imprecise and incomplete. In this paper, we study the relationship between wireless network performance and available network information. It is assumed that each node in the network can obtain the information about other nodes within its information collection range, and a distributed graph coloring algorithm is employed to perform scheduling with the available information. The analytical result on the quantitative relationship between the information collection range and the network throughput is derived. We also consider the communication overhead of collecting information, and analyze the tradeoff between network capacity improvement and information collection overhead. Based on the derived result, an optimal information collection range which maximizes the net data rate can be found. Since wireless networks are typically mobile, and the collected information may be inaccurate due to the dynamics of the networks, we analyze the effect of information for mobile wireless networks by considering the information updating rate, and the result can be used to determine the information collection range as well as the information updating period.