Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
4954358 | Computer Communications | 2017 | 11 Pages |
Abstract
RI-MAC is the most representative receiver-initiated (RI) duty-cycled MAC, designed to improve the throughput and minimize the power consumption of wireless sensor networks (WSNs). With RI-MAC, the receiver always initiates each transmission opportunity for all intended senders, while the senders compete for channel access based on the same backoff windows (BWs) controlled by the receiver. Such a WSN may encounter a throughput bottleneck under the star topology around a sink node in practice, and hence investigating the performance of RI-MAC under such a condition is of great importance. As opposite to the conventional sender-initiated MACs such as 802.11 and 802.15.4, in RI-MAC, data transmissions of all senders delivered to the same receiver are coupled closely due to the RI nature. Consequently, the well-known Bianchi model and its extensions for sender-initiated MACs are inapplicable for RI-MAC. In this paper, we develop a brand-new model from the perspective of a receiver. We study the system saturation throughput and the power consumption of RI-MAC with the model under a star topology. To the best of our knowledge, it is the first model for any kind of receiver-initiated duty-cycled MACs. Our model captures the most essential characteristics of receiver-initiated duty-cycled MACs and therefore has a wide application scope. With this model, we deeply reveal the relationship between the parameter settings and the RI-MAC performance, and further optimize the RI-MAC design. Extensive simulations verify that our model is very accurate. For instance, the average errors between the theoretical and simulation results are less than 6% for both the system saturation throughput and power consumption.
Keywords
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Networks and Communications
Authors
Rongchang Duan, Qinglin Zhao, Hanwen Zhang, Yujun Zhang, Zhongcheng Li,