Simple and Complete Resynchronization for Wireless Sensor Networks

Hiromi YAGIRI  Takeshi OKADOME  

Publication
IEICE TRANSACTIONS on Communications   Vol.E102-B   No.4   pp.679-689
Publication Date: 2019/04/01
Online ISSN: 1745-1345
DOI: 10.1587/transcom.2018SEP0002
Type of Manuscript: Special Section PAPER (Special Section on Sensing, Wireless Networking, Data Collection, Analysis and Processing Technologies for Ambient Intelligence with Internet of Things)
Category: 
Keyword: 
asynchronous neighbor discovery,  resynchronization,  ultra-limited resources,  wireless sensor network,  

Full Text: FreePDF(1013.8KB)


Summary: 
The methods proposed in this paper enable resynchronization when a synchronization deviation occurs in a sensor node without a beacon or an ack in a wireless sensor network under ultra-limited but stable resources such as the energy generated from tiny solar cell batteries. The method for a single-hop network is straightforward; when a receiver does not receive data, it is simply placed in recovery mode, in which the receiver sets its cycle length TB to (b±γ)T, where b is non-negative integer, 0 < γ < 1, and T is its cycle length in normal mode, and in which the receiver sets its active interval WB to a value that satisfies WBW + γT, where W is its active interval in normal mode. In contrast, a sender stays in normal mode. Resynchronization methods for linear multi-hop and tree-based multi-hop sensor networks are constructed using the method for a single-hop network. All the methods proposed here are complete because they are always able to resynchronize networks. The results of simulations based on the resynchronization methods are given and those of an experiment using actual sensor nodes with wireless modules are also presented, which show that the methods are feasible.