Paper Info
Title
M-MAC: Motion Sensor Assisted MAC Protocol for Body Area Network with Periodical Movement.
Abstract
When a person is performing repetitive daily activities (e.g. walking, cycling etc.) with equipped motion sensor such as accelerometer, its measured motion signals can go up or down and fluctuates in a periodical manner. In addition, recent researches indicate that the Received Signal Strength (RSS) through wireless channels between on-body sensors and the hub in a Wireless Body Area Network (WBAN) also shares a similar periodical pattern regards to the same body movement. Finding the connections of the two, can enable low-power tracking of body movement and infer corresponding channel changes to provide optimized MAC communications between sensor nodes and the hub. However, since nodes can be placed at different parts of the body experiencing different motion intensities and velocities, it is not straightforward to directly map the motion sensor measurements to the Channel RSS. In this paper, we proposed the notion of motion index which is an effective mechanism to achieve synchronization between the channel periodicity and the corresponding body motion. The proposed M-MAC protocol then reacts to such WBAN channel periodicity and gives high medium access priority to sensors with good channel condition hence increases communication reliability, while buffering data otherwise. We conducted experiments based on real-life on-body channel measurements, and our initial results yield promising results with 10% - 20% PDR increase and 3 dB transmitting power reduction compared to the conventional methods.
Year
Venue
Field
2016
IEEE Global Communications Conference
Key distribution in wireless sensor networks,Synchronization,Wireless,Computer science,Accelerometer,Computer network,Communication channel,Real-time computing,Body area network,Wireless sensor network,RSS
DocType
ISSN
Citations 
Conference
2334-0983
0
PageRank 
References 
Authors
0.34
0
2
Name
Order
Citations
PageRank
Yichao Jin151.12
Stephen Lingfeng Wang2183.74