Paper Info
Title
EC-CENTRIC: An Energy- and Context-Centric Perspective on IoT Systems and Protocol Design.
Abstract
The radio transceiver of an Internet of Things (IoT) device is often where most of the energy is consumed. For this reason, most research so far has focused on low-power circuit and energy-efficient physical layer designs, with the goal of reducing the average energy per information bit required for communication. While these efforts are valuable per se, their actual effectiveness can be partially neutralized by ill-designed network, processing, and resource management solutions, which can become a primary factor of performance degradation, in terms of throughput, responsiveness, and energy efficiency. The objective of this paper is to describe an energy-centric and context-aware optimization framework that accounts for the energy impact of the fundamental functionalities of an IoT system and that proceeds along three main technical thrusts: 1) balancing signal-dependent processing techniques (compression and feature extraction) and communication tasks; 2) jointly designing channel access and routing protocols to maximize the network lifetime; and 3) providing self-adaptability to different operating conditions through the adoption of suitable learning architectures and of flexible/reconfigurable algorithms and protocols. After discussing this framework, we present some preliminary results that validate the effectiveness of our proposed line of action, and show how the use of adaptive signal processing and channel access techniques allows an IoT network to dynamically tune lifetime for signal distortion, according to the requirements dictated by the application.
Year
DOI
Venue
2017
10.1109/ACCESS.2017.2692522
IEEE ACCESS
Keywords
Field
DocType
Context-awareness,energy-efficiency,internet of things,protocol design
Resource management,Transceiver,Computer science,Efficient energy use,Communication channel,Computer network,Context awareness,Physical layer,Throughput,Routing protocol,Distributed computing
Journal
Volume
ISSN
Citations 
5
2169-3536
8
PageRank 
References 
Authors
0.48
45
7
Name
Order
Citations
PageRank
Alessandro Biason1588.10
Chiara Pielli280.48
Michele Rossi322826.33
Andrea Zanella41920131.55
Davide Zordan51017.67
Kelly, M.691.51
Michele Zorzi77079736.49