SGER: Scalable Cooperative Communications in Adaptive Large Scale Networks Inspired by Natural Swarms

点击翻译按钮获取中文摘要

ABSTRACT0347514Anna ScaglioneCornell UElectronic devices become more reliable and can be integrated at very small scales. Sensors combined with embedded intelligence meet an increasing number of applications. The value of gathering and processing information locally is tied to the ability to communicate the information itself remotely. The scale of the communication devices developed thus far and their purpose do not meet the requirement of distributed, dynamic, energy limited networks of gigantic scale. I addition, most of the research at the physical layer has been focused on channel access. Cellular networks and wireless local area networks are designed following the principles developed in years of research on multiple access.The backbone of these networks is not wireless and this has clearly polarized the research on mobile communications towards solving the access problem. Networking large numbers of these devices with traditional methods requires prohibitive costs that soon become incompatible with the dynamic nature of the network and the limited capabilities of each node. Recent results also indicate that as the density of the nodes increases peer to peer communications become unfeasible, even under the best multiple access and routing scheme. This results reinforce the need of a paradigm shift in the design and this is what this project is trying to pioneer.The key inspiration in this project stems from the observation that distributed sensing ability and communications ver large scale networks are common in nature and are at the basis of any biological architecture referred to as swarm. Human intelligence is believed to be a remarkable example of such architecture. The communications within the network occur in a very different fashion compared to man-made communication networks and emerge from very simple building blocks. The nodes cooperate and aggregate their signals operating as distributed sources and receivers, where peer to peer transmissions are completely lost.