In Partial Fulfillment of the Requirements for the Degree of
Master of Science
Will defend his thesis
Precisely-timed concurrent transmission(CX) is relatively new in wireless sensor network(WSN) area, and protocols based on CX can implement reliable and conceptually simple flooding mechanism. These protocols work well; however, they are intrinsically energy inefficient for non-flooding traffic models, because in such cases not every mote needs those broadcasted packets, but all of them have to receive and then forward. Efforts have been spent in reducing forwarder set size for those cases, typically by removing those motes which are not “between” the packet source and destination so that they are “useless”. These efforts are effective, but they have to keep some unnecessary forwarders because they rely on topology information and this information is subject to network dynamic and then not precise.
Here we present a new mechanism based on Forwarder Selection(CXFS) protocol, to further remove unnecessary forwarders. We implement 3 variants of this mechanism and evaluate them on our 27-node testbed. Results show that our new protocol can reduce forwarder set size by up to 40% and reduce ducy cycle by 40%, at the expense of lower packet reception ratio(less than 10% lower).Our new protocol is a feasible option when energy efficiency is top priority and packet reception ratio is not extremely important.
Date: Monday, November 24, 2014
Time: 2:30 PM
Place: PGH 362
Faculty, students, and the general public are invited.
Advisor: Prof. Omprakash Gnawali