Monday, March 14, 2011
Hierarchical Power Management in Disruption Tolerant Networks with Traffic-Aware Optimization
In this paper, the author investigates power management in DTNs with high randomness in the node mobility. Author presents a hierarchical power management framework, in which nodes control two radio interfaces to discover contacts. In addition, provides traffic aware approximation algorithms to save energy while discovering enough contacts to deliver the traffic load in the network. Simulation results from two mobility models show that generalized power management mechanism could achieve better energy efficiency than mechanisms relying only on one radio for contact discovery. In addition, when the traffic load can be predicted, approximation algorithms helps nodes to save significant amount of energy while handling the expected traffic load. Finally, the additional information allows the one-radio architecture to save equivalent energy to the two-radio architecture, overcoming the disadvantage of not having the additional radio.
Subscribe to:
Post Comments (Atom)
This paper gives an excellent explanation on power management schemes in disruption tolerant networks especially for energy optimization. But the simulation results do provide an insufficient explanation.Since the criterion is to go for energy optimization so Algorithm assumes high power radio for Detection and transmission of beacons while low power radio only used for detection instead the algorithm should also be focussing on transmission through low power radio as well. Nevertheless, author clears the fundamentals and theoretical explanation for various power management schemes which made the paper easy to follow.
ReplyDeleteThis is a well explained and easy to undersatnd paper. The graphs plotted to evaluate the performance of the protocol are not well explained. As Anshika pointed when low radio is able to contact nodes with in the radio range then why the need of high radio to send data to the nodes when low radio can do by itself ?
ReplyDeleteI think the idea of dual radio in this paper is to ensure reliable communication among the nodes and also not to waste the energy of the nodes. The low power radio is enough to identify its neighbor and high power is used for better packet transmission. Though this protocol focuses on saving energy by using dual radio, it is still not energy efficient due to the beacons and idle listening intervals.
ReplyDeleteAgain, like in protocols such as PTW you could have to pay a lot more money to deploy a network that consists of dual radio sensor nodes. But if cost is no object I still feel that PTW might be a better bet because like everyone else is saying the graphs are a little confusing and aren't explained as well as in PTW.
ReplyDelete