Kakiuchi, Hirofumi (2004) Dynamic Load Balancing in Sensor Networks. Technical Report. Stanford.
A sensor network with a tree topology faces challenges in efficient data aggregation, especially in normalizing energy consumption. Our paper introduces a new load-balancing algorithm that works autonomously to reduce overload and to distribute energy requirements without an energy-aware module in the sensor. We investigate the differences produced by two approaches to decision-making: a semi-centralized approach or a fully-distributed one. We also discuss how to avoid cascading reorganization as loads are balanced.
|Item Type:||Techreport (Technical Report)|
|Subjects:||Computer Science > Distributed Systems|
|Deposited By:||Import Account|
|Deposited On:||13 Jun 2004 17:00|
|Last Modified:||23 Dec 2008 09:12|
Repository Staff Only: item control page