Adaptive Online Data Compression
HPDC '02 Proceedings of the 11th IEEE International Symposium on High Performance Distributed Computing
An Experimental Comparison of Min-Cut/Max-Flow Algorithms for Energy Minimization in Vision
IEEE Transactions on Pattern Analysis and Machine Intelligence
Cross-Layer Collaborative In-Network Processing in Multihop Wireless Sensor Networks
IEEE Transactions on Mobile Computing
Multiprocessor Scheduling with the Aid of Network Flow Algorithms
IEEE Transactions on Software Engineering
| Hi-index | 0.00 |
Wireless sensor-based healthcare systems consist of hierarchically organized components with varying energy and performance capabilities, such as sensors, local aggregators (e.g. cell phones) and servers. This paper proposes a distributed graph-based approach to partition tasks across these various components with the goal of optimizing the available energy. State of the art implementations assume that all the data is gathered and forwarded for computation to the servers. We show in this work that significant gains in energy efficiency can be obtained if some of the processing tasks are assigned to sensors and local data aggregators. Our DynAGreen algorithm takes the graph associated with the workload and successively partitions it between the server, cell phones and sensors such that the overall system energy utilization for computing and communication tasks is minimized. Our experiments show that the task assignment given by DynAGreen reduces the overall system energy by 30% with respect to an optimal static design time assignment when minor run time variations are considered.