Optimal Policies for Distributed Data Aggregation in Wireless Sensor Networks
Authors:
- Z. YeZYZ. Ye
- A. A. AbouzeidAAA. A. Abouzeid
- J. AiJAJ. Ai
- 1 January 2007
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 0743166X,p. 1676-1684
Abstract
We consider the scenario of distributed data aggregation in wireless sensor networks, where each sensor can obtain and estimate the information of the whole sensing field through local data exchange and aggregation. The intrinsic trade-off between energy and delay in aggregation operations imposes a crucial question on nodes to decide optimal instants for forwarding their samples. The samples could be composed of the information from their own sensor readings or an aggregation of information with other samples forwarded from neighboring nodes. By considering the randomness of the sample arrival instants and the uncertainty of the availability of the multiaccess communication channel due to the asynchronous nature of information exchange among neighboring nodes, we propose a decision process model to analyze this problem and determine the optimal decision policies at nodes with local information. We show that, once the statistics of the sample arrival and the availability of the channel satisfy certain conditions, there exist optimal control-limit type policies which are easy to implement in practice. In the case that the required conditions are not satisfied, we provide two learning algorithms to solve a finite-state approximation model of the decision problem. Simulations on a practical distributed data aggregation scenario demonstrate the effectiveness of the developed policies, which can also achieve a desired energy-delay tradeoff.Keywords
This publication has 11 references indexed in Scilit:
- MAC protocols for wireless sensor networks: a surveyIEEE Communications Magazine, 2006
- On the optimal density for real-time data gathering of spatio-temporal processes in sensor networks2005
- Optimized scheduling for data aggregation in wireless sensor networks2005
- AIDAACM Transactions on Embedded Computing Systems, 2004
- Aggregation in sensor networks: an energy–accuracy trade-offAd Hoc Networks, 2003
- Asynchronous stochastic approximation and Q-learning2002
- TAG2002
- Directed diffusion2000
- Adaptive protocols for information dissemination in wireless sensor networks1999
- Learning to act using real-time dynamic programmingArtificial Intelligence, 1995