Model refinement and data filtering in high-tunnel greenhouse sensor network

Authors:
Ju Wang;Kostadin Damevski;Hui Chen
Affiliations:
Virginia State University, Petersburg, VA, USA;Virginia State University, Petersburg, VA, USA;Virginia State University, Petersburg, VA, USA
Venue:
Proceedings of the 7th ACM symposium on QoS and security for wireless and mobile networks
Year:
2011

Citing 9
Cited 0

An analysis of a large scale habitat monitoring application

SenSys '04 Proceedings of the 2nd international conference on Embedded networked sensor systems
A macroscope in the redwoods

Proceedings of the 3rd international conference on Embedded networked sensor systems
Approximate Data Collection in Sensor Networks using Probabilistic Models

ICDE '06 Proceedings of the 22nd International Conference on Data Engineering
Spatial correlation-based collaborative medium access control in wireless sensor networks

IEEE/ACM Transactions on Networking (TON)
Adaptive model selection for time series prediction in wireless sensor networks

Signal Processing
Model-driven data acquisition in sensor networks

VLDB '04 Proceedings of the Thirtieth international conference on Very large data bases - Volume 30
Sensor network data fault types

ACM Transactions on Sensor Networks (TOSN)
Murphy loves potatoes: experiences from a pilot sensor network deployment in precision agriculture

IPDPS'06 Proceedings of the 20th international conference on Parallel and distributed processing
PAQ: time series forecasting for approximate query answering in sensor networks

EWSN'06 Proceedings of the Third European conference on Wireless Sensor Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Precise modeling of networked sensor data is an important link to improve the quality of service of wireless actuating networks in physical world sensing and actuating. It allows the accurate prediction of environmental conditions with partial knowledge and provides means to assess the quality of sensor readings, while reducing system activity for lower battery consumption. A data model is established through past observations, extracting both temporal and spatial correlation between sensors. The model exhibits high sensitivity to the selection of observed attributes for posterior estimation. A novel data model framework is then proposed that integrates both validation and prediction of sensor readings. The framework consists of clusters of self-evolving submodels and each cluster represents a group of closely-related sensor attributes. The sub-model clusters are dynamically formed and optimized for maximum prediction accuracy.