Data filtering for wireless sensor networks using forecasting and value of information

Abstract

Energy constitutes a scarce resource in wireless sensor networks, making energy-efficient operation mandatory. Data transmission has been identified as one of the most energy consuming operations. Consequently, different approaches to reduce data transmissions have been proposed, like data filtering.Recently, the value of information of sensor data has been identified for data filtering, explicitly incorporating application-specific and context-dependent information needs. The filtering is done according to the benefit a data transmission would induce at the recipient. We propose an on-mote filtering approach, which relies on local multi-step assessment of sensor data with forecasting and assessing value of information. We apply our approach to logistics transport processes and evaluate it concerning number of data transmissions and energy efficiency. Our simulation results showed that with our approach the number of data transmissions and the energy consumption can be reduced by over 25% to over 60%, while simultaneously accounting for user-specific information desires.

Value of Information in Wireless Sensor Network Applications and the IoT: A Review

“…2) Stochasticity: The stochasticity (column 12 in Tables I and II) describes whether the VoI is calculated using probabilistic or deterministic methods. Zöller et al [66] discuss how the two types of models differ. The deterministic model can be valued and monetized.…”

Data Communication Optimization for the Evaluation of Multivariate Conditions in Distributed Scenarios

“…Furthermore, the proposed method is related to forecasting techniques on time signals at a sink node. Some representative examples in this scope are [10], [23], [37]. Nevertheless, all these works deal with reducing network traffic by filtering signal data in the sensor domain (usually, integer or real numbers).…”