Evaluating Short-Term Forecasting of Multiple Time Series in IoT Environments

TL;DR

This paper evaluates short-term forecasting methods for multiple IoT sensor streams under resource constraints, comparing statistical, machine learning, and neural network models on real datasets to improve decision-making.

Contribution

It introduces a unified experimental protocol for forecasting multiple time series in resource-limited IoT environments, assessing various models' accuracy.

Findings

Neural networks outperform traditional models in accuracy.

The proposed framework aids in establishing reliable forecasting strategies.

Resource constraints significantly impact forecasting performance.

Abstract

Modern Internet of Things (IoT) environments are monitored via a large number of IoT enabled sensing devices, with the data acquisition and processing infrastructure setting restrictions in terms of computational power and energy resources. To alleviate this issue, sensors are often configured to operate at relatively low sampling frequencies, yielding a reduced set of observations. Nevertheless, this can hamper dramatically subsequent decision-making, such as forecasting. To address this problem, in this work we evaluate short-term forecasting in highly underdetermined cases, i.e., the number of sensor streams is much higher than the number of observations. Several statistical, machine learning and neural network-based models are thoroughly examined with respect to the resulting forecasting accuracy on five different real-world datasets. The focus is given on a unified experimental protocol especially designed for short-term prediction of multiple time series at the IoT edge. The proposed framework can be considered as an important step towards establishing a solid forecasting strategy in resource constrained IoT applications.