Prediction of time series using wavelet Gaussian process for wireless sensor networks
- PDF / 3,634,377 Bytes
- 8 Pages / 595.276 x 790.866 pts Page_size
- 1 Downloads / 182 Views
(0123456789().,-volV)(0123456789(). ,- volV)
Prediction of time series using wavelet Gaussian process for wireless sensor networks Jose Mejia1
•
Alberto Ochoa-Zezzatti1
•
Oliverio Cruz-Mejı´a2
•
Boris Mederos1
Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract The detection and transmission of a physical variable over time, by a node of a sensor network to its sink node, represents a significant communication overload and consequently one of the main energy consumption processes. In this article we present an algorithm for the prediction of time series, with which it is expected to reduce the energy consumption of a sensor network, by reducing the number of transmissions when reporting to the sink node only when the prediction of the sensed value differs in certain magnitude, to the actual sensed value. For this end, the proposed algorithm combines a wavelet multiresolution transform with robust prediction using Gaussian process. The data is processed in wavelet domain, taking advantage of the transform ability to capture geometric information and decomposition in more simple signals or subbands. Subsequently, the decomposed signal is approximated by Gaussian process one for each subband of the wavelet, in this manner the Gaussian process is given to learn a much simple signal. Once the process is trained, it is ready to make predictions. We compare our method with pure Gaussian process prediction showing that the proposed method reduces the prediction error and is improves large horizons predictions, thus reducing the energy consumption of the sensor network. Keywords Sensor networks Time series Gaussian process
1 Introduction Sensor networks generated time series are increasingly significant for emerging applications that analyze this data, however, the acquisition for long periods of time depends on the network making a proper management of its energy resources. One of the main sources of energy expenditure occurs in the transmission of data within the network [1, 2]. Efforts to reduce radio transmissions communication include data aggregation [3, 4] and data reduction via prediction of the sensed magnitude [5], the latter consists in the use of algorithms to analyze and predict the sensed magnitude, in this way if the prediction is within a certain range of error, the data is not transmitted, allowing energy savings. This scheme depends on the model used to predict & Oliverio Cruz-Mejı´a [email protected] 1
Universidad Auto´noma de Ciudad Jua´rez, Ciudad Jua´rez, Mexico
2
Universidad Auto´noma del Estado de Me´xico, Nezahualcoyotl, Mexico
the time series. The analysis of time series is an important area of research in general, in the last decades there has been a growing activity in trying to develop and improve time series forecast models [6]. Time series prediction on sensor networks has been analyzed with different statistical methods, from classical prediction methods such as autoregressive moving averages (ARMA) and integrated autoregressive moving averages [7],
Data Loading...
