Practical Data Prediction for Real-World Wireless Sensor Networks

Abstract

Abstract-Data prediction is proposed in wireless sensor networks (WSNs) to extend the system lifetime by enabling the sink to determine the data sampled, within some accuracy bounds, with only minimal communication from source nodes. Several theoretical studies clearly demonstrate the tremendous potential of this approach, able to suppress the vast majority of data reports at the source nodes. Nevertheless, the techniques employed are relatively complex, and their feasibility on resource-scarce WSN devices is often not ascertained. More generally, the literature lacks reports from real-world deployments, quantifying the overall system-wide lifetime improvements determined by the interplay of data prediction with the underlying network. These two aspects, feasibility and system-wide gains, are key in determining the practical usefulness of data prediction in real-world WSN applications. In this paper, we describe Derivative-Based Prediction (DBP), a novel data prediction technique much simpler than those found in the literature. Evaluation with real data sets from diverse WSN deployments shows that DBP often performs better than the competition, with data suppression rates up to 99% and good prediction accuracy. However, experiments with a real WSN in a road tunnel show that, when the network stack is taken into consideration, DBP only triples lifetime-a remarkable result per se, but a far cry from the data suppression rates above. To fully achieve the energy savings enabled by data prediction, the data and network layers must be jointly optimized. In our testbed experiments, a simple tuning of the MAC and routing stack, taking into account the operation of DBP, yields a remarkable seven-fold lifetime improvement w.r.t. the mainstream periodic reporting.

A Spatial-Temporal Correlation Approach for Data Reduction in Cluster-Based Sensor Networks

“…This filter consists of multiple layers of regular Least Mean Square (LMS) filters, each layer takes feedback from the previous layer in the hierarchy aiming to minimize the prediction error of the model. Another technique called Derivative Based Prediction (DBP) was introduced in [25], it is less complex than the adaptive-filter based methods. The prediction model is simply a straight line that interpolates a fixed window of data of size m using the first and last l values in the window.…”

“…Figure 12 shows the maximum memory size needed by the CH in function of the number of nodes belonging to the cluster. The maximum memory size required by the CH is 8 × M ax(N (SR max + 1 2 N + 4) + 1, 6N + 1) bytes, since the values stored in the first part of the Algorithm (1-17), could be cleared once the sensors have been matched (Algorithm 1, line [17][18][19][20][21][22][23][24][25][26][27][28].…”

FROG: A Robust and Green Wireless Sensor Node for Fog Computing Platforms

“…Furthermore, the data acquisition requirements, time, and data amount to be processed, greatly vary from one application to another. In this work, we follow the latest trends on developing a multiple-purpose sensor platform capable of dynamically adapting its duty cycle to the application [28]. Different from the early proposals introduced in the literature, the proposed power management mechanism consists basically on turning on/off the power to be delivered to the sensors, rather than dealing with the radio system.…”

Unbalanced Threshold Based Distributed Data Collection Scheme in Multisink Wireless Sensor Networks

“…Therefore, data collection schemes in wireless sensor network should be light weight and energy efficient. Two representative types of data collection schemes in wireless sensor networks are spatial-temporal correlation based data predication schemes [3][4][5] and distributed source coding schemes [6][7][8].…”