An Interactive Context-aware Power Management Technique for Optimizing Sensor Network Lifetime

Abstract

A key problem in sensor networks equipped with renewable energy sources is deciding how to allocate energy to various tasks (sensing, communication etc.) over time so that the deployed network continues to gather high-quality data. The state-of-the-art energy allocation algorithm takes into account current battery level and harvesting energy and fairly allocates as much energy as possible along the time dimension. In this paper we show that by not considering application-context this approach leads to very high and uniform sampling rates. However, sampling the environment at fixed predefined intervals is neither possible (need to accommodate system failures) nor desirable (sampling rate might not capture an important event with desired fidelity). To that end, in this paper we propose a novel interactive power management technique that adapts sampling rate as a function of both application-level context (e.g., user request) and system-level context (e.g harvesting energy availability). We vary several key parameters including application request patterns, geographic locations, time slot length, battery end point voltage and evaluate the performance of our approach in terms of energy efficiency and accuracy. Our simulations use sensor data and system specifications (battery and solar panel specs, sensing and communication costs) from a real sensor network deployment. Our results show that the proposed approach saves significant amounts of energy by avoiding oversampling when application does not need it while using this saved energy to support sampling at high rates to capture events with necessary fidelity when needed. The computational complexity of our approach is lower (O(n)) than the state-of-the-art noninteractive energy allocation algorithm (O(n 2)).

A distributed real-time data prediction and adaptive sensing approach for wireless sensor networks

“…Oppositely, if the prediction does not respect the error budget, the sensor discards it and transmits to the Sink the real reading, which we will refer to as "correction packet". Once the Sink receives the packet, they both update the value of d, by subtracting N R from the correction value x[k] and dividing the result by the time difference between the reception of these tow measurements, as shown in the equation (4). And once again to keep track of the last received value for potential future update, the old value of N R is replaced in the memory by the newly received measurement…”

“…In previous studies, authors studied several energy-saving schemes for wireless monitoring operations such as: data aggregation [16,17], data compression [18,19], adaptive sampling [2,3,4,5,6,7,20] and data prediction [8,9,10,11,12,13].…”

“…In [4] the sampling rate of the sensor node is adapted by taking into consideration both the system and the application context levels. For instance, the availability of the energy for harvesting represents the system context.…”

“…Due to the nature of WSNs, sensor data tend to change smoothly over time and it contains a significant chunk of redundant information. Therefore, to reduce the number of sampled data some researchers proposed several adaptive sampling techniques [2,3,4,5,6,7] that dynamically increase or decrease the sampling rate of a sensor according to the level of variance between collected data over a certain period of time. This approach prevents the sensor from collecting redundant information.…”

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

“…Compression [12]- [15] and aggregation [16]- [18] are two techniques aiming to reduce the amount of data routed through the network [19]. The former focus on compressing the data before transmission to the upper node in the network hierarchy and the latter filters and clean the data by removing redundant information before routing these data to the Sink station.…”