3 results
Filters
Settings
Now showing1 - 3 of 3
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Scheduling recurring and dependent tasks in EH-WSNs(2020-09); The recent growth of application variety for Energy Harvesting Wireless Sensor Network (EH-WSNs) poses new demands on matching energy consumption of sensors and actuators with energy harvest. High precision sensing—e.g., measuring gas concentration or fine particle matter—offers valuable insight into the environment yet increases program complexity due to dependent deadlines of multiple involved sensing tasks. Execution of all tasks at once meets all deadlines but causes a high energy demand in a short time period. Due to the small energy buffers of EH-WSNs, solely executing all tasks in batch and adapting the execution interval likely prevents completion of all tasks and threatens availability of the system. This raises the need for task scheduling that guarantees dependencies and deadlines inherited from physical phenomena. Standard mathematical solvers meet this end but are too computationally complex for sensor nodes. For this reason, we introduce and discuss light-weight scheduling strategies for dependent tasks restricted by both time and energy constraints. We show via simulations and a real-world experiment that our algorithm outperforms existing scheduling techniques in both execution time and node downtime. We decrease the latter by a factor of 5 maintaining equal activity or increase activity by up to 28% at equal downtimes. Our approach paves the ground for the emerging variety of modern, complex applications for energy-constrained sensors and actuators.Publicationtype: Journal Article154 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Online energy assessment with supercapacitors and energy harvesters(2014-03); ; Combining energy harvesting with energy-aware load adaptation and scheduling enables perpetually operating sensor networks. The practical realization of this goal yet requires methods for reliable and precise holistic online energy assessment. While the building blocks - assessing residual energy, predicting energy intake, and tracing energy consumption - have been studied in detail, the analysis of their interaction on a real platform has been neglected. This paper answers the question, whether these techniques can be easily joined to give a precise and correct picture of a sensor node's energetic state and behavior. For this purpose, we model the energy flow of a prototype energy-harvesting and supercapacitor-powered sensor node. We show that in a real deployment simple supercapacitor models suffice for energy assessment, while capacity calibration is mandatory yet practicable. We evaluate the joint performance of state-of-the-art energy assessment based on an outdoor field test. We verify the system model and show the feasibility of holistic online energy assessment, which tolerates small configuration errors, achievable with a combination of generic configuration and online calibration. We analyze the feasibility of forecasting a node's future energy reserve and find that the presented method produces accurate results for uniformly distributed consumption profiles.Publicationtype: Journal Article21 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Adaptive energy-harvest profiling to enhance depletion-safe operation and efficient task scheduling(2012-03); Forecasting the expected energy harvest enables small-sized energy-harvesting sensor nodes to schedule tasks or adapt the radio duty cycle. This ability ensures depletion-safe and efficient operation. Most energy sources exhibit cyclic patterns of intensity, e.g., the Sun. These patterns show periods with unequal - low versus high and stable versus varying - energy production and heavily depend on a node's location as well as seasonal and environmental changes. Existing forecast algorithms do not exploit these patterns, but create and update forecasts at static and arbitrary points in time, the main knob being the number of updates per cycle. We present a method enabling sensor nodes to adapt to harvesting patterns at runtime. It is designed for seamlessly replacing the static scheme to improve the accuracy of a wide range of existing forecast algorithms. In our evaluation, we show that (i) the adaptive method traces the energy pattern in real-world deployments accurately, (ii) reacts to seasonal and environmental changes, (iii) increases forecast accuracy, and (iv) reduces the number of prediction updates. These achievements enhance depletion-safe operation and efficient task scheduling with fewer recalculations and adjustments of the duty cycle. They also facilitate the exchange of harvesting forecasts for collaborative node tasks, since less information has to be shared.Publicationtype: Journal Article37