More Like this

1. REHASH: A Flexible, Developer Focused, Heuristic Adaptation Platform for Intermittently Powered Computing

   https://doi.org/10.1145/3478077  

   Bakar, Abu; Ross, Alexander G.; Yildirim, Kasim Sinan; Hester, Josiah (September 2021, Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies)

   Battery-free sensing devices harvest energy from their surrounding environment to perform sensing, computation, and communication. This enables previously impossible applications in the Internet-of-Things. A core challenge for these devices is maintaining usefulness despite erratic, random or irregular energy availability; which causes inconsistent execution, loss of service and power failures. Adapting execution (degrading or upgrading) seems promising as a way to stave off power failures, meet deadlines, or increase throughput. However, because of constrained resources and limited local information, it is a challenge to decide when would be the best time to adapt, and how exactly to adapt execution. In this paper, we systematically explore the fundamental mechanisms of energy-aware adaptation, and propose heuristic adaptation as a method for modulating the performance of tasks to enable higher sensor coverage, completion rates, or throughput, depending on the application. We build a task based adaptive runtime system for intermittently powered sensors embodying this concept. We complement this runtime with a user facing simulator that enables programmers to conceptualize the tradeoffs they make when choosing what tasks to adapt, and how, relative to real world energy harvesting environment traces. While we target battery-free, intermittently powered sensors, we see general application to all energy harvesting devices. We explore heuristic adaptation with varied energy harvesting modalities and diverse applications: machine learning, activity recognition, and greenhouse monitoring, and find that the adaptive version of our ML app performs up to 46% more classifications with only a 5% drop in accuracy; the activity recognition app captures 76% more classifications with only nominal down-sampling; and find that heuristic adaptation leads to higher throughput versus non-adaptive in all cases. 

   more » « less
2. Greentooth: Robust and Energy Efficient Wireless Networking for Batteryless Devices

   https://doi.org/10.1145/3649221  

   Babatunde, Simeon; Alsubhi, Arwa; Hester, Josiah; Sorber, Jacob (May 2024, ACM Transactions on Sensor Networks)

   Communication presents a critical challenge for emerging intermittently powered batteryless sensors. Batteryless devices that operate entirely on harvested energy often experience frequent, unpredictable power outages and have trouble keeping time accurately. Consequently, effective communication using today’s low-power wireless network standards and protocols becomes difficult, particularly because existing standards are usually designed to support reliably powered devices with predictable node availability and accurate timekeeping capabilities for connection and congestion management. In this article, we present Greentooth, a robust and energy-efficient wireless communication protocol for intermittently powered sensor networks. It enables reliable communication between a receiver and multiple batteryless sensors using Time Division Multiple Access–style scheduling and low-power wake-up radios for synchronization. Greentooth employs lightweight and energy-efficient connections that are resilient to transient power outages, while significantly improving network reliability, throughput, and energy efficiency of both the battery-free sensor nodes and the receiver—which could be untethered and energy constrained. We evaluate Greentooth using a custom-built batteryless sensor prototype on synthetic and real-world energy traces recorded from different locations in a garden across different times of the day. Results show that Greentooth achieves 73% and 283% more throughput compared to Asynchronous Wake-up on Demand MAC and Receiver-Initiated Consecutive Packet Transmission Wake-up Radios, respectively, under intermittent ambient solar energy and over 2× longer receiver lifetime. 

   more » « less
3. On the Accuracy of Network Synchronization Using Persistent Hourglass Clocks

   https://doi.org/10.1145/3362053.3363497  

   Çürük, Eren; Yıldırım, Kasim Sinan; Pawelczak, Przemyslaw; Hester, Josiah (January 2019, Proceedings of the 7th International Workshop on Energy Harvesting & Energy-Neutral Sensing Systems)

   Batteryless sensor nodes compute, sense, and communicate using only energy harvested from the ambient. These devices promise long maintenance free operation in hard to deploy scenarios, making them an attractive alternative to battery-powered wireless sensor networks. However, complications from frequent power failures due to unpredictable ambient energy stand in the way of robust network operation. Unlike continuously-powered systems, intermittently-powered batteryless nodes lose their time upon each reboot, along with all volatile memory, making synchronization and coordination difficult. In this paper, we consider the case where each batteryless sensor is equipped with a hourglass capacitor to estimate the elapsed time between power failures. Contrary to prior work that focused on providing a continuous notion of time for a single batteryless sensor, we consider a network of batteryless sensors and explore how to provide a network-wide, continuous, and synchronous notion of time. First, we build a mathematical model that represents the estimated time between power failures by using hourglass capacitors. This allowed us to simulate the local (and continuous) time of a single batteryless node. Second, we show--through simulations--the effect of hourglass capacitors and in turn the performance degradation of the state of the art synchronization protocol in wireless sensor networks in a network of batteryless devices. 

   more » « less
4. Protean: An Energy-Efficient and Heterogeneous Platform for Adaptive and Hardware-Accelerated Battery-Free Computing

   https://doi.org/10.1145/3560905.3568561  

   Bakar, Abu; Goel, Rishabh; de Winkel, Jasper; Huang, Jason; Ahmed, Saad; Islam, Bashima; Pawełczak, Przemysław; Yıldırım, Kasım Sinan; Hester, Josiah (November 2022, Proceedings of the 20th ACM Conference on Embedded Networked Sensor Systems (SenSys'22))

   Battery-free and intermittently powered devices offer long lifetimes and enable deployment in new applications and environments. Unfortunately, developing sophisticated inference-capable applications is still challenging due to the lack of platform support for more advanced (32-bit) microprocessors and specialized accelerators---which can execute data-intensive machine learning tasks, but add complexity across the stack when dealing with intermittent power. We present Protean to bridge the platform gap for inference-capable battery-free sensors. Designed for runtime scalability, meeting the dynamic range of energy harvesters with matching heterogeneous processing elements like neural network accelerators. We develop a modular "plug-and-play" hardware platform, SuperSensor, with a reconfigurable energy storage circuit that powers a 32-bit ARM-based microcontroller with a convolutional neural network accelerator. An adaptive task-based runtime system, Chameleon, provides intermittency-proof execution of machine learning tasks across heterogeneous processing elements. The runtime automatically scales and dispatches these tasks based on incoming energy, current state, and programmer annotations. A code generator, Metamorph, automates conversion of ML models to intermittent safe execution across heterogeneous compute elements. We evaluate Protean with audio and image workloads and demonstrate up to 666x improvement in inference energy efficiency by enabling usage of modern computational elements within intermittent computing. Further, Protean provides up to 166% higher throughput compared to non-adaptive baselines. 

   more » « less
5. MANIC: A $$19\mu\mathrm{W}$$ @ 4MHz, 256 MOPS/mW, RISC-V microcontroller with embedded MRAM main memory and vector-dataflow co-processor in 22nm bulk finFET CMOS

   https://doi.org/10.1109/ISCAS46773.2023.10181809  

   Gobieski, Graham; Atli, Oguz; Erbagci, Cagri; Mai, Ken; Beckmann, Nathan; Lucia, Brandon (May 2023, 2023 IEEE International Symposium on Circuits and Systems (ISCAS))

   Whether powered by a battery or energy harvested from the environment, low-power (LP) sensor devices require extreme energy efficiency. These sorts of devices are becoming pervasive, running increasingly sophisticated applications in inhospitable environments. We present Manic, an energy-efficient microcontroller (MCU) augmented with a vector-dataflow (VDF) co-processor. The testchip taped out on a 22nm bulk finFET CMOS process demonstrates that Manic is 60% more energy-efficient than a baseline, scalar, low-power MCU, achieving peak efficiency of 256 MOPS/mW (2.6× prior work) while consuming only 19.1μW (@4MHz). To make the system viable for intermittently powered applications that require non-volatile storage, Manic includes a 256KB embedded MRAM. 

   more » « less