Embedding Artificial Intelligence onto low-power devices is a challenging task that has been partly overcome with recent advances in machine learning and hardware design. Presently, deep neural networks can be deployed on embedded targets to perform different tasks such as speech recognition, object detection or Human Activity Recognition. However, there is still room for optimization of deep neural networks onto embedded devices. These optimizations mainly address power consumption, memory and real-time constraints, but also an easier deployment at the edge. Moreover, there is still a need for a better understanding of what can be achieved for different use cases. This work focuses on quantization and deployment of deep neural networks onto low-power 32-bit microcontrollers. The quantization methods, relevant in the context of an embedded execution onto a microcontroller, are first outlined. Then, a new framework for end-to-end deep neural networks training, quantization and deployment is presented. This framework, called MicroAI, is designed as an alternative to existing inference engines (TensorFlow Lite for Microcontrollers and STM32Cube.AI). Our framework can indeed be easily adjusted and/or extended for specific use cases. Execution using single precision 32-bit floating-point as well as fixed-point on 8- and 16 bits integers are supported. The proposed quantization method is evaluated with three different datasets (UCI-HAR, Spoken MNIST and GTSRB). Finally, a comparison study between MicroAI and both existing embedded inference engines is provided in terms of memory and power efficiency. On-device evaluation is done using ARM Cortex-M4F-based microcontrollers (Ambiq Apollo3 and STM32L452RE).
Wireless sensor networks (WSNs) require an extremely energy-efficient design. As sensor nodes have limited power sources, the problem of autonomy is crucial. Energy harvesting provides a potential solution to this problem. However, as current energy harvesters produce only a small amount of energy and their storage capacity is limited, efficient power management techniques must also be considered. In this article we address the problem of modeling and simulating energy harvesting WSN nodes with efficient power management policies. We propose furthermore a framework that permits to describe and simulate an energy harvesting sensor node by using a high level modeling approach based on power consumption and energy harvesting. The node architectural parameters as well as the on-line power management techniques will also be specified. Two new power management architectures will be introduced, taking into account energy-neutral and negative-energy conditions. Simulations results show that the throughput of a sensor node can be improved up to 50% when compared to a state of the art power management algorithm for solar harvesting WSN. The simulation framework is then used to find an efficient system sizing for a solar energy harvesting WSN node.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.