Abstract-This paper presents the implementation and deployment of a compute/memory intensive non-parametric Bayesian machine learning algorithm on a microcontroller unit (MCU) to estimate room occupancy in a Smart Room using a single analogue PIR sensor. We envisage an IoT device consisting of a resource-constrained MCU, PIR sensor and a battery running the occupancy estimation algorithm and operating over days or months without recharging or replacing the battery. Both hardware-independent and hardware-dependent optimizations are performed to reduce memory footprint and yet provide acceptable real-time performance while consuming less energy. We show a significant reduction in the on-chip memory usage in the MCUs by the algorithm through optimisation of the machine learning models and of the static memory footprint and dynamic memory usage. We also show that a low-end MCU does not meet the real-time requirements of the application without causing high average power consumption. However, a moderately highperformance MCU with a higher clock frequency and hardware floating-point unit provides 19x improvement in the execution time of the algorithm, better meeting the real-time specification of the application and reducing power consumption. Further, we estimate the battery lifetime of the IoT device if it operates continuously in a Smart Room. With a typical size battery, an IoT device consisting of a Cortex-M4F MCU and PIR sensor can operate for more than a month without replacement or recharging of the battery while running the compute-intensive Bayesian machine learning algorithm.

Abstract-Multi/Many-core systems are prevalent in several application domains targeting different scales of computing such as embedded and cloud computing. These systems are able to fulfil the ever-increasing performance requirements by exploiting their parallel processing capabilities. However, effective power/energy management is required during system operations due to several reasons such as to increase the operational time of battery operated systems, reduce the energy cost of datacenters, and improve thermal efficiency and reliability. This article provides an extensive survey of learning-based run-time power/energy management approaches. The survey includes a taxonomy of the learning-based approaches. These approaches perform design-time and/or run-time power/energy management by employing some learning principles such as reinforcement learning. The survey also highlights the trends followed by the learning-based run-time power management approaches, their upcoming trends and open research challenges.

Abstract-Stereo vision is a methodology to obtain depth in a scene based on the stereo image pair. In this paper we introduce a Discrete Wavelet Transform (DWT) based methodology for a state-of-the-art disparity estimation algorithm, that resulted in significant performance improvement in terms of speed and computational complexity. In the initial stage of the proposed algorithm, we apply DWT to the input images, reducing the number of samples to be processed in subsequent stages by 50%, thereby decreasing computational complexity and improving processing speed. Subsequently the architecture has been designed based on this proposed methodology and prototyped on a Xilinx Virtex-7 FPGA. The performance of the proposed methodology has been evaluated against four standard Middlebury Benchmark image pairs viz. Tsukuba, Venus, Teddy and Cones. The proposed methodology results in improvement of about 44.4% cycles per frame, 52% frames per second and 61.5% and 59.6% LUT and register utilization respectively, compared with state-of-the-art designs.

No abstract