Energy Measurement and Profiling of Internet of Things Devices

As technological improvements in hardware and software have grown in leaps and bounds, the presence of IoT devices has been increasing at a fast rate. Profiling and minimizing energy consumption on these devices remains to be an an essential step towards employing them in various application domains. Due to the large size and high cost of commercial energy measurement platforms, the research community has proposed alternative solutions that aim to be simple, accurate, and user friendly. However, these solutions are either costly, have a limited measurement range, or low accuracy. In addition, minimizing energy consumption in IoT devices is paramount to their wide deployment in various IoT scenarios. Energy saving methods such as duty-cycling aim to address this constraint by limiting the amount of time the device is powered on. This process needs to be optimized, as devices are now able to perform complex, but energy intensive tasks due to advancements in hardware. The contributions of this paper are two-fold. First we develop an energy measurement platform for IoT devices. This platform should be accurate, low-cost, easy to build, and configurable in order to scale to the high volume and varying requirements for IoT devices. The second contribution is improving the energy consumption on a Linux-based IoT device in a duty-cycled scenario. It is important to profile and optimize boot up time and shutdown time, and improve the way user applications are executed. EMPIOT is an accurate, low-cost, easy to build, and flexible power measurement platform. We present the hardware and software components that comprise EMPIOT and then study the effect of various design parameters on accuracy. In particular, we analyze the effect of driver, bus speed, input voltage, and buffering mechanisms on sampling rate, measurement accuracy, and processing demand. In addition to this, we also propose a novel calibration technique and report the calibration parameters under different settings. In order to demonstrate EMPIOT\u27s scalability, we evaluate its performance against a ground truth on five different devices. Our results show that for very low-power devices that utilize 802.15.4 wireless standard, measurement error is less than 4%. In addition, we obtain less than 3% error for 802.11-based devices that generate short and high power spikes. The second contribution is the optimization the energy consumption of IoT devices in a duty cycled scenario by reducing boot up duration, shutdown duration, and user application duration. To this end, we study and improve the amount of time a Linux-based IoT device is powered on to accomplish its tasks. We analyze the processes of system boot up and shutdown on two platforms, the Raspberry Pi 3 and Raspberry Pi Zero Wireless, and enhance duty-cycling performance by identifying and disabling time consuming or unnecessary units initialized in the userspace. We also study whether SD card speed and SD card capacity utilization affect boot up duration and energy consumption. In addition, we propose Pallex, a novel parallel execution framework built on top of the systemd init system to run a user application concurrently with userspace initialization. We validate the performance impact of Pallex when applied to various IoT application scenarios: (i) capturing an image, (ii) capturing and encrypting an image, (iii) capturing and classifying an image using the the k-nearest neighbor algorithm, and (iv) capturing images and sending them to a cloud server. Our results show that system lifetime is increased by 18.3%, 16.8%, 13.9% and 30.2%, for these application scenarios, respectively