High Precision Hybrid RF and Ultrasonic Chirp-based Ranging for Low-Power IoT Nodes

Abstract

Hybrid acoustic-RF systems offer excellent ranging accuracy, yet they typically come at a power consumption that is too high to meet the energy constraints of mobile IoT nodes. We combine pulse compression and synchronized wake-ups to achieve a ranging solution that limits the active time of the nodes to 1 ms. Hence, an ultra low-power consumption of 9.015 µW for a single measurement is achieved. Measurements based on a proof-of-concept hardware platform show median distance error values below 10 cm. Both simulations and measurements demonstrate that the accuracy is reduced at low signal-to-noise ratios and when reflections occur. We introduce three methods that enhance the distance measurements at a low extra processing power cost. Hence, we validate in realistic environments that the centimeter accuracy can be obtained within the energy budget of mobile devices and IoT nodes. The proposed hybrid signal ranging system can be extended to perform accurate, low-power indoor positioning.

Zero-Crossing Chirp Frequency Demodulation for Ultra-Low-Energy Precise Hybrid RF-Acoustic Ranging of Mobile Nodes

“…The collected data is transmitted back to the base station for further processing. The approach in [9] involves limited processing at the mobile node, yet relies on an energy expensive ADC that prohibits its application in real-time energy constraint devices. We here propose a method to further optimize the mobile nodes' energy consumption.…”

“…In our experiments, we extend the analysis of the acoustic power consumption [9] with the evaluation of the energy needed for the wireless communication and processing. These power measurements were performed with a Happy Gecko with the "Energy Profiler" in Simplicity Studio at a fixed supply voltage of 3.3 V. Table 1 summarizes the power and energy consumed during the sensing, computing and communication operations for the binary sampling method (41 samples) and the ADC sampling method (200 samples).…”

“…This takes almost double the time for the ADC sampling method, while the current drawn in both cases is similar. The numbers for the acoustic power consumption are taken from previous research [9]. The only difference is that in the binary sampling method a comparator is added to the circuit, hence the higher drawn current in active mode.…”