Indoor air temperature measurements using ultrasonic travel-time tomography

Acoustic travel time TOMography (ATOM) is a remote measuring technique that uses sound velocity to monitor indoor air temperature distribution. The sound velocity can be derived experimentally from the travel times estimation along the sound propagation paths with known lengths. Previous works often deal with the ATOM measurements within the audible range (20Hz - 20kHz) based on the travel times of early reflections up to third-order reflections. In this study, the focus is shifted from the audible to the ultrasonic range. Ultrasonic tomography, when compared to the audible range, offers greater practicality for air temperature monitoring in real -world scenarios such as office environments. A major advantage is that the room's occupants remain undisturbed by the signal transmission during measurements. Moreover, the smaller dimensions of ultrasound sensors allow for more precise positioning of the sound sources and receivers in the room compared to conventional audible sensors. On the other hand, the strong directivity of the ultrasonic sound sources presents new challenges such as reducing the number of early reflections. This study evaluates the performance of the ATOM measuring system in the ultrasonic frequency range by reconstructing indoor temperature under several thermal conditions. Accordingly, four ultrasonic sound sources and one receiver are mounted within a climate chamber. The main contributions of the developed setup involve enhancing signal-to-noise ratios (SNR) for individual travel times and taking advantage of early reflections to ensure a suitable number of travel times crucial for tomography reconstruction while utilizing a minimal number of transducers. The measurement outcomes demonstrate a 5 dB enhancement in SNR for the ultrasonic measurement configuration compared to the audible measurements. Moreover, the superiority of the activated high -pass filter of the utilized preamplifier in handling temperature measurements is demonstrated. An average root mean square error of 0.31 K reveals the applicability of the ATOM-measuring system for high -resolution monitoring of air temperature distribution in the ultrasonic range.