Energy consumption reduction of indoor temperature measurements with asymmetric numeral system compression
Michał Markiewicz, Henryk Telega, Jarosław Duda, Aleksandra Dujović, Marek Skomorowski, Josef Pieprzyk, Tadeusz Stasiak · Energy and Buildings · 2025
• Power consumption of a wireless temperature sensor decreased by 6 % to 37 %, depending on the transmission intervals. • The asymmetric numeral system used for the compression of the temperature measurements decreased to 103 % of the theoretical limit for compression rate loss (compared to 117 % for Huffman coding). • The back-end server resources offload: resulted in a 66 % reduction of storage space and up to 33 % reduction in processing power. About half of building energy consumption is attributed to heating, ventilation, and air conditioning (HVAC) systems. One method to improve building performance in terms of energy efficiency is the installation of additional sensors that provide more information to the HVAC control. Considering the trade-offs between sensor installation, maintenance cost and resulting benefits, one potential solution is to install wireless sensors. The purpose of this article is to investigate when battery life of wireless sensors can be extended by the asymmetric numeral system data compression algorithm. We used the following method: we measured the temperatures with battery-powered wireless sensors inside a multifamily residential building throughout the period of one year. Then we prepared two variants of sensor firmware that transmits compressed and uncompressed temperature readouts. We measured how much energy was consumed by a physical device during the sensor activity and we estimated the total energy consumption. We found out that it is possible to reduce the power consumption of a wireless temperature sensor by between 6 % and 37 %, depending on the transmission intervals. Additionally, we measured the resources consumed by the backend server required to handle the compressed and uncompressed temperature readouts, achieving a 66 % reduction in storage space and up to 33 % reduction in processing power.