engleski

Assessing compression algorithms on IoT sensor nodes

With the ever-growing presence of IoT devices in everyday applications, the sheer volume of data exchanged is increasing rapidly. Communication channels' bandwidth used in these applications often presents a bottleneck and limits the frequency at which data can be exchanged, making application of compression beforehand a viable approach. In this paper, we analyze some of the well known lossless compression algorithms capable of seamless integration as software modules on embedded devices used in (near) realtime applications. Low processing and memory resources available on them as well as unsteadiness of aforementioned communication channels were taken into consideration in algorithm selection. Attained compression ratio is evaluated on data harvested from smart meters, whilst time and energy consumed during processing were analyzed within a measurement framework purposefully built for this research. We show that considerable gains can be attained in the volume of data that can be sent from devices based on commonly used microcontrollers, with only a modest toll on time and overall energy used.

IoT, compression

nije evidentirano