Real-time, content-based communication load reduction in the Internet of Multimedia Things.

Abstract

There is an increasing number of devices available for the Internet of Multimedia Things (IoMT). The demands these ever-more complex devices make are also increasing in terms of energy efficiency, reliability, quality-of-service guarantees, higher data transfer rates, and general security. The IoMT itself faces challenges when processing and storing massive amounts of data, transmitting it over low bandwidths, bringing constrained resources to bear and keeping power consumption under check. This paper’s research focuses on an efficient video compression technique to reduce that communication load, potentially generated by diverse camera sensors, and also improve bit-rates, while ensuring accuracy of representation and completeness of video data. The proposed method applies a video content-based solution, which, depending on the motion present between consecutive frames, decides on whether to send only motion information or no frame information at all. The method is efficient in terms of limiting the data transmitted, potentially conserving device energy, and reducing latencies by means of negotiable processing overheads. Data are also encrypted in the interests of confidentiality. Video quality measurements, along with a good number of Quality-of-Service measurements demonstrated the value of the load reduction, as is also apparent from a comparison with other related methods.