M-IIoT System for Reducing Bandwidth Costs in Cloud-Hybrid Multimedia Pipelines

With the advent of cloud computing and 5G, Outside Broadcasting (OB) vehicles and mobile multimedia capturing workflows are moving towards cloud-hybrid approaches for processing and distribution. These evolving multimedia pipelines are championed by broadcasters for their software-enabled flexibility and reduced capita costs. However, they are subject to the Internet's indeterministic traffic and path conditions that undermine the performance of the new workflows. In this paper, a novel Multimedia Industrial Internet-of- Things (M-IIoT) real-time lossless compression scheme is proposed to reduce the bandwidth request to Internet service providers. The proposed scheme is designed to losslessly compress User Datagram Protocol (UDP)-based payloads generated by real-time broadcasts prior to flight over the Internet and similar networks. As the total amount of data exchanged decreases due to the compression process, the bandwidth utilization by intermediate nodes to transmit and retransmit lost packets due to the Internet's environment is reduced due to smaller over-all packet sizes and short average wait times. Using D/M/1 queues, the expected cost savings enabled by the proposed scheme are approximated and are found as a function of transmission bitrates. The proposed solution is simulated using Network Simulator 3 (NS3) with FFMPEG-based source nodes as well as implemented on in-the-field playout devices. The results show a decrease in bandwidth utilization and transmission costs in error free networks as well as networks with induced errors of a minimum of 40.0% and a maximum of 97.0%.