Answers are grounded in this paper and cite their sources.
We propose an innovative framework to improve communication reliability of IoT sensory traffic, when the network is subject to congestion. The proposed architecture is based on information dispersion and relies on the fact that congestion occurs in space-and time-differing parts of the network. Our approach uses encoding and spreading of data flows across various parts of the network, while ensuring that the original data could be recovered at the destination with sufficient accuracy, even if only parts of the encoded data survive the congestion losses. The encoding is based on spatial use of erasure coding, while the recovery is based on Dictionary-based Compressed Sensing. We term this approach Congestion-Tolerant Framework. Our work is specifically targeted at the paradigm of IoT that promises to provide wide-area sensory connectivity to devices on a massive scale, interconnecting billions of such devices. Our framework is applicable to both, real-time and non-real-time IoT networking paradigms.
No takes yet. Share an insight, caveat, or question.
Haas et al. (2019) studied this question.