Adaptive Transmission Rate Control for Decentralized Sensing Applications

Cellular sidelink can serve as an effective means to enable device-to-device communication for local information sharing, which enables an efficient implementation of, e.g., transportation safety and decentralized mobile crowdsensing applications. Such applications rely on periodic broadcast messages and thus may face scalability issues as network utilization increases with the number of users within a resource sharing domain (RSD). In this paper, we propose an adaptive transmission rate control algorithm for decentralized sensing applications (ARC-DSA) to limit the combined data rate, irrespective of the number of devices within an RSD. Our algorithm leverages device density information that is locally collected, aggregated, and shared. Implemented at the application layer, it limits the application-specific rate to a target fraction of the sidelink resources. To ascertain effective information sharing despite these resource limits, we present three different content selection strategies based on the age of information and distance. We evaluate our designs in a system-level simulation using CrowNet, a framework based on the OMNeT++ ecosystem. We first choose a single RSD for microbenchmarking, followed by a realistic urban scenario comprising multiple RSDs. The results indicate that ARC-DSA can successfully limit the total rate consumed by applications for a wide range of target rates and node densities in single and multi-RSD scenarios. Furthermore, content selection strategies allow controlling the trade-off between the freshness of the information and the range in which the information is available. The findings significantly contribute to the understanding of how adaptive rate control and content-selection strategies can be applied to existing and future decentralized sensing applications.