Scheduling of Wireless Edge Networks for Feedback-Based Interactive Applications

Interactive applications with automated feedback will largely influence the design of future networked infrastructures. In such applications, status information about an environment interest is captured and forwarded to a compute node, which analyzes generates message. Timely processing forwarding must ensure be still applicable; thus, quality-of-service parameter for end-to-end latency over entire loop. By modelling communication loop as two-hop network, we address problem allocating network resources in order minimize delay violation probability (DVP), i.e. exceeding target value. We investigate queue states along path on performance semi-static dynamic scheduling policies. The former determine schedule prior transmission packet, while latter benefit from time evolves reallocate slots depending queue’s evolution. proposed policies evaluated variations several system parameters comparison baselines. Results show that policy achieves close-to-optimal DVP outperforms state-of-the-art algorithms.