Spatial–temporal routing for supporting end-to-end hard deadlines in multi-hop networks

Abstract

We consider the problem of routing packets with end-to-end hard deadlines in multihop communication networks. This is a challenging problem due to the complex spatial–temporal correlation among flows with different deadlines. To tackle this problem, we introduce the concepts of virtual links and virtual routes to incorporate end-to-end deadline constraints into routing and propose a novel virtual queue architecture to guide the spatial–temporal routing where the routing algorithm specifies where and when a packet should be routed. For the frame-based periodic traffic, the proposed policies can support any arrival within throughput region when the ratio between the link capacity and the packet size is sufficiently large. For the general traffic, we integrate a constrained resource-pooling heuristic into the spatial–temporal routing, which improves the delivery ratio and performs well. Our extensive simulations show that the policies outperform traditional policies such as backpressure and earliest-deadline-first (EDF) for more general traffic flows in multihop communication networks.