SPEED: Stateless Real-time Communication for Timely Response


SPEED design is guided by the key observation that unlike wired networks, where the delay is independent of the physical distance between the source and destination, in multi-hop wireless sensor networks, the end-to-end delay depends on not only single hop delay, but also on the distance a packet travels. In view of this, the key design goal of the SPEED algorithm is to support a soft real-time communication service with a desired delivery speed across the sensor network, so that end-to-end delay is proportional to the distance between the source and destination.

SPEED Architecture

     SPEED is an adaptive real-time routing protocol that aims to reduce the end-to-end deadline miss ratio in sensor networks. Similar to geographic routing, each node only  maintains the states of one-hop neighbors. The core of SPEED is feedback-based adaptation  algorithms that enforce per-hop speed in face of unpredictable traffic. The first adaptation mechanism is a neighborhood feedback loop on each node that periodically computes the probability of forwarding a packet to every neighbor  based on its measured delay and per-hop speed violation in the last sampling  period. The feedback loop ensures that more congested neighbors (with longer  delays and higher miss ratios) get lower probabilities of receiving packets.  When all of its neighbors have deadline misses in the last sampling period, a  node performance rate control to drops packets. The packet-drops subsequently cause upstream neighbors to  redirect packets away from it, a process called back pressure rerouting. The  back pressure can propagate upstream until it reaches outside the congestion  region or the sources. The combination of neighborhood feedback loop and  back pressure rerouting significantly enforces the per-hop deadline in steady  states and reduces the end-to-end deadline miss ratio.

feedback-based adaptation controller

     Simulation experiments  given in  showed that SPEED can achieve significantly lower deadline  miss ratio than geographic routing, DSR, and AODV in face of sudden congestion.  Meanwhile, SPEED's number of overhead packets is comparable to geographic routing and significantly smaller than DSR and AODV. SPEED demonstrates that localized feedback control is a promising approach for real-time communication  in sensor networks. Remaining challenges in this direction include establishing  stability analysis and providing statistical guarantees on end-to-end delays.

For more detail see PDF


กก

กก