PROJECT TITLE :

Enhancing the Delay Performance of Dynamic Backpressure Algorithms

ABSTRACT:

For general multi-hop queueing networks, delay optimal network management has sadly been an outstanding drawback. The dynamic backpressure (BP) algorithm elegantly achieves throughput optimality, however does not yield smart delay performance in general. In this paper, we tend to obtain an asymptotically delay optimal control policy, which resembles the BP algorithm in basing resource allocation and routing on a backpressure calculation, however differs from the BP algorithm in the form of the backpressure calculation employed. The distinction suggests a doable reason for the unsatisfactory delay performance of the BP algorithm, i.e., the myopic nature of the BP management. Motivated by this new affiliation, we tend to introduce a new class of enhanced backpressure-based mostly algorithms that incorporate a general queue-dependent bias perform into the backpressure term of the ancient BP algorithm to improve delay performance. These enhanced algorithms exploit queue state data beyond one hop. We tend to prove the throughput optimality and characterize the utility-delay tradeoff of the improved algorithms. We more focus on two specific distributed algorithms within this category, that have demonstrably improved delay performance furthermore acceptable implementation complexity.