Name
Abstract | ||
|---|---|---|
We attempt to bridge queuing theory and information theory by considering queue lengths as communication channels conveying information about system parameters. A toy problem is proposed and an information-theoretic lower bound is derived on the queue-length/cost tradeoff which matches (in an order sense) the performance of back-pressure algorithms. The proof is based on the idea of “communication simulation,” which gives a “reduction” that uses a queuing system with a certain performance to simulate a communication system with the corresponding performance. The result suggests that there is a fundamental performance penalty to following the minimalist aesthetic of avoiding explicit protocol information. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1109/Allerton.2013.6736626 | Allerton |
Keywords | Field | DocType |
protocols,communication simulation,queueing theory,implicit communication channel,queue lengths,information theory,protocol information,bridge queuing theory,system parameters,back-pressure algorithms | Information theory,Mathematical optimization,Bulk queue,Toy problem,Computer science,Queue,Communications system,Communication channel,Queueing theory,Queue management system,Distributed computing | Conference |
ISSN | ISBN | Citations |
2474-0195 | 978-1-4799-3409-6 | 0 |
PageRank | References | Authors |
0.34 | 7 | 2 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Se Yong Park | 1 | 48 | 21.78 |
| Anant Sahai | 2 | 35 | 4.97 |