On an Adaptive-Quasi-Deterministic Transmission Policy Queueing Model

Autorzy Bergquist J.; Gelenbe E.; Sigman K.
Tytuł On an Adaptive-Quasi-Deterministic Transmission Policy Queueing Model
Czasopismo 32nd IEEE MASCOTS'24 Conference on Modeling, Analysis and Simulation of Computer and Telecommunication Systems, October 21-23, 2024, Krakow Poland
Rok 2024
Status Published
URL https://www.researchgate.net/publication/384288968_On_an_Adaptive-Quasi-Deterministic_Transmission_Policy_Queueing_Model#fullTextFileContent
Abstrakt <p>We analyze, further and deeper, a recently proposed<br />
technique for addressing the Massive Access Problem (MAP), an<br />
issue in telecommunications which arises when too many devices<br />
transmit packets to a gateway in quick succession. This technique,<br />
the Adaptive-Quasi-Deterministic Transmission Policy (AQDTP)<br />
is a special case of “traffic shaping” which involves delaying<br />
some packets at the points of origin to alleviate congestion at<br />
the routers. One nice feature of AQDTP is that it loses no<br />
packets and allows an infinite buffer. In this work, to clarify<br />
the approach in a general queueing theory framework, and to<br />
move beyond the original telecommunications application, we<br />
frame these potential delays as time spent at a caf´e by customers<br />
before proceeding to a service facility. We first present some<br />
sample-path results that significantly refine and expand upon<br />
what was shown in previous work, and then present further<br />
results under a general stationary ergodic stochastic framework.<br />
In the sample-path realm, we give conditions that ensure AQDTP<br />
will not change the total delay and sojourn time of any customer<br />
as compared to what that customer would have experienced if<br />
there was no caf´e; but we also prove that AQDTP can never<br />
reduce the total delay. The difference is that, under AQDTP, some<br />
of that delay is spent at the caf´e instead of in the queue/line at the<br />
service facility. In a stochastic framework, our focus is on stability<br />
and constructing proper stationary versions of the model. Under<br />
i.i.d. assumptions we dig deeper by proving Harris recurrence<br />
of an underlying two-dimensional Markov process, and explicitly<br />
find positive recurrent regeneration points.</p>
Wydawca IEEEXplore
PDF Bergquist2024On an Adaptive.pdf