TY - GEN
T1 - Delay analysis of selective repeat ARQ for a markovian source over a wireless channel
AU - Kim, Jeong Geun
AU - Krunz, Marwan
PY - 1999/8/1
Y1 - 1999/8/1
N2 - In this paper, we analyze the delay performance for a Marko- vian source transported over a wireless channel with time- varying error characteristics. To improve the reliability of the channel, the end points of the wireless link implement a selective-repeat (SR) ARQ error control protocol. We provide an approximate discrete-time analysis of the end- to-end mean packet delay, which consists of transport and resequencing delays. The transport delay, in turn, consists of queueing, transmission/retransmission, and propagation delays. In contrast to previous studies, our analysis accommodates the inherent autocorrelations in both the input traffic and the channel state. Our approximation of the mean transport delay is based on decoupling the dependence of the queueing behavior from the past channel conditions. The exact probability generating function (PGF) of the queue length under ideal SR ARQ is obtained and is combined with the retransmission delay to obtain the mean transport delay. For the resequencing delay, our analysis is performed under heavy-traffic assumptions, hence providing an upper bound on the actual mean resequencing delay. Numerical results and simulations indicate that our approximate analysis is sufficiently accurate for a wide range of parameter values.
AB - In this paper, we analyze the delay performance for a Marko- vian source transported over a wireless channel with time- varying error characteristics. To improve the reliability of the channel, the end points of the wireless link implement a selective-repeat (SR) ARQ error control protocol. We provide an approximate discrete-time analysis of the end- to-end mean packet delay, which consists of transport and resequencing delays. The transport delay, in turn, consists of queueing, transmission/retransmission, and propagation delays. In contrast to previous studies, our analysis accommodates the inherent autocorrelations in both the input traffic and the channel state. Our approximation of the mean transport delay is based on decoupling the dependence of the queueing behavior from the past channel conditions. The exact probability generating function (PGF) of the queue length under ideal SR ARQ is obtained and is combined with the retransmission delay to obtain the mean transport delay. For the resequencing delay, our analysis is performed under heavy-traffic assumptions, hence providing an upper bound on the actual mean resequencing delay. Numerical results and simulations indicate that our approximate analysis is sufficiently accurate for a wide range of parameter values.
UR - http://www.scopus.com/inward/record.url?scp=85028999808&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85028999808&partnerID=8YFLogxK
U2 - 10.1145/313256.313278
DO - 10.1145/313256.313278
M3 - Conference contribution
AN - SCOPUS:85028999808
T3 - Proceedings of the 2nd ACM International Workshop on Wireless Mobile Multimedia, WOWMOM 1999
SP - 59
EP - 66
BT - Proceedings of the 2nd ACM International Workshop on Wireless Mobile Multimedia, WOWMOM 1999
PB - Association for Computing Machinery, Inc
T2 - 2nd ACM International Workshop on Wireless Mobile Multimedia, WOWMOM 1999
Y2 - 20 August 1999
ER -