TY - JOUR
T1 - Factor interaction on service delivery in mobile ad hoc networks
AU - Vadde, Kiran K.
AU - Syrotiuk, Violet
N1 - Funding Information:
Manuscript received October 1, 2003; revised March 15, 2004. This paper was supported in part by the National Science Foundation (NSF) under Grant ANI-0240524. The authors are with the Computer Science and Engineering Department, Arizona State University, Tempe, AZ 85287-8809 USA (e-mail: kvadde@asu.edu; syrotiuk@asu.edu). Digital Object Identifier 10.1109/JSAC.2004.829351
PY - 2004/9
Y1 - 2004/9
N2 - Delay sensitive applications are driving the need for the support of quality-of-service (QoS) in mobile ad hoc networks. In this paper, we use statistical design of experiments to study the impact of factors and their interaction on the service delivered. We consider the factors of QoS architecture, routing protocol, medium access control protocol, offered load, and mobility, at mixed levels. Real-time throughput, total throughput, and average delay are used as the measures of service delivery. A statistical analysis of the data collected by simulation using analysis of variance techniques is performed. This allows us to identify both main effects and interactions of factors that best explain the response variables. For average delay several factors of the experiment interact. For both forms of throughput, the impact of the routing protocol is not apparent except as it interacts with other factors, and the factor interactions are Hot as extensive as for delay. Thus, for all response variables the factors cannot be studied in isolation. As well, the analysis provides a means for a system architect to determine the level to set the factors to optimize a specific service delivery metric, or combination of metrics.
AB - Delay sensitive applications are driving the need for the support of quality-of-service (QoS) in mobile ad hoc networks. In this paper, we use statistical design of experiments to study the impact of factors and their interaction on the service delivered. We consider the factors of QoS architecture, routing protocol, medium access control protocol, offered load, and mobility, at mixed levels. Real-time throughput, total throughput, and average delay are used as the measures of service delivery. A statistical analysis of the data collected by simulation using analysis of variance techniques is performed. This allows us to identify both main effects and interactions of factors that best explain the response variables. For average delay several factors of the experiment interact. For both forms of throughput, the impact of the routing protocol is not apparent except as it interacts with other factors, and the factor interactions are Hot as extensive as for delay. Thus, for all response variables the factors cannot be studied in isolation. As well, the analysis provides a means for a system architect to determine the level to set the factors to optimize a specific service delivery metric, or combination of metrics.
KW - Design of experiments
KW - Factor interactions
KW - Mobile ad hoc networks (MANETs)
KW - Service delivery
UR - http://www.scopus.com/inward/record.url?scp=4444382971&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=4444382971&partnerID=8YFLogxK
U2 - 10.1109/JSAC.2004.829351
DO - 10.1109/JSAC.2004.829351
M3 - Article
AN - SCOPUS:4444382971
SN - 0733-8716
VL - 22
SP - 1335
EP - 1346
JO - IEEE Journal on Selected Areas in Communications
JF - IEEE Journal on Selected Areas in Communications
IS - 7
ER -