Performance Evaluation of Heterogeneous Network Scenarios

Packet loss, delay and jitter degrade the quality of services like VoIP (Voice over IP) or Video Streaming over IP networks. In real networks, an experimental measure of these parameters is fundamental in the planning process of new services over novel network infrastructures. Furthermore, currently networks are complex heterogeneous systems (in terms of access network technologies, end-users’ devices, Operating Systems and finally end-users’ application). This heterogeneity exacerbates even more the need of a real assessment of the Quality of Service metrics. Stepping fro this consideration, in this Chapter we provide an empirical performance study of real heterogeneous wireless network with respect to delay, jitter, throughput and packet loss, in UDP and TCP environments, by using the innovative tool D-ITG (Distributed Internet Traffic Generator). We also use the concept of “Service Condition” as a mechanism to cope with issues related to dynamically changing network service scenario and to formally define a parametric and systemic approach to the performance evaluation. Results presented in this Report can be used as performance references for development of wireless communication applications over multi-service heterogeneous networks. Indeed, we perform a large set of measurements with a wide set of hardware. Then we take the results of our measurements and by analyzing the similarities and the differences we understand what we can generalize and what not. We investigate the impact of mechanisms and protocols on the performance metrics and we try to prove which protocol element is a limiting factor or which devices characteristics affect the measured metrics. In addition, thanks to the wide range of provided results (as well as the public traffic traces, available on request), a reader could actually make use of the information provided in this Report for his purposes. More precisely, this Report presents results of our performance evaluation activities in the following scenario:

[1]  Haitao Wu,et al.  Performance of reliable transport protocol over IEEE 802.11 wireless LAN: analysis and enhancement , 2002, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[2]  Kishor S. Trivedi,et al.  Composite performance and availability analysis of wireless communication networks , 2001, IEEE Trans. Veh. Technol..

[3]  Kaveh Pahlavan,et al.  Performance monitoring of a wireless campus area network , 1997, Proceedings of 22nd Annual Conference on Local Computer Networks.

[4]  Petri Mähönen,et al.  TCP performance issues over wireless links , 2001, IEEE Commun. Mag..

[5]  J. J. Garcia-Luna-Aceves,et al.  Delay analysis of IEEE 802.11 in single-hop networks , 2003, 11th IEEE International Conference on Network Protocols, 2003. Proceedings..

[6]  H. Balakrishnan,et al.  A comparison of mechanisms for improving TCP performance over wireless links , 1999, SIGCOMM '96.

[7]  Chai-Keong Toh,et al.  Evaluating the communication performance of an ad hoc wireless network , 2002, IEEE Trans. Wirel. Commun..

[8]  George C. Polyzos,et al.  TCP and UDP performance over a wireless LAN , 1999, IEEE INFOCOM '99. Conference on Computer Communications. Proceedings. Eighteenth Annual Joint Conference of the IEEE Computer and Communications Societies. The Future is Now (Cat. No.99CH36320).

[9]  Rui Pedro Lopes,et al.  Performance analysis of a wireless MAN , 2001, Proceedings IEEE International Symposium on Network Computing and Applications. NCA 2001.

[10]  Ahmed K. Elhakeem,et al.  Performance analysis of ad hoc wireless LANs for real-time traffic , 2003, IEEE J. Sel. Areas Commun..

[11]  B. Bing Measured performance of the IEEE 802.11 wireless LAN , 1999, Proceedings 24th Conference on Local Computer Networks. LCN'99.

[12]  Randy H. Katz,et al.  A trace-based approach for modeling wireless channel behavior , 1996, Winter Simulation Conference.

[13]  D. Clawin,et al.  Wireless LAN performance under varied stress conditions in vehicular traffic scenarios , 2002, Proceedings IEEE 56th Vehicular Technology Conference.

[14]  M. Garetto,et al.  Performance Analysis of the 802 . 11 Distributed Coordination Function under Sporadic Traffic , 2004 .