Traffic Measurement and Analysis


Effective and Available bandwidth Estimation

This research theme investigates the development of algorithms for effective and available bandwidth measurements. The effective bandwidth of a traffic flow is the maximum bandwidth required by a link to meet the quality of service targets on packet delay of that traffic flow. Understanding this measurement for different traffic types is fundamental for developing QoS provisioning strategies.

Available bandwidth of an end-to-end path is usually defined as its remaining capacity, that is the amount of traffic that can be sent along the path without incurring congestion. Common methods for estimating this are based on injecting probe trains of traffic into the path at different rates and observing the packet train behaviour at the receiving end. This short-term measurement can be used for a variety of applications such as admission control, rate control and server selection over the end-to-end path.

Effective Capacity Estimation of Wireless Links

Current approaches to managing Quality of Service (QoS) of delay sensitive services within the wireless domain rely on complicated analytical models, which are designed to represent the requirements of both the traffic and the capabilities of the wireless channel. Such approaches can be suitable within particular environment conditions where traffic and wireless conditions are known. However when conditions such as traffic characteristics and wireless technologies are continuously changing, the methods of identifying traffic requirements and wireless capabilities must be agnostic of application, and environment condition.

The objective of this research is to develop a QoS control framework, which can be independent of these heterogeneities. Our proposed methodology focuses on measurement based techniques to identify both QoS requirements of traffic and QoS provisioning capabilities of wireless channels under various scenarios such as metropolitan area wireless mesh networks, mobile wireless vehicular networks and LTE deployments. The research focuses on methods of estimating the effective capacity of the wireless link. This is the upper bound of throughput the wireless link can support while meeting stochastic Quality of Service targets.

Intrinsic Montoring in IPv6 Networks

In general there is a disconnect between the logical network path and the physical network path traffic flows through as routing protocols can allow traffic with a particular flow to take different physical paths through a network. It is therefore difficult to directly monitor the characteristics of the actual path taken by a traffic flow through the network. With the benefits of IPv6, additional hop by hop instructions can be triggered along a path using the hop by hop extension header. This research focuses on investigating the possibility of an Intrinsic network monitoring technique where packets will trigger and collect monitoring information from the network on a hop by hop basis. The technique can lead to the possibility of directly mapping traffic flows to network paths and also discover node specific information related to a physical network path.

Members

Dr. Alan Davy, Dr. Lei Shi, Brian Meskill

Collaborators

Prof Jordi Domingo Pascual, Dr. Albert Cabellos – Universitat Politecnica de Catalynua

Projects

(IRC) MAVEN, (EI) QOSPLAN, EI LANTERN, SFI FAME

Recent Publications

Haytham Assem, Mohamed Adel, Brendan Jennings, David Malone, Jonathan Dunne, and Pat O Sullivan. A Generic Algorithm for Mid-call Audio Codec Switching. In Proc. 1st IFIP/IEEE International Workshop on QoE Centric Management (QCMan 2013), to appear, 2013.

Alan Davy, Brian Meskill, and Jordi Domingo Pascual. An Empirical Study of Effective Capacity Throughputs in 802.11 Wireless Networks. In Proc. IEEE Global Communications Conference (Globecom 2012), pages 1788 – 1793, 2012.

Alan Davy, Brendan Jennings, and Dmitri Botvich. QoSPlan: A Measurement Based Quality of Service aware Network Planning Framework. Journal of Network and Systems Management, to appear, –:1–36, June 2013. (doi:10.1007/s10922-012-9243-3)

Brian Meskill, Alan Davy, and Brendan Jennings. Revenue-Maximizing Server Selection and Admission Control for IPTV Content Servers using Available Bandwidth Estimates. In Proc. 2012 IEEE/IFIP Network Operations and Management Symposium (NOMS 2012), pages 319–326. IEEE, 2012.

Cathal O Connor, Alan Davy, and Brendan Jennings. Controlling the Transfer of Kinect Data to a Cloud-hosted Games Platform. In Proc. 23rd ACM Workshop on Network and Operating Systems Support for Digital Audio and Video (NOSSDAV 2013), 2013.

Runxin Wang, Lei Shi, and Brendan Jennings. Ensemble Classifier for Traffic in Presence of Changing Distributions. In Proc. 18th IEEE Symposium on Computers and Communications (ISCC 2013), to appear, 2013.