Enhancing user experience by exploiting real-time QoS measurements

Jarmo Prokkola
VTT Technical Research Centre of Finland
jarmo.prokkola(at)vtt.fi

The project goal was to enhance user experience in heterogeneous networks with solutions that are easy to use and can be made commercially available already today via existing systems. This goal should be achieved without requiring architectural network modifications, special network protocols, or application support. This is different from an earlier ITEA Easy Wireless project (2004–2007), which targeted end-to-end QoS: The project proved that end-to-end QoS can be reached, but it requires serious updates to the network infrastructure, which is not commercially feasible in the near future.

EW-2 playground

EW-2 approached its goal in different ways, as shown in figure 1. Before starting to search solutions for the problems, they first need to be understood. Thus, a major focus was put on network and QoS monitoring and analysis, which is done both by heavy-weight network analyses and light-weight QoS monitoring solutions. By better understanding the network problems, we can, for example, set up the wireless networks in an optimized way. One of the key aspects in the project was to find new ways in exploiting the measured real-time performance information.

From QoS measurements to enhanced user experience

We know that the prevailing network conditions are the key to enhance the user experience. QoS monitoring shows accurately how the user’s applications are performing over the network path. When monitoring in real-time one can react quickly to the changing conditions. Monitoring is realized, for example, through a light-weight distributed passive QoS monitoring tool named Qosmet. Triggering reactions is, however, not straightforward, since one needs to understand what the results mean. This leads to situation-aware measurements, i.e., we need to understand which type of applications are running, and then, how different QoS parameters interact with those applications. For example, high delay is not a problem for streaming video, but it is a problem for real-time conversational video.

For this purpose, the project participated in developing a light-weight two-phased traffic classifier, which runs even on terminal devices, recognizing the user’s applications based on statistical analysis. With application information, the newly developed GQoSM (Generic QoS Measure) algorithm indicates with a single numeric value, how the user’s application performs over the network path. EW-2 also produced the QoS Measurement Control Protocol (QMCP), which enables easy remote control of QoS measurements and results collection for any QMCP-capable third party software, as shown in figure 2.

Figure 2: A high-level architectural view of Qosmet solution

Intelligent mobility

EW-2 worked on intelligent mobility, where the measured QoS information was utilized to trigger handovers. There exist already good mobility enabling protocols like mSCTP, Mobile IP, and HIP, but none of them can determine how the applications are really performing over the network. Thus, by including this information with an external measurement tool and handover intelligence, one can reach QoS based mobility, enabling the user to enjoy an access network whose performance is always fit for the user’s applications. The developed Intelligent Vertical Handover (IVHO) solution enables also additional policies, e.g., one can favour low-cost networks over the high-cost ones, keeping in mind also the performance. The IVHO system is mainly mobile-driven, but includes also an MIH-interoperable (Media Independent Handover) interface for network (e.g., operator) assisted/driven mobility.

Conclusions

EW-2 developed novel solutions, from network planning optimization, through QoS measurements, and access control, to intelligent mobility, all heading to enhance user experience. The promising topics are still at the technology demonstration level, like the new Regressive Access Control method. They will be further explored in other research projects. Many of the solutions are already beyond demonstration level, like the IVHO solution, and will be fine-tuned further. Then, some solutions, like Qosmet, are reaching commercial level, and new commercial pilots, in addition to the industry partners’ product releases, will follow the EW-2 project.

You can find more information on EW-2 at http://www.it.uc3m.es/ew2/, and on QoS measurements at http://www.cnl.fi/qosmet.html.