TCP Performance in Heterogeneous Wireless Networks

The TCP protocol is used by most Internet applications today, including the recent mobile wireless terminals that use TCP for their World-Wide Web, E-mail and other traffic. The recent wireless network technologies, such as GPRS, are known to cause delay spikes in packet transfer. This causes unnecessary TCP retransmission timeouts. This dissertation proposes a mechanism, Forward RTO-Recovery (F-RTO) for detecting the unnecessary TCP retransmission timeouts and thus allow TCP to take appropriate follow-up actions. We analyze a Linux F-RTO implementation in various network scenarios and investigate different alternatives to the basic algorithm. The second part of this dissertation is focused on quickly adapting the TCP's transmission rate when the underlying link characteristics change suddenly. This can happen, for example, due to vertical hand-offs between GPRS and WLAN wireless technologies. We investigate the Quick-Start algorithm that, in collaboration with the network routers, aims to quickly probe the available bandwidth on a network path, and allow TCP's congestion control algorithms to use that information. By extensive simulations we study the different router algorithms and parameters for Quick-Start, and discuss the challenges Quick-Start faces in the current Internet. We also study the performance of Quick-Start when applied to vertical hand-offs between different wireless link technologies.

Analysis of Heterogeneous IP Traffic in Wireless Environment

Wireless access is expected to play a crucial role in the future of the Internet. The demands of the wireless environment are not always compatible with the assumptions that were made on the era of the wired links. At the same time, new services that take advantage of the advances in many areas of technology are invented. These services include delivery of mass media like television and radio, Internet phone calls, and video conferencing. The network must be able to deliver these services with acceptable performance and quality to the end user. This thesis presents an experimental study to measure the performance of bulk data TCP transfers, streaming audio flows, and HTTP transfers which compete the limited bandwidth of the GPRS/UMTS-like wireless link. The wireless link characteristics are modeled with a wireless network emulator. We analyze how different competing workload types behave with regular TPC and how the active queue management, the Differentiated services (DiffServ), and a combination of TCP enhancements affect the performance and the quality of service. We test on four link types including an error-free link and the links with different Automatic Repeat reQuest (ARQ) persistency. The analysis consists of comparing the resulting performance in different configurations based on defined metrics. We observed that DiffServ and Random Early Detection (RED) with Explicit Congestion Notification (ECN) are useful, and in some conditions necessary, for quality of service and fairness because a long queuing delay and congestion related packet losses cause problems without DiffServ and RED. However, we observed situations, where there is still room for significant improvements if the link-level is aware of the quality of service. Only very error-prone link diminishes the benefits to nil. The combination of TCP enhancements improves performance. These include initial window of four, Control Block Interdependence (CBI) and Forward RTO recovery (F-RTO). The initial window of four helps a later starting TCP flow to start faster but generates congestion under some conditions. CBI prevents slow-start overshoot and balances slow start in the presence of error drops, and F-RTO reduces unnecessary retransmissions successfully.

Helsingin seudun ruuhkamaksut yritysnäkökulmasta

Congestion of traffic is one of the biggest challenges for urban cities in global perspective. Car traffic and traffic jams are causing major problems and the congestion is predicted to worsen in the future. The greenhouse effect has caused a severe threat to the environment globally. On the other hand from the point of view of companies and other economic parties time and money has been lost because of the congestion of traffic. This work studies some possible traffic payment systems for the Helsinki Metropolitan area introducing three optional models and concentrating on the point of view of the economic parties. Central part of this work is formed by a research questionnaire, which was conducted among companies located in the Helsinki area and where more than 1000 responses were gained. The study researches the approaches of the respondents to the area s current traffic system, its development and urban congestion pricing and the answers are analyzed according to the size, industry and location of the companies. The economic aspect is studied by economic theory of industrial location and by emphasizing the meaning of smoothly running traffic for the economic world. Chapter three presents detailed information about traffic congestion, how today s car-centered society has been formed, what concrete things congestion means for economic life and how traffic congestion can be limited. Theoretically it is examined how urban traffic payment systems are working using examples from London and Stockholm where successful traffic payment experiences exist. The literature review analyzes urban development, increasing car traffic and Helsinki Metropolitan area on a structural point of view. The fourth chapter introduces a case study, which concentrates on Helsinki Metropolitan area s different structures, the congestion situation in Helsinki and the introduction of the traffic payment system clarification. Currently the region is experiencing a phase where big changes are happening in the planning of traffic. The traffic systems are being unified to consider the whole region in the future. Also different advices for the increasing traffic congestion problems are needed. Chapter five concentrates on the questionnaire and theme interviews and introduces the research findings. The respondents overall opinion of the traffic payments is quite skeptical. There were some regional differences found and especially taxi, bus and cargo and transit enterprises shared the most negative opinion. Economic parties were worried especially because of the traffic congestion is causing harm for the business travel and the employees traveling to and from work. According to the respondents the best option from the traffic payment models was the ring model where the payment places would be situated inside the Ring Road III. Both the company representatives and other key decision makers see public transportation as a good and powerful tool to decrease traffic congestion. The only question, which remains, is where to find investors willing to invest in public transportation if economic representatives do not believe in pricing the traffic by for example traffic payment systems.