András Veres

The chaotic nature of TCP congestion control

In this paper we demonstrate how TCP congestion control can show chaotic behavior. We demonstrate the major features of chaotic systems in TCP/IP networks with examples. These features include unpredictability, extreme sensitivity to initial conditions and odd periodicity. Previous work has shown the fractal nature of aggregate TCP/IP traffic and one explanation to this phenomenon was that traffic can be approximated by a large number of ON/OFF sources where the random ON and/or OFF periods are of length described by a heavy-tailed distribution. In this paper we show that this argument is not necessary to explain self-similarity, neither is randomness is required. Rather, TCP itself as a deterministic process creates chaos, which generates self-similarity. This property is inherent in todays TCP/IP networks and it is independent of higher layer applications or protocols. The two causes, heavy-tailed ON/OFF and chaotic TCP together contribute to the phenomenon, called the fractal nature of Internet traffic.

Supporting service differentiation for real-time and best-effort traffic in stateless wireless ad hoc networks (SWAN)

We propose SWAN, a stateless network model which uses distributed control algorithms to deliver service differentiation in mobile wireless ad hoc networks in a simple, scalable and robust manner. The proposed architecture is designed to handle both real-time UDP traffic, and best effort UDP and TCP traffic without the need for the introduction and management of per-flow state information in the network. SWAN supports per-hop and end-to-end control algorithms that primarily rely on the efficient operation of TC/IP protocols. In particular, SWAN uses local rate control for best-effort traffic, and sender-based admission control for real-time UDP traffic. Explicit congestion notification (ECN) is used to dynamically regulate admitted real-time sessions in the face of network dynamics brought on by mobility or traffic overload conditions. SWAN does not require the support of a QoS-capable MAC to deliver service differentiation. Rather, real-time services are built using existing best effort wireless MAC technology. Simulation, analysis, and results from an experimental wireless testbed show that real-time applications experience low and stable delays under various multihop, traffic, and mobility conditions.

A large-scale, passive analysis of end-to-end TCP performance over GPRS

In this paper a passive methodology for TCP performance evaluation over general packet radio service (GPRS) networks is presented that relies on traffic monitoring at the GPRS ingress/egress router interface (Gi). Based on the IP and TCP headers of the packets we estimate the end-to-end performance of TCP connections such as connection setup behavior and data transfer goodput. In order to identify the effects behind the measured performance the introduced algorithms estimate round trip delays, packet loss ratios, available channel rates, throughput and cany out bottleneck analysis. Large-scale GPRS measurements in seven countries are presented to analyze TCP performance and demonstrate the applicability of the method. The effects of different TCP parameters such as maximum segment size, selective acknowledgements, timestamp usage and receiver window size are also quantified. GPRS measurement results are compared to a wireline dial-up network to identify the effects specific to the wireless environment

A passive method for estimating end-to-end TCP packet loss

This paper presents a passive end-to-end loss monitoring method, which relies on traffic monitoring at a core or ingress router interface. The monitoring node captures the packets of TCP connections generated by end-hosts. Based on the seen sequence number pattern the loss ratios are estimated for the two segments of end-to-end path divided by the monitor. This feature is especially useful if the monitoring node is placed at the border of an autonomous system, e.g., at the egress point of an ISP. When applied to mobile Internet access, packet losses on the radio interface can be distinguished from losses in the core network or in the public Internet. The packet loss estimation algorithm is rigorously validated using simulations and testbed measurements. Additionally, measurement results and an application example in a dial-up ISP are presented.

A family of measurement-based admission control algorithms

In this paper we identify a set of the requirements for an efficient admission control algorithm and propose new algorithms. These methods need only aggregate traffic measurements, work with simple FIFO scheduling and take only minimal assumptions on the pattern of the traffic. The requirements define a family of measurement based admission control algorithms of which three key members are discussed. We give effective bandwidth formulae for buffered and bufferless systems with token bucket and peak rate limited sources. It is also shown how the utilization can be improved by measuring the variance of the traffic rate while avoiding the limitations of MBAC methods based on the Central-limit theorem.

TCP's role in the propagation of self-similarity in the Internet

This paper analyzes how TCP congestion control can propagate self-similarity between distant areas of the Internet. This property of TCP is due to its congestion control algorithm, which adapts to self-similar fluctuations on several timescales. The mechanisms and limitations of this propagation are investigated. It is demonstrated that if a TCP connection shares a bottleneck link with a self-similar background traffic flow, it propagates the correlation structure of the background traffic flow asymptotically, above a characteristic timescale. The cut-off timescale depends on the end-to-end path properties, e.g. round-trip time and average window size, and the receiver window size in case of high-speed connections. It is also shown that even short TCP connections can propagate long-range correlations effectively. In case when TCP encounters several bottleneck hops, the end-user perceived end-to-end traffic is also long-range dependent and it is characterized by the largest Hurst exponent. Through simple examples, it is shown that self-similarity of one TCP stream can be passed on to other TCP streams that it is multiplexed with. The limitations of propagation depend on network conditions. Propagation complements the widespread scaling phenomena reported in the literature. Our arguments are supported with a combination of analytic techniques, simulations and statistical analyzes of a number of wide area Internet traffic measurements.

Ping-Pong Reduction Using Sub Cell Movement Detection

In this paper we demonstrate an effective way to classify repeated handovers (ping-pongs) in mobile broadband networks. The paper also presents a method that can significantly reduce unwanted ping-pongs in the network. The method combines a sub cell movement detection method and ping-pong detection to decide when it is most effective to apply handover threshold tuning (pinning) without increasing the risk of late or failed handovers. The algorithm was evaluated based on live network measurements.

Measuring high-speed TCP performance during mobile handovers

We analyse throughput adaptation of high-speed TCP protocols during handovers in an emulation testbed. We use emulation characteristics reflecting current mobile networks and future network specifications as set by the relevant standards. We found that the performance of TCP protocols is satisfactory, except as follows. Traditional loss-based TCP-s like BIC, highspeed and scalable adapt slowly to sudden link capacity increases caused by handovers. Fast TCP adapts much faster, however depending on parameter settings it may under-utilise after a capacity decrease coupled with RTT increase. We also calculate the probability of buffer overflows at handover as the function of the buffer size and transmission parameters at the bottleneck link. This probability turns out to be significant for reasonable scenarios, and we suggest a dynamic buffer re-allocation method to eliminate such buffer overruns in mobile networks.

Movement detection for location based network management

This paper presents a sub-cell movement detection method for mobile networks based on standardized terminal reports. Movement detection is complementary to the well-known problem of location detection, but, compared to localization, movement detection puts less stringent constraints on data sources, e.g., network synchronization is not required. Statistics on moving terminals could potentially enhance a set of advanced network management applications. The algorithm is demonstrated on real network data, we present an analysis of mobility patterns for different types of mobile broadband terminals.

Towards Understanding the Evolution of Wars in Virtual and Real Worlds

Emerging Massively Multiplayer Online Real Time Strategy games require complex game server architecture to make the transmission of the state information of a huge number of units generated by a lot of players feasible. This architecture design is supported by network traffic simulations based on the accurate characterization of player behavior. However, these characteristics of player behavior in RTS games have not been investigated yet, thus, this is the main motivation of this paper. In particular, we introduce a method that can identify the war periods in real-time strategy game sessions based on non-intrusive measurements thus, it is possible to analyze a vast number of game plays. The paper also presents some early results comparing the real and gaming worlds.