K. Sleurs

Fast link adaptation for TFRC after a handover

TCP-friendly rate control(TFRC) is an equation-based congestion control mechanism that competes fairly with TCP but has a much lower throughput variation, which makes it a better choice for streaming over the Internet. It is known that in a mobile network environment, after a handover TFRC can overshoot or under utilize the new link if the conditions there differ from those of the old link. There are different factors that affect the TFRC performance during a handover. In this paper we focus on the impact of the change in the round trip time, an aspect that is largely overlooked in the literature compared with bandwidth disparity. We find that even if the same bandwidth is assured on the new link, changes in the link latency will still cause TFRC performance degradation after the handover. A fast link adaptation mechanism is thus proposed to address the problem. Simulations show that this mechanism helps TFRC to adjust quickly to the new link without consecutive packet loss or long-time bandwidth under utilization.

A fast adaptation mechanism for TCP vertical handover

In this paper we propose a vertical handover mechanism for TCP to deal with the delay and bandwidth change during the handover. The proposed mechanism relies on the interaction between the sender and the receiver during the handover and requires no a prior information of the new path. The adaptation is triggered by a handover notification at the receiver and finishes in about two round-trip time. Furthermore, it solves the problem of packet reordering and spurious retransmission timeout as well which are also common during vertical handovers. Simulation results show that our mechanism improves TCP performance in various vertical handover scenarios.

Training Sequence Versus Cyclic Prefix for CPM with Frequency Domain Equalization

Frequency domain equalization (FDE) of continuous phase modulations (CPM) has been thoroughly investigated lately. To enable this low-complexity FDE, all known techniques use a cyclic prefix (CP). However, using a training sequence (TS) of known symbols instead of a CP offers some advantages: the additional known symbols can be used to improve synchronization and channel estimation, with the same performance of a CP. Nevertheless, using a TS for CPM-FDE is not trivial because the memory in a CPM waveform has to be taken into account to guarantee cyclicity and phase continuity after insertion of the TS into a block of input symbols. In this paper, we therefore propose a technique for constructing a TS for CPMFDE. Simulation results in a 60 GHz environment show that the proposed technique satisfies all requirements. The 60 GHz case is chosen because CPM with FDE has recently been proposed for communications at 60 GHz and the latest IEEE and ECMA standards for these frequencies mandate the use of a TS rather than a CP.

A Qualitative Description of the Effect of Single Queues on Bin Counts

A large number of research articles are dedicated to queuing theory and queuing systems. Most of these articles employ a continuous representation of network traffic, in the form of timestamps or interarrival times. In this, there is a contradiction with more recent traffic models capable of capturing the multi-fractal nature of network traffic e.g. the Conservative Cascade model. These models often represent packet streams in a discrete way by calculating the bin count vector. Directly describing the effect of queuing systems on the variance-time behavior of this discrete representation of traffic is relatively unexplored terrain. This paper presents and analyzes some qualitative results on the altering of a bin count vector when passing it through a basic queuing system. The scenario is also extended with the influence of correlated background traffic.

How Different Queuing Systems Affect the Discrete Representation of a Packet Stream

A large number of research articles are devoted to queuing theory and queuing systems. Most of these articles employ a continuous representation of network traffic, in the form of timestamps or interarrival times. In this, there is a contradiction with more recent traffic models capable of capturing the multi-fractal nature of network traffic e.g. the conservative cascade model. These models often represent packet streams in a discrete way by calculating the bin count vector. Directly describing the effect of queuing systems on the variance-time behavior of this discrete representation of traffic is relatively unexplored terrain. This paper presents and analyzes some qualitative results on the altering of a bin count vector when passing it through a queuing system. After the detailed analysis of a basic fixed service time queue, some considerations are made on real networking components. This leads to extending the basic queue model with variable service times, based on the packet size distribution. Finally, the influence of the shape of this distribution on the queuing effects is studied.

Improving RED Performance during Handovers in Wireless IP Networks

The random early detection (RED) is a widely used active queue management (AQM) algorithm for congestion avoidance. It monitors the average queue length to detect incipient congestion and notifies the connections of congestion to adjust their sending rate. During a handover, all the active connections of the mobile node will be diverted to the new network and increase the total traffic there immediately. Because RED uses an exponential weighted moving average (EWMA) to calculate the queue length, it will be too slow to track this rapid change and fail to react correctly. In this paper we propose a dual RED algorithm to accommodate the coming handed-over connections quickly while keeping the link utilization high. The improvement is verified with simulations.

Evaluation of network traffic workload scaling techniques

Performance testing on a network that is loaded by realistic background traffic, is an important issue in the design of network applications. To obtain this background traffic, a packet stream can be captured on a real network. For testing the application under varying network loads, straightforward techniques can be applied to alter the originally captured traffic trace. An example thereof is faster replay of the packet stream. The drawback of these simple techniques is that the packet stream will be altered in many ways, and thus a simple multiplication of the load that is inflicted by the packet stream on a network cannot be guaranteed. This paper presents and analyzes some frequently used load scaling methods. Next to the computationally more demanding use of traffic models, a novel technique is introduced based on the rotation of the bin vector, that combines better accuracy with computational simplicity.

Bridging the Gap between Mathematical Traffic Models and Operational Parameters

Over the past decade and more, several accurate but complex traffic models have been developed by just as many researchers. Among the most accurate models are those based on mathematical principles that are able to model the multifractal nature of network traffic. Unfortunately these models are hardly even usable by network engineers because they lack a connection to operational parameters that can easily be estimated based on the knowledge of the network. In this paper, we try to bridge the gap between high-level parameters describing the concerning network and the inputs the modern traffic models need to generate artificial traffic. We first build a model of the behavior of TCP over the duration of a flow, and we then approximate this behavior based on simple parameters such as for example the packet transmission time and the RTT.

Design considerations of a modular testbed for mobile IP networks

In this paper we present a research testbed being developed at K.U.Leuven, Belgium, aiming at providing a flexible and convenient testing infrastructure for researchers working on mobility issues over IP networks. The testbed is built on general-purpose PCs, which makes it suitable for most laboratories. Its user-level modularity makes it easy for both developers and users to configure and extend even for unusual mobility scenarios and new mobility techniques. Movement is realized in a way that makes mobility experiment easy to be carried out and automated