A fuzzy threshold controller for priority traffic in ATM networksis presented The controller is based on fuzzy theory and PED control principle. The new approach can perform much better than the fixed threshold mechanism through adjusting the threshold periodically and adaptivelly and adaptively according to the arriving traffic load and charac teristics.

This paper fist revistits the previously proposed NSAD (New Self-Adapt DCF) mechanism. Some modifications are presented to further enhance the performance of NSAD in the error-prone environment. Then a new NAC mechanism is proposed that can realize bandwidth guarantee by assigning different. self-adapt parameters to users at different priority levels. the bandwidth guarantee property of this new mechanism is analyzed and the highpriority users are found to have bandwidth guaranteed even in heavy contention condition, which is proved true not only by theoretical analysis but also by simulatin results. At the same time the new scheme keeps the self-adapt character of NSAD, so the overall system utilization is kept very high in heavy contention condition compared with the previously studied DCF-based QoS mechanisms.

This paper presents a more complete analytical model of multiple TCP connections established between a busy server and multiple receivers. In such case, the server may become the bottleneck of the whole system, and TCP behaviors are different from the model presented before. Based on the analysis of working flows in the system and a M/G/1 queueing model, the RTT and long-term TCP throughput formulae are derived in terms of number of TCP connections, packet loss rate, and end-to-end delay. And the effect of maximum window size is also investigated. Finally, the simulation results confirm the effectiveness of analytical model.

本文介绍了Internet端到端测量技术的最新进展，包括性能拓扑推测、时延测量、丢包率测量。最后讨论了未来Internet端到端测量的研究趋势和应用前景。

Distributed Coordination Function (DCF) is the basis of the IEEE 802.11 WLAN MAC protocols. This paper proposes a scheme named DCF+, which can be regarded as an option to DCF, to enhance the performance of reliable transport protocol over WLAN. To analyze the performance of DCF and DCF+, this paperalso introduces an analytical model to compute the saturated throughput performance of WLAN. Compared with other models, this model is more accurate, which is verified by elaborate simulations. Moreover, DCF+ is shown to be able to improve the performance of TCP over WLAN by both modeling and simulations.

Queuing/scheduling algorithm is one of the most important mechanisms to provide guaranteed Quality of Service(QoS) in high speed packet-switched networks. By computing the system virtual time and per packet/connection virtual start/finish time, a number of Packet Fair Queueing(PFQ) algorithms are designed to simulate GPS (Generalized Processor Sharing)[1]. The difference in computation complexity is due to the variable ways to compute system time. Many algorithms, including WFQ/WF2Q, SCFQ, SPFQ, WF2Q+, VC, etc., have been proposed to use different system Virtual Time(also known as system potential) function. This paper proves that all the Latency-Rate (LR) servers only need to calculate their system virtual times once per packet service time, no matter how many packet arrivals occur in this interval. Thus, it is a general scheme that benefits all the well-known LR PFQ algorithms.

Tis paper proposes an algorithm named WLFVC, which is fully distributed wireless packet scheduling algorithm based on LFVC (LeapForward Virtual Clock), to provide streams with fairness, especially in the short time scale. This algorithm is able to allocate bandwidth to streams proportional to their weights. Thus, betterdelay and jitter performance can be achieved. since collisions and wireless link error degrade the performance of MAC protocols in the WLAN environment, WLFVC also takes them into consideration. Finally, this algortthm works in a fully distributedway and it can coexist with other stations that only support DCF (Distributed Coordination Function).