Evolutionary learning methods have been applied to a variety of different problems. In this paper, a new algorithm for active queue management based on an evolutionary learning model is proposed. This novel algorithm generates packet marks for the purpose of improving robustness and responsiveness of congestion control in the Internet routers, while maintaining a reasonable degree of queueing performance such as utilization, delay, and packet drops due to buffer overflow. Simulation results demonstrate the effectiveness of the proposed algorithm and compare the performance of various algorithms.

TCP congestion control is receiving increased attention in recent years due to their usefulness for network stability, robustness use of network buffer and bandwidth resources on an end-to-end per-connection basis. The RED scheme was designed for a network where a single dropped packet is sufficient to signal the presence of congestion to the TCP protocol. This paper applies matrix-analytic approach to analyze both the long-term and the short-term drop behaviors of a queue with RED scheme and uses this model to quantify the benefits brought about by RED. The result shows that the drop probability between RED and Drop-Tail is very close under heavy load conditions. This indicates that RED not only can resolve the synchronization problem but also has the same loss performance with Drop-Tail scheme under the heavy load circumstances. Our findings also show that the rate oscillation behavior of RED is better than Drop-Tail when TCP applies the additive-increase and multiplication-decrease mechanism. As a consequence, it can help reduce the required buffer capacity in the RED router.

Explicit Congestion Notification (ECN) supports the binary congestion information of the network for adjusting the window size. However, this results in the oscillation of the window size and the queue length due to the insufficient congestion information. In this paper, we propose the window-based congestion control mechanism with the modified ECN mechanism. The proposed scheme is based on extracting the network status from the consecutive binary congestion information provided by ECN. From the explicit network information, we estimate the allowable window size to achieve better performance. Through the simulations, the effectiveness of the proposed algorithm is shown as compared with the ECN algorithm.

We derived the design requirements that wireless systems and congestion control algorithms must satisfy to transmit best-effort Internet protocol (IP) packets over wireless systems. We proved that, if these requirements are satisfied, congestion control algorithms are robust against unfairness in the systems and can provide near-maximum throughputs in various environments. From the viewpoint of the design requirements, we investigated the effect of automatic repeat request (ARQ) on the throughputs of best-effort IP connections, and showed why ARQ can improve the throughputs while too large a number of retransmissions degrade them. We also investigated the effect of variance in packet transmission rates and clarified what kind of congestion control algorithm degrades the throughputs.

We developed a distributed control algorithm to solve the problem of a trade-off between transient response and stability. We applied it to a congestion control algorithm for transmitting best-effort packets such as transmission control protocol (TCP) packets over the Internet. A new transmission power control algorithm suitable for transmitting best-effort packets over the wireless Internet was also developed using the distributed control algorithm. We showed that in a steady state, TCP connections can use the bandwidth efficiently over both wired and wireless Internet when the proposed control algorithms are used. The transient response was also evaluated and it was found that the packet transmission rate and the transmission power adjusted by the proposed control algorithms converge to a steady state faster than when adjusted by conventional control algorithms while maintaining the stability of network systems.

In this paper, we propose a new consolidation algorithm called the Selective Backward Resource Management (BRM) cell Feedback (SBF) algorithm. It achieves a fast response and low consolidation noise by selectively forwarding BRM cell from the most congested branch to the source instead of waiting from all branches. Mathematical models are derived to quantitatively characterize the performance, i.e. the response time and ACR of the source, of SBF and previously proposed algorithms. The interoperation of consolidation algorithms in point-to-multipoint available bit rate (ABR) is investigated. We address response time, consolidation noise and the effect of asymmetrical round trip delay (RTD) from branch point to destinations aspects. All combinations of four different consolidation algorithms are interoperated in both local/metropolitan area network (LAN/MAN) and wide area network (WAN) configuration. By a simulation method, we found that the consolidation algorithm used at the uppermost stream branch point, especially in WAN configuration, plays an important role in determining the performance of the network. However, consolidation algorithm used at the lower stream branch point affects the network performance insignificantly. Hence, in order to achieve a good and effective performance of the consolidation algorithms interoperated network, a fast response with low consolidation noise algorithm should be used at the uppermost stream branch point and a simple and easy to implement algorithm should be used at the lower stream branch point.

This paper proposes a credit-based congestion control scheme for multicast communication which employs application-specific processing at intermediate network nodes. The control scheme was designed not only to take advantage of credit-based flow control for unicast communication, but also to achieve flexibility supported by active network technology. The resultant active multicast congestion control scheme is able to meet the different requirements of various multicast applications in terms of reliability and end-to-end latency. The performance of the proposed control scheme was evaluated using both discrete-event simulations and experiments on a prototype active network implementation. The results show that the proposed scheme performs very well in terms of fairness, responsiveness, and scalability. The implementation experiences also confirmed the feasibility of the scheme in practice.

This paper describes IP traffic, especially the control of VoIP traffic, on the carrier-scale, and proposes algorithms for it. It examines a case that has already been introduced in the United States and discusses the trend of standardization for this control. Control techniques that will be introduced into the IP network in the future are considered from the viewpoints of both "quality" that users receive and the "control" that carriers perform.

In the paper, we propose a congestion control scheme for reliable multicast communication which enables TCP fairness and prevents a drop-to-zero problem. The proposed congestion control scheme is rate-based one based on NAKs from receivers and cooperatively works with a flow control scheme. The congestion control scheme consists of two components of a rate-based controller and a selection mechanism of a representative. The rate-based controller runs between the sender and the representative and achieves TCP fairness and fast response to losses at the representative. The selection mechanism of the representative allows the sender to select the representative in a scalable manner, in which the sender makes use of NAKs from receivers to select it. In the paper, we also propose the switchover mechanism of the flow and congestion control schemes which enables the sender to use either of them adaptively based on network situations. When the network is congested, the congestion control scheme works to share network resources fairly with competing TCP flows. Otherwise, the flow control scheme works to adapt the transmission rate to the slowest receiver. We verify the performance of our proposed schemes by using computer simulation.

Layered multicast approach enables IP multicast to adapt to heterogeneous networks. In layered multicast, each layer of a session is sent to separate multicast groups. These layers will be transmitted on the same route, or on different routes. However, traditional congestion control schemes of layered multicast do not consider the case when layers of a session are transmitted on different routes. In this paper, at first we show that in sparse-mode routing protocols like PIM-SM and CBT, layers of a session can be mapped to different Rendezvous Points or cores due to the bootstrap mechanism. It means that layers of a session can be transmitted on different routes. We then show that traditional congestion control schemes of layered multicast do not work properly in sparse-mode routing regions. At last we introduce Rendezvous Point based Layered Multicast (RPLM), a novel congestion control scheme suitable for sparse-mode routing regions, and show that RPLM works efficiently in regions using sparse mode routing protocols. RPLM uses per-RP packet loss rate instead of the overall one to detect congestion on each route, and can react to congestion quickly by dropping the highest layer on the congested route. In addition, RPLM simultaneously drops all the layers those are useless in quality's improvement to prevent bandwidth waste.

This paper proposes two modes of the congestion control scheme to improve its behavior during the start-up period of networks in current TCP over ATM-UBR implementation. The proposed two modes are a single packet loss mode and a multiple packet losses mode. The proposed algorithm is to minimize the number of cell losses in the ATM switch during specially the start-up period. During the start-up period, multiple packet losses often happens because a TCP sender starts with default parameters. It often ends up sending too many packets and too fast, leading to multiple losses is packet burstiness which occurs right after fast recovery ends. We analyze the transition behavior during fast recovery algorithm and estimate the number of new packets sent when multiple packet losses detected. We present a simple simulation model and numerical results to investigate its performance of the proposed algorithms.

In this paper, we survey the fairness issues in the congestion control mechanisms of TCP, which is one of most important service aspects in the current and future Internet. We first summarize the problems from a perspective of the fair service among connections. Several solution methods are next surveyed. Those are modifications of TCP congestion control mechanism and router support for achieving the fair service among TCP connections. We finally investigate the fair share of resources at endhosts.

In order to ease the impact of the packet fragmentation problem and to avoid network congestion in TCP over UBR, packet discard schemes in ATM layer (such as PPD and EPD) have been proposed. These schemes drop packets before they reach their intended destinations if the network is congested and the packets are to be partially discarded. On the other hand, TCP also regulates data flow with its own flow control method. Due to restriction of data flow at the TCP layer, buffer space is not fully used in an ATM switch. In order to make use of more buffer resources, this paper generalizes the PPD and EDP schemes. From this generalization, an optimistic packet discard scheme named the "Probability-Based Delayed Packet Discard" (PDPD) scheme is proposed. Depending on a particular probability, this scheme sets a discard flag to delay actual discard operation. This paper presents the results of several simulated models to find out the potential of improvement of goodput by PDPD. The results of these simulations indicate that PDPD obtains higher goodput than ordinary schemes when the packet size is large and the input load is not light. This author concludes that a PDPD scheme should achieve effective goodput and link utilization while using more buffer resources effectively.

In the future Internet, hierarchically classified Quality of Service (QOS) controls will be effective because various connections requiring different QOS are mixed. However, even in such an environment, among the same class connections, performance protection to harmful impact from the other connections and quality differentiation between connections will be required furthermore. In this paper, from this point of view, we focus on the active connections succession time (age of active connections) as a new dimensional criterion for buffer controls. To be concrete, the packet discarding control of congested router's buffer based on active connections is proposed. Moreover, its performance is evaluated through TCP/IP level simulation from the viewpoint of file transfer time. Conventional Internet can be regarded as the environment where only one class traffic exists (unit class environment). The proposed control scheme can provide powerful differentiation capability to avoid the performance disruption of total connections even in the conventional Internet.

In this article, we first discuss QoS metrics of the data networks, followed by raising the challenging problems for the next-generation Internet with high-performance and high-quality. We then discuss how the WDM technology can be incorporated for resolving those problems. Several research issues for the IP over WDM networks are also identified.

Flow control algorithm in high speed networks is a resource-sharing policy implemented in a distributed manner. This paper introduces a novel concept of backlog balancing and demonstrates its application to network flow control and congestion control by presenting a rate-based flow control algorithm for ATM networks. The aim of flow control is to maximize the network utilization for achieving high throughput with tolerable delay for each virtual circuit (VC). In a resource-sharing environment, this objective may also cause network congestion when a cluster of aggressive VC's are contending for the same resource at a particular node. The basic idea of our algorithm is to adjust the service rate of each node along a VC according to backlog discrepancies between neighboring nodes (i.e., to reduce the backlog discrepancy). The handshaking procedure between any two consecutive nodes is carried out by a link-by-link binary feedback protocol. Each node will update its service rate periodically based on a linear projection model of the flow dynamics. The updated service rate per VC at a node indicates its explicit demand of bandwidth, so a service policy implementing dynamic bandwidth allocation is introduced to enforce such demands. Simulation study has validated the concept and its significance in achieving the goal of flow control and yet preventing network congestion at the same time.

Although in recent years, considerable efforts have been exerted on treating the congestion control problems of ABR services in the ATM networks, the focus has been so far mostly on unicast applications. The inclusion of the emerging multicast services in the design of congestion control schemes is still at its infancy. The generic rate-based closed-loop congestion control scheme proposed by the ATM Forum for ABR services suffers from large delay-bandwidth product. VS/VD behavior is therefore proposed by the Forum as an supplement. In this paper, two VS/VD behavior congestion control schemes for multicast ABR services in the ATM networks are examined : forward explicit congestion notification (FECN) and backward explicit congestion notification (BECN). Their performances are analyzed and compared. We further observe that both VS/VD schemes alleviate the problem of consolidation noise and consolidation delay of the RM cells returning from the downstream nodes. The alleviation of consolidation noise and consolidation delay is a major concern of most present researches. Simulation results are also given to support the validity of our analysis and claims.

This paper presents a design and implementation of a ATM multicast service based on programmable and active network concepts. It aims to address the design and implementation issues of creating new network services--multicast in this case--through a set of corba-based network interfaces, and with a java based user codes injection mechanism for supporting customization of network services. We demonstrate the feasibility of our prototype through the implementation of a wavelet video multicast application with active filters implanted at intermediate nodes for supporting heterogeneous receivers and the implementation of a congestion control scheme. The performance of the prototype over an ATM test-bed is measured and evaluated.

Recently, it is required to transfer continuous media over networks without QoS guarantee. In these networks, network congestion will cause transmission delay variance which degrades the quality of continuous media itself. This paper proposes a new protocol using a congestion control with two level rate control in the data transfer level and the coding level. It introduces a TCP-like congestion control mechanism to the rate control of data transfer level, which can detect the QoS change quickly, and adjust the coding rate of continuous media with time interval long enough for its quality. The performance evaluation through software simulation with multiplexing continuous media traffics and TCP traffics shows that the proposed protocol works effectively in the case of network congestion.

Congestion Control Scheme of TCP/IP protocol suite is established by Transmission Control Protocol (TCP). Using the self-clocking scheme, TCP is able to maintain a quick optimum connection status for the network path, unless it is given an excessive load to carry to the network. However, in wide area networks, there are some obstructive factors for the self-clocking scheme of TCP. In this paper, we describe the obstructive factors for the self-clocking scheme. We propose a new congestion control scheme using a packet pair scheme and a traffic-shaping scheme. In combining these schemes with TCP, new TCP options and a modification for TCP congestion control algorithms are added. Using our scheme, TCP is able to maintain smooth self-clocking. We implemented this scheme on a network simulator for evaluation. Compared with normal TCP, this scheme was demonstrated to be over 20% more efficient in symmetric communication and over 40% more efficient in asymmetric communication.