Receiver-driven layered multicast has been designed to support scalable multimedia applications and very recently used to support bulk file transfers. Layered multicast congestion control assumes that all the receivers behind a bottleneck are equally able to subscribe to a particular number of cumulative layers. This paper explores the requirement placed on heterogeneous receivers behind a bott...

Some of the promising End-To-End solutions to improve TCP performance over networks suffering from wireless transmission errors are error discriminators. An error discriminator is any technique that tries to differentiate between error types and deal with each type differently. By distinguishing between error types, TCP can avoid unnecessary cuts of congestion window in the case of wireless err...

Transmission Control Protocol (TCP) carries most of the traffic on the Internet these days. There are several implementations of TCP, and the most important difference among them is their mechanism for controlling congestion. One of the methods for determining type of a TCP is active probing. Active probing considers a TCP implementation as a black box, sends different streams of data to the ap...

In this paper we describe a study of an end-to-end architecture in a packet network where congestion is signalled to all contributing users, who then react accordingly. We assume a generic form of packet marking, where congested nodes set a single bit in the packets flowing through them using ECN (Explicit Congestion Notification). We briefly describe how allowing greater freedom to the end-sys...

Wireless Mesh Network (WMN), is a upcoming promising technology, that is continuing to boom because of its various advantages like low deployment cost, scalability, efficiency, characteristics of each node being to function as a router, etc.. But there are many challenges that are not addressed, which includes congestion prevention and avoidance. Congestion in any network will or may result in ...

In the current Internet, TCP’s congestion control is performed end-to-end possibly assisted by simple and limited congestion feedback mechanisms in routers, e.g., Explicit Congestion Notification (ECN) or Random Early Detection (RED). In order to control network traffic load more accurately by reducing or even preventing congestion in (parts of) the network and improving the overall performance...

Network throughput increase is usually associated with replacement of communication links and appropriate network devices. This approach is usually intrusive, expensive and nevertheless time consuming. On the other hand it is necessary to bear in mind that effective and less intrusive increase of network throughput can be achieved via the improvement of existing protocol stack especially networ...

In this paper, we examine an emerging combination of challenges for TCP: increasingly bursty background traffic that is not subject to flow and congestion control, higher bandwidth networks and small buffers at network routers. As a result, TCP experiences short-term bursty packet losses that may not reflect long-term congestion. In this work, we propose a balanced approach that uses TCP conges...

Residual bit-errors in wireless environments are well known to cause difficulties for congestion controlled protocols like TCP. Newer protocols, such as the non-congestion controlled UDP-Lite and the congestion controlled DCCP use partial checksums to adapt and make better use of a channel with biterrors. For TCP there exists many possible approaches to handle the difficulties, and in this stud...