The dramatically increased bandwidths and processing capabilities of future high-speed networks make possible many distributed real-time applications, such as sensor-based applications and multimedia services. Since these applications will have traffic characteristics and performance requirements that differ dramatically from those of current data-oriented applications, new communication network architectures and protocols will be required. In this paper we discuss the performance requirements and traffic characteristics of various real-time applications, survey recent developments in the areas of network architecture and protocols for supporting real-time services, and develop frameworks in which these, and future, research efforts can be considered.

IP based groupware applications, such as net-based learning environments, rely on robust audio transport for efficient communication between users. This paper therefore gives an overview and an initial evaluation of how to achieve robust transport of real-time audio streams over Internet connections without service guarantees. Due to hardware jitter and network congestion these connections face loss, packet delay and delay variation. Audio streams are especially sensitive due their realtime characteristics, where the end result is degradation of the perceived quality. Packet loss can be repaired using receiveronly, sender-initiated or receiver-initiated techniques. Depending on the actual network condition, an optimal technique can be selected using adaptive behavior together with loss-recovery techniques in the applications. Studies have shown that loss rates up to 20% can be effectively repaired using fairly simple techniques. The paper gives initial results from subjectively evaluat...

By conventional wisdom dating back to the genesis of the Internet, TCP is not suitable for realtime applications because it favors reliability over timeliness. The result has been a proliferation of application-specific and candidate real-time transport protocols, none of which has achieved the maturity and acceptance of TCP. To our knowledge, there has not been much prior study to ask the basic engineering question: What changes in TCP and its use would be required to allow it to be used for real-time network applications of interest? In this paper, we investigate exactly this question. Our testbed-based and public Internet experiment results show that the combination of a relatively modest extension to TCP, what we call real-time mode (RTM), and several application techniques allow real-time applications to operate well over TCP, making all the benefits of TCP, including error recovery, available to the real-time application developers. The TCP-RTM approach also leads to real-time applications that are responsive to network congestion, sharing the network resources fairly with other TCP applications. 1

We consider policies for deciding which cells will be lost or dropped when losses occur at a finite buffer ATM node. The performance criteria of interest are the delay of transmitted (nonlost) cells, the jitter (or variability in the delay of transmitted cells), and the burstiness of lost cells. We analyze the performance tradeoffs for various cell dropping policies. We show the usual the "rear dropping" in which cells that arrive to a full buffer are lost stochastically maximizes delay, while "front dropping", in which cells at the front of the buffer are lost, stochastically minimizes delay. On the other hand, rear dropping stochastically minimizes the jitter. We also propose policies that have both stochastically smaller delay and less lost cell burstiness in a stochastic majorization sense than the rear dropping policy.

We consider policies for deciding which cells will be lost or dropped when losses occur at a finite buffer ATM node. The performance criteria of interest are the delay of transmitted (non-lost) cells, the jitter (or variability in the delay of transmitted cells), and the burstiness of lost cells. We analyze the performance tradeoffs for various cell dropping policies. We show the usual the "rear dropping" in which cells that arrive to a full buffer are lost stochastically maximizes delay, while "front dropping", in which cells at the front of the buffer are lost, stochastically minimizes delay. On the other hand, rear dropping stochastically minimizes the jitter. We also propose policies that have both stochastically smaller delay and less lost cell burstiness in a stochastic majorization sense than the rear dropping policy.

i telekommunikation fredagen den 5 juni 2009 vid KTH.