Although the Internet was designed to handle non-real time data traffic, it is being used increasingly to carry voice and video. One important class of contributors to this growth are Internet telephones. Critical to more widespread use of Internet telephony is smooth interoperability with the existing telephone network. This interoperability comes through the use of Internet Telephony Gateway’s (ITG’s) which perform protocol translation between an IP network and the Public Switched Telephone Network (PSTN). In order for an IP host to call a user on the PSTN, the IP host must know the IP address of an appropriate gateway. We consider here the problem of finding these gateways. An analysis of a number of protocol architectures is presented, including hierarchical databases, multicast advertisement, routing protocols, and centralized databases. We propose a new protocol architecture, called Brokered Multicast Advertisements (BMA) which serves as a lightweight, scalable mechanism for locating ITG’s. The BMA architecure is general, and can be applied to location of any service across a wide area network. 1

This paper examines current header compression techniques, CRTP & ROCCO, for multimedia services over IP and examines their performance in mobile channel environments

This paper was written for an Independent Study course. Princy Mehta Overview of Voice over IP Professor Udani February 2001 1 Table of Contents ACRONYMS AND DEFINITIONS............................................................................................................... 3 INTRODUCTION ........................................................................................................................................... 5 IMPLEMENTATION OF VOICE OVER IP............................................................................................... 6 OVERVIEW OF TCP/IP ................................................................................................................................... 6 PACKETIZATION............................................................................................................................................. 7 COMPONENTS OF VO

Voice-over-IP (VoIP) uses packetized transmission of speech over the Internet (IP network). However, at the receiving end, packets are missing due to network delay, network congestion (jitter) and network errors. This packet loss degrades the quality of speech at the receiving end of a voice transmission system in an IP network. Since the voice transmission is a real-time process, the receiver cannot request for retransmission of the missing packets. Concealment algorithms, either transmitter or receiver based, are used to replace these lost packets. The packet loss concealment (PLC) techniques described in the standards ANSI TI.521 (Annex B) and ITU-T Rec. G.711 (Appendix I), have good performance, but these algorithms do not use subsequent packets for reconstruction. Furthermore, there are discontinuities between the reconstructed and the subsequent packets, especially at the transitions from voiced to unvoiced and phoneme to phoneme. The goal of this work is to develop an improved PLC algorithm, using the subsequent packet information when available. For this, we use the Time-Scale Modification (TSM) technique based on Waveform Similarity Over-Lap Add (WSOLA) to reconstruct

Abstract—In this paper we analyze packet traces of widely used voice codecs and present analytical source models which describe their output by stochastic processes. Both the G.711 and the G.729.1 codec yield periodic packet streams with a fixed packet size, the G.723.1 as well as the iLBC codec use silence detection leading to an on/off process, and the GSM AMR and the iSAC codec produce periodic packet streams with variable packet sizes. We apply all codecs to a large set of typical speech samples and analyze the output of the codecs statistically. Based on these evaluations we provide quantitative models using standard and modified on/off processes as well as memory Markov chains. Our models are simple and easy to use. They are in good accordance with the original traces as they capture not only the complementary cumulative distribution function (CCDF) of the on/off phase durations and the packet sizes, but also the autocorrelation function (ACF) of consecutive packet sizes as well as the queuing properties of the original traces. In contrast, voice traffic models used in most of today’s simulations or analytical studies fail to reproduce the ACF and the queueing properties of original traces. This possibly leads to underestimation of performance measures like the waiting time or loss probabilities. The models proposed in this paper do not suffer from this shortcoming and present an attractive alternative for use in future performance studies. Index Terms—traffic models, voice codecs, autocorrelation, queuing behavior

A solution to the problem of speech recognition with signals distorted by low-bit rate coders is presented in this paper. A model for the coding-decoding distortion, a HMM compensation method to include this model, and an EM-based adaptation algorithm to estimate this distortion are proposed here. Medium vocabulary continuous-speech speaker-independent recognition experiments with 8 kbps G.729(CS-CELP), 13 kbps RPE-LTP (GSM), 5.3 kbps G723.1, 4.8 kbps FS-1016 and 32 kbps G.726(ADPCM) coders show that the approach described in this paper is able to dramatically reduce the effect of the coding distortion and, in some cases, gives a word accuracy higher than the baseline system with uncoded speech. Finally, the EM estimation algorithm requires only one adapting utterance and the approach described is certainly The evolution and popularity of cellular and TCP/IP networks has created the problem of improving the recognition accuracy for speech distorted by low-bit rate coders. The distortion of coding schemes in speech recognizers is difficult to model and is an open problem that cannot be solved by applying conventional noise cancelling techniques [1] such as spectral subtraction [2], cepstral mean subtraction [3] and RASTA

A System on Chip (SoC) is presented, which is wellsuited for both conventional telephony via the Public Switched Telephone Network (PSTN) and Voice over Internet Protocol (VoIP).

Previous work in wireless speech recognition has focused on two methods, namely, quantizing recognition features (e.g.

Abstract—We present simple approximation formulae for the distribution of the packet waiting time of multiplexed periodic traffic. The multiplexed streams may have different periods and packet sizes. We show by extensive simulations the accuracy of the proposed methods. They are simpler than other existing formulae which make them attractive for engineers, applicable in practice, and easy to implement in switching devices. Packetized speech traffic has a periodic flow structure. Many compression techniques preserve it during talk phases but suppress the generation of packets during silence phases. When such on/off streams are multiplexed, advantage can be taken of their reduced flow rates by overbooking the link bandwidth. We adapt the proposed formulae to cope with on/off traffic and overbooking and validate them by extensive simulations. They can be applied for admission control in networks carrying different types of real-time traffic. Index Terms—waiting time distribution; on/off streams; multiplexing; overbooking. I.