Estágio realizado na Novabase e orientado pelo Eng.º Pedro Faúlha / Tese de mestrado integrado. Engenharia Informátca e Computação. Faculdade de Engenharia. Universidade do Porto. 2008

Ces travaux portent sur la sécurité de la téléphonie déployée dans les réseaux Internet. Ce service est sans aucun doute, après le Web et la messagerie, l'application qui imposera l'infrastructure IP (Internet Protocol) comme le standard de transport de tout type d'information ou de média. Cette migration de la téléphonie classique vers le tout IP semble être incontournable mais elle pose des problèmes en matière de sécurité. Si des attaques existaient déjà avec la téléphonie classique, l'usage d'un réseau IP les rend plus facilement réalisables. Notre analyse souligne les limites des solutions usuelles, principalement au travers des problèmes d'interopérabilité. De plus eu égard à l'hétérogénéité des infrastructures de ToIP, la protection de bout-en-bout des appels n'est pour le moment pas considérée, sauf par les services étatiques. Dans un premier temps, nous avons cherché les possibilités de renforcer la sécurité de SIP (Session Initiation Protocol) de l'IETF, protocole actuellement massivement adopté dans les infrastructures de téléphonie sur IP. Nous avons proposé des solutions innovantes et validées pour consolider les mécanismes existants de manière complètement transparente pour les infrastructures. Nous avons choisi de nous focaliser sur l'authentification, car c'est le premier mécanisme rencontré par les usagers ou les systèmes. Les solutions présentées ci-après proposent de nouvelles propriétés de sécurité en définissant une sémantique pour des champs dit " opaques ". Ces contributions consolident la sécurité entre l'usager et son serveur. Dans un second temps, nous nous sommes intéressés aux solutions permettant une sécurité de bout-en-bout des appels. L'analyse des solutions applicatives comme " Future Narrow Band Digital Terminal " et " Secure Voice over IP Simple Protocol " nous a permis de formaliser les spécifications d'une architecture permettant la protection des conversations quelque soient les spécificités et l'hétérogénéité des réseaux de ToIP. Cette approche utilise le flux d'informations voix pour mette en oeuvre une signalisation de sécurité, ce qui rend cette solution complètement compatible avec les infrastructures existantes. Par ailleurs notre étude atteste de l'intérêt de mettre en place des entités de confiance dédiées à la sécurité des appels. Enfin la conclusion reprend et positionne les différentes contributions relatives à ces travaux dans le contexte de la téléphonie sur IP. Notre volonté d'être interopérable avec les infrastructures sous-jacentes voire indépendantes peut être considérée comme un service à valeur ajoutée.

<p>The trends in todays tele- and datacommunication market point toward using IP for all sorts of service delivery ranging from voice calls to TV. The next natural step in this evolution is to provide the same set of services to the end users independent of the access technology and device used. The IP Multimedia Subsystem (IMS) is an IP based telecommunications platform which targets this and lets the operators develop new services once which can then be used on many different devices.</p><p>This thesis examines the integration of IPTV and IMS. Can IMS be used to deliver TV services and can the IPTV set-top-boxes of today be used as clients in IMS? Since this is a new and previously unexamined area an explorative approach is taken. The aim is to identify how such an integration could be performed and the possible problems which have to be solved. To assist in this exploration a TV-push service based on IMS technology is constructed.</p><p>Based on the experiences from this service a general architecture for IPTV in IMS is suggested.</p><p>A number of problems crucial to solve for a successful integration are identified and possible solutions to these are discussed.</p>

<p>In the thesis the advantages with Text over IP (ToIP) is explained and it is motivated why it is a good idea to integrate this in Asterisk. It also presents an implementation of a ToIP extension in Asterisk.</p><p>ToIP means communicating over a network based on Internet protocols with real-time text. Real-time text means a character is sent to the receiving terminal as soon the sender has typed it or with a small delay.</p><p>In the thesis IM and ToIP is compared in a survey. The result point at IM is not better than ToIP even though it is much more commonly used. VoIP can not replace ToIP either because there are occasions when ToIP is better for instance if the person using it is deaf or if a person want to make a private conversation in a noisy room.</p><p>Asterisk is an IP-PBX. PBX stands for Private Branch Exchange which means a private telephone system which is part of a larger network system that exchange information.</p><p>An IP-PBX is a PBX based on the Internet. Asterisk and many other IP-PBX can also exchange calls between the PSTN ant the Internet. By including ToIP in Asterisk it will be possible to exchange ToIP calls.</p><p>The implementation described is not only including ToIP in Asterisk but also a translation function between the text format called t140 and another text format called t140 with redundancy.</p><p>The idea is to extend the translation function in the future to more text formats.</p>

<p>Gateways handle many functions in todays telecommunication networks and as the move towards IP-based telecommunication networks continue, their importance is growing. Many vendors offer a tiered architecture where logic is separated from the gateways for easy extensibility. Currently gateway control and gateway communication is handled using the H.248 protocol. As more and more equipment starts moving over to SIP based communication there has been a degree of interest in homogenizing the system and possibly replacing the H.248 protocol with a SIP based protocol.</p><p>In this paper we examine how the communication between a gateway controller and gateway may look if implemented in SIP. We also examine the performance characteristics of the proposed protocol from both an execution time, communication size and memory consumption perspective. Implementation and tests will be performed using a language, developed by Ericsson specifically for the telecommunication sector, called Erlang. The protocol designed herein is not intended for production use and is only examined for viability. It is NOT the official stance of Ericsson that H.248 will be replaced with the protocol or any like it.</p>

<p>H.323 och Session Initiation Protocol är två olika protokoll som kan användas t ex för att koppla upp röstsamtal eller videosamtal via Internet. Det är ofta önskvärt i en uppkoppling mellan två personer att personerna kan autentisera sig för varandra. Denna autentisering är avsedd att garantera identiteten på deltagarna i kommunikationen. Den här undersökningen jämför protokollens struktur vid autentiseringen och visar skillnader i säkerhetssynpunkt. Autentisering finns i 3 skikt i de båda protokollen. I applikationsskiktet skiljer sig protokollen åt då SIP använder sig av lösenord för autentisering medan H.323 både kan använda lösenord och en PKI-baserad lösning med utbyte av nyckelcertifikat. I transportskiktet och nätverksskiktet kan båda protokollen använda TLS och IPSec för autentisering och därmed är det ingen större skillnad på protokollen i dessa skikt.</p>

Voice over IP (VoIP) is a real time application that allows transmitting voice through the Internet network. Recently there has been amazing progress in this field, mainly due to the development of voice codecs that react appropriately under conditions of packet loss, and the improvement of intelligent jitter buffers that perform better under conditions of variable inter packet delay. In addition, there are other factors that indirectly benefited VoIP. Today, computer networks are faster due to the advances in hardware and breakthrough algorithms. As a result, the quality of VoIP calls has improved considerably. However, the quality of VoIP calls under extreme conditions of packet loss still remains a major problem that needs to be addressed for the next generation of VoIP services. This thesis concentrates in making an analysis of the effects that network impairments, such as: delay, jitter, and packet loss have in the quality of VoIP calls and approaches to solve this problem. Finally, we analyze the impact of introducing forward error correction (FEC) Piggybacking and Reed Solomon codes for VoIP. To measure the mean opinion score of VoIP calls we develop an application based on the E-Model, and utilize perceptual evaluation of speech quality (PESQ).

The trends in todays tele- and datacommunication market point toward using IP for all sorts of service delivery ranging from voice calls to TV. The next natural step in this evolution is to provide the same set of services to the end users independent of the access technology and device used. The IP Multimedia Subsystem (IMS) is an IP based telecommunications platform which targets this and lets the operators develop new services once which can then be used on many different devices. This thesis examines the integration of IPTV and IMS. Can IMS be used to deliver TV services and can the IPTV set-top-boxes of today be used as clients in IMS? Since this is a new and previously unexamined area an explorative approach is taken. The aim is to identify how such an integration could be performed and the possible problems which have to be solved. To assist in this exploration a TV-push service based on IMS technology is constructed. Based on the experiences from this service a general architecture for IPTV in IMS is suggested. A number of problems crucial to solve for a successful integration are identified and possible solutions to these are discussed.

With the development of Internet, more and more people believe that the future telecommunication network will be constructed based on IP technology. Session Initiation Protocol (SIP), which has advantages of simple entrainment method, good scalability and open protocols, is the main research topic on Voice-over-IP (VoIP). Although the client-server architecture currently used by SIP is simple and easy to maintain, it has limitation wherein service quality needs to rely on server performance. To improve this, the Internet Engineering Task Force (IETF) has created a draft to discuss the application of P2P (Peer-to-Peer) architecture in SIP, and we hope that the draft can help to provide good SIP service quality on P2P architecture, such as good fault tolerance and transmission performance. Our research is based on Chord architecture and aims to make P2P SIP architecture in an embedded User Agent. For the SIP internet telephone feature, we adjusts Chord algorithm to meet SIP internet telephone requirements. Furthermore, the adjustment to Chord makes it more applicable to the environment that users continuously join or leave, so that the revised Chord can be implemented with SIP protocol to achieve the P2P SIP goal.

UDP (User Datagram Protocol) and RTP (Real-time Transport Protocol), using fixed bit rate to convey data every time period, are the most pervasive transport protocols for multimedia traffic in communications networks. However, unexpected packet delay/jitter may occur when network becomes congested or channel interference remains unresolved. To reduce packet delay and packet loss for real-time traffic in a hybrid network from wired to wireless ad-hoc, this thesis presents RTP rate control with an ad-hoc gateway to dynamically adjust the transmission rate according to network conditions. With the proposed scheme, a source node can distinguish the two network conditions, congestion and interference, by monitoring RTCP (RTP control protocol) packets regularly reported from destination nodes and the associated ad-hoc gateway. Based on the RTCP reports, a sender node can dynamically change its encoding bit rate to improve the quality of real-time traffic. For the purpose of demonstration, we implement the proposed adaptive rate control scheme on a Linux platform for SIP-phone communications. The experimental results have shown that our proposed scheme not only relieves traffic congestion but also increases the number of received data even in the case of severe channel interference.