Data communication and computer networking

The growth of the Internet mirrors the rapid development of new protocols, mechanisms and the remarkable increase of Internet users, for data communication and computer networking. This growth, in turn, has been liberated by the exponential growth of the World Wide Web (WWW), or Web for short historically the Web has triggered explosive escalation in the Internet as e-mail has been triggered fast growth after ARPANET establishment. The great increase in traffic volume generated from the Web, real-time, multimedia, and multicasting applications was motivating the researchers for developing new technologies that can support better Quality of Service (QoS) [reference].

In terms of QoS principles, the Internet and Internet Protocol (IP) was established to offer a best-effort and fair delivery service under which all packets are treated equally. But this delivery is slowed down as a result of the increase of traffic volume, the traffic congestions and the random drop property of normal routers behavior. This issue contributed the foundation of new QoS mechanisms, Traffic Engineering (TE) and traffic classification (elastic and inelastic traffic). The elastic traffic, like TCP/IP, can be tuned and adjust to changes in QoS parameters such as delay and throughput across an Internet and still practical for the needs of its applications (like file transfer, electronic mail, remote login and Web access). On the other hand, it is not easy to adapt the inelastic traffic, if at all, to variations in delay and throughput, this difficulty appears clearly in real time traffic adaptation such as voice and video. QoS is an emerging feature of modern networks and becoming one of the hottest topics in IP convergence technology for multimedia services and real time applications like VoIP, video conferencing and IP TV.

IPv6 stands for Internet Protocol Version 6. It is the next generation protocol for the Internet, designed to offer several advantages over current Internet Protocol Version 4 (or IPv4) and to brings QoS that is essential for several new applications such as VoIP, video/audio, interactive games or ecommerce. Whereas IPv4 designed to provide best effort service, IPv6 designed to ensure QoS, support a set of service requirements in order to deliver performance guarantee while transporting traffic over the network. Also IPv6 addresses the main limitation of IPv4, that is, the exhaustion of addresses to connect computers or host in a packet-switched network. IPv6 distinguished in covering a very large address space and consists of 128 bits compared to 32 bits in IPv4[ reference]. The main properties of IPv6 are increased address space, more efficient routing, reduced management requirement, improved methods to change ISP, better mobility support, multi-homing, security, scoped address; link-local, site-local and global-address space

In the literature [1], three transition mechanisms to IPv6 have been launched: Dual Stack, Tunneling, and Translations. In this thesis, two of these transitions approaches are considered at MPLS-enabled backbone; CE-to-CE tunnel and dual stack.

The traffic engineered MPLS technologies is highly distinguished as the most modern approach that guarantee the high level of quality and reliability that we expect from telephony services [2]. MPLS architecture is defined in RFC3031 [3]. Moreover the coexisting of MPLS facilities: (speed, traffic engineering, QoS, VPN, and resiliency), with IPv6 features like QoS insurance, wide addressing coverage security and others, both together (IPv6 & MPLS) was presented by IETF as one of the robust infrastructures (solutions) for QoS Internet backbone requirements and mainly supported by Cisco and juniper vendors.

Actually most of the real time applications are delay sensitive and need guaranteed bandwidth to work effectively. Practically different services require different demands. For example, video/audio telephony requires low latency delay, whereas digital TV demands low bit error loss, and the low delay demand for gamming. As a result the glow of IPv6 insurance in supporting the QoS demands of real time applications push for more research in bandwidth utilization and header compression schemes. Traditionally various types of codecs were used to compress and transmit the VoIP packets of real time applications, and most of these codecs produce small packets sizes. They are used to convert an analog voice signal to digitally encoded version; also they vary in the sound quality, the bandwidth required, and the computational requirements. Obviously voice datagram(s) [4] are on the order of 20 bytes while IPv6/UDP/RTP headers are on the order of 100 bytes. GSM codec is considered as a sample VoIP codec for this thesis.

Many compression and suppression techniques were developed to save the bandwidth especially for satellite (the most costly), wireless (the limited and costly one) and also at the high speed networks. Typically whereas compression schemes use encoding, decoding and feedback-updating processes of packet header in order to synchronies their compressor state with corresponding decompressor state, suppression schemes built on striping out the header fields which are redundant in successive packets of a certain stream.

The challenges for the present thesis work come from the limitations of previous works which can be summarized in these points:

  • Limited comparative studies between compression and suppression schemes in terms analysis and complexities.
  • There is a seldom (if not at all) studies that explores the limitations of header compression techniques in terms of different domains. The studies mainly focus on wireless side.
  • There is a partial fair in highlighting between the disadvantages and advantages of header compression and header suppression in terms of the extra complexities.
  • Still there is seldom (if not at all) implementation for header suppression of IPv6 over MPLS for real time applications.
  • How this these can produce competitive scheme compared to the existing strong schemes.
  • How the proposed methodology satisfies the requirements of header compression over MPLS.

Problem Statement

One of the most critical aspects in transmitting real time streams over IPv6 networks is the increase in packetsize compared to the small payload size, which represents extra overhead (or successive headers overhead problem). These overheads are considered additional costs in terms of complexities metrics like: the time complexity for packet header processing, the storage resources like queuing requirement, and the transmission bandwidth requirements. As a result these overheads might contribute negatively in the networks performance (QoS) and increases the happening probability of traffic congestion problem. In terms of QoS parameters, real time applications are very sensitive to the delay and jitter. For example, the successive headers overheads appear clearly in VoIP stream, it uses the packetization of protocols: voice/RTP/UDP/IP (where voice transporting over the Real Time protocol (RTP) of network application layer, over User Datagram Protocol (UDP) of network transport layer, over Internet Protocol (IP) of network layer.

Research Motivation

Three trends motivate this article: first, the growth of telecommunications industry interest in the implementation of MPLS as an efficient transport technology, the second, the increasing needs for the investigation of IPv6 performances with challenges of bigger header sizes compared to the smaller sizes of the packets payloads inside the internet backbone infrastructures like MPLS, and third, looking for more IPv6 research challenges in exploring the effect of the total size of static fields over the non-static fields size ratio of packet's header, on the choosing of compression techniques.

Thesis scope and limitations

The important question that needs to be answered is "How to improve the QoS parameters and utilize the bandwidth with a suitable compression technique for the high speed transmission media like MPLS-enabled backbone which requires low and fast complexity approach for Header Compression (HC) and Header Decompression (HD)". Also it is necessary for any proposed technique to be compatible with the symmetry behavior of Ingress and Egress points (the push and pop) of MPLS labels manipulation. Thesis scope is intersection field of four main research domains, from the largest domain to the narrow one, are QoS, IPv6, MPLS and header suppression.

The theses limitations can be summarized as follows:

  • Tested for GSM-FR codec of VoIP applications.
  • One MPLS cloud
  • Support tunneling approaches which use protocol 41.

Thesis objectives

Thesis objectives are summarized as follows:-

  • To Study the QoS performance metric for the native IPv6, 6-in-4 tunneling (which uses protocol 41) and investigates the domain limitations for applying the existing compression techniques in order to choose the proper technique to support the real time applications at the MPLS-enabled backbone.
  • To Study the IPv6-MPLS facilities and the effect of the compressed packet size on the QoS parameters like throughput, delay and jitter metrics.
  • To investigate the performance of IPv6 at the MPLS-enabled backbone and the possibility of applying packet header suppression to support the real time applications like VoIPv6.
  • To propose a new framework (MPHS-MPLS service) to enhance the QoS performance of real time applications at the MPLS-based backbone. The proposal advances the PHS approach of WiMAX domain which is limited for supporting the BS to MS link, compared to the proposed framework (MPHS_MPLS) at which multi-Ingresses to multi-Egresses MPLS-LSPs connectivities are found.
  • To support the Explicit Label Switch Path (E-LSP) of MPLS cloud with MPHS service. E-LSP is one of the MPLS properties that permit the booking of an explicit LSP which is not necessary a shortest path. E-LSP can be deployed for different situations, like fast restoration path (in failure cases of node/links), for MPLS-Traffic Engineering usage, load balancing and so on.

Thesis Contributions:

The present work offered better bandwidth utilization by supporter larger number of real time streams under acceptable QoS parameters This thesis presenting two main contributions for MPLS-IPv6 domain:

  • First is introducing a comparative study between compression and suppression schemes in terms analysis and complexities.
  • Second is supporting the Native IPv6 packets with header suppression service called MPHS-N algorithm,
  • Third is supporting the 6-in-4 Tunneling streams between Customer Edges (CE-to-CE) over MPLS-enabled backbone with header suppression service called MPHS-T algorithm,
  • Fourth is supporting the Explicit Label Switch Path (E-LSP) of MPLS cloud with MPHS service (E-LSP is one of the MPLS properties that permit the booking of an explicit LSP which is not necessary a shortest path. E-LSP can be deployed for different situations, like fast restoration path (in failure cases of node/links, for MPLS-Traffic Engineering usage, load balancing and so on).
  • Fifth is testing the effectiveness of the developed MPHS Algorithm using NS-2.33 over different topologies load. Analyze and compare the proposed framework to the existing standard approaches of header reduction like RoHC and to the normal MPLS work in order to evaluate its reduction performance.

Thesis Organization:

The rest of this thesis is organized as follows:-

  • Chapter 2 covers background on IP header facilities, header suppression and compression schemes algorithms, MPLS technology and requirements for header compression over MPLS, QoS approaches, and real time applications.
  • Chapter 3 defines the design of MPHS algorithm, and all its features including its components and functions, how it is working, and the integration with the existing modules of MPLS Ingress and Egress.
  • Chapter 4 describes simulation environments, network scenarios, and QoS performance metrics used, while simulation results, analysis and discussion are presented in Chapter 5.
  • Chapter 6 provides the over conclusion and future research work.

Please be aware that the free essay that you were just reading was not written by us. This essay, and all of the others available to view on the website, were provided to us by students in exchange for services that we offer. This relationship helps our students to get an even better deal while also contributing to the biggest free essay resource in the UK!