SIP Handbook

Services, Technologies, and Security of Session Initiation Protocol

Syed A. Ahson, Mohammad Ilyas, Syed A. Ahson, Mohammad Ilyas

• Publisher: CRC Press
• Year: 2018
• eBook ISBN: 9781351834827
• Topic: Technology & Engineering
• Subtopic: Computer Networking
• Index: Technology & Engineering

Widely adopted by service providers to enable IP telephony, instant messaging, and other data services, SIP is the signaling protocol of choice for advanced multimedia communications signaling. Compiled by noted engineering experts Syed Ahson and Mohammad Ilyas, SIP Handbook: Services, Technologies, and Security of Session Initiation Protocol presents a thorough technical review of all aspects of SIP. It captures the current state of IP Multimedia Subsystem technology and provides a unique source of comprehensive reference material on this subject.

SIP Applications for Today and Tomorrow

The scope of this volume ranges from basic concepts to future perspectives. Divided into three sections, the book begins with a discussion of SIP in peer-to-peer networks and then goes on to examine advanced media integration, migration considerations, mobility management, and group conferencing, while also reviewing home networking and compliance issues.

The middle section of the book focuses on the underlying technologies of SIP. Chapters review network architecture, vertical handoffs, NAT traversals, multipoint extensions, and other areas at the forefront of research. Finally, the text examines various security vulnerabilities and provides perspectives on secure intelligent SIP services with a future outlook on a fraud detection framework in VoIP networks.

Insights from International Researchers

Authored by 65 experts from across the world, this text is sure to advance the field of knowledge in this ever-changing industry and provide further impetus for new areas of exploration. Because of the editors' pivotal influence and their proximity to both the current market and the latest science, this work is certain to become the definitive text on this emerging technology.

Part II

TECHNOLOGIES

9

P2P SIP: Network Architecture and Resource Location Strategy

Guiran Chang, Chuan Zhu, and Wei Ning

Northeastern University

CONTENTS

9.1 Introduction

9.2 Peer-to-Peer Network Technologies

9.3 Early Development of P2P SIP Network Architecture

9.4 P2P SIP Standardization

9.5 Protocol Development and Implementation

9.6 Concluding Remarks

References

9.1 Introduction

Session Initiation Protocol (SIP) has been widely used in Internet telephony and has been chosen as the protocol for Internet Protocol (IP)-based multimedia call control for third generation (3G) wireless networks. As defined in RFC 3261 [1], SIP is a client server-based control protocol above the transport layer for creating, modifying, and terminating sessions between two or more participants. However, SIP-based telephony can be viewed as an application of peer-to-peer (P2P) architecture where the user agents (UAs) form a self-organizing P2P overlay network to locate and communicate with each other. In consideration of the characteristics of P2P, it is expected to be a perfect method to solve the scalability, robustness, and fault tolerance problems in traditional SIP networks. Skype is a free P2P application based on the Kazaa architecture for Internet telephony and instant messaging [2,3]. The protocol is proprietary and the system has centralized elements for login authentication [2]. Researchers have proposed several pure P2P architectures for SIP-based IP telephony systems [4,5]. The P2P SIP working group has been formed within the Internet Engineering Task Force (IETF) for adapting P2P with features suitable for SIP.

9.2 Peer-to-Peer Network Technologies

An overlay network is a computer network built on top of another network. Nodes in the overlay are connected by virtual or logical links, each of which corresponds to a path in the underlying network [6]. An example overlay network built on top of another IP computer network is shown in Figure 9.1. The overlay network resides at the edge of IP network and is ignorant of the underlying IP network. The nodes of the overlay network use only the IP addresses from the underlying network for discovery and routing on the application layer.

The term “peer-to-peer” (P2P) refers to a class of systems and applications that employ distributed resources to perform a function in a decentralized manner. A P2P network relies primarily on the computing power and bandwidth of the participants in the network rather than concentrating it in a number of servers. Because it exists on top of a physical layer network, a P2P network is often called a “P2P overlay” or “P2P overlay network.” A typical client-server network is shown in Figure 9.2a, while a P2P network is illustrated by Figure 9.2b. In comparison with the client-server network, P2P networks are inherently scalable and reliable because there is no single point of failure. Peer-to-peer systems operate without centralized organization or control and, in the purest form, have no concept of servers. All participants are peers and communicate in a distributed, potentially untrusted environment to achieve a certain objective [6,7].

FIGURE 9.1
Example of an overlay network.

FIGURE 9.2
Client-server vs. peer-to-peer network.

The core operation in most peer-to-peer systems is efficient location of data items. The resource location strategy of a P2P network can be classified into four types: centralized unstructured directory model, pure decentralized unstructured flooded requests model, hybrid unstructured model, and decentralized structured distributed hash table (DHT) Model [8]. Currently, DHT-based structured P2P systems are the main technology used in P2P SIP researches, and the most popular algorithms include Chord, Pastry, and Bamboo, which can be easily implemented and be used to solve the problems faced by P2P SIP.

9.2.1 Chord

Chord [9,10] is a ring-based DHT with an identifier space of size N. A Chord node with identifier u has a pointer to the first node following it clockwise on the identifier space (Succ(u)) and the first node preceding it (Pred(u)). The nodes form a doubly linked list. A node keeps M = log₂(N) pointers called fingers. The finger points to the edge on the Chord, which will be used for quick resource location algorithm. The set of fingers of a node u is:

$$F_u=\left\{\left(u,succ(u+2^{i-1})\right)\right\}1\le i\le M,$$ (9.1)

here the arithmetic is modulo N.

With this choice of edges, a node perceives the circular identifier space as if it starts from its identifier. Then the edges are chosen in such a way that the space can be partitioned into two halves, one of the halves can be partitioned into two quarters, and so on.

As an example, let us look at the network in Figure 9.3a with an identifier space N = 16. Each node has:

$$M={\log }_2(N)=4\text{edges}.$$ (9.2)

FIGURE 9.3
(a) A chord network with N = 16 populated with 6 nodes and 5 items. (b) The general policy for Chord’s routing tables. (c) Example of inserting a new item.

The network contains nodes with identifiers 0, 2, 5, 9, 12. The policy for constructing routing tables is shown in Figure 9.3b. As pointed out earlier, Node n chooses its pointers by assuming that the identifier space starts from its identifier. It chooses to have the pointers to the successors of the identifiers:

$$n+2^0,n+2^1,n+2^2\text{and}n+2^3.$$

In this way, the last pointer n + 2³ divides the space into two halves. The one before it, n + 2², divides the first half into two quarters, and so on. When no node exists at the desired position, its successor will be taken instead.

As shown in Figure 9.3a, an item is stored at the first node that follows clockwise on the identifier space. If items with identifiers 3, 6, 9, 11, and 15 are to be stored in this network, then {3} will be stored at 5; {6, 9} at 9; {11} at 12; and {15} at 0.

The lookup process follows the result of the way the identifier space is partitioned. Both the insertion and querying of items depend on how to find the successor of an id. As illustrated in Figure 9.3c, if node 12 wants to insert a new item with an identifier of 7, the lookup is forwarded to node 5, which is the closest preceding finger to the identifier 7 from the point of view of 12. Similarly, node 5 will lookup its set of fingers to find the closest successor of 7. Node 5 finds that 7 is between itself and its successor 9, and therefore, returns 9 as an an...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Preface
• Editors
• Contributors
• I Concepts & Services
• II Technologies
• III Security
• Index