An example of the usage of wireless LAN technologyfor multimedia: see Berlin live!

TKN - Telecommunication Networks Group TU-Berlin
Head of Group: Prof. Adam Wolisz Faculty of Electrical Engineering and Computer Science

Handover in All-IP Mobile Networks

1. Introduction

Handover describes a mechanism in cellular networks that transfers the association of a mobile end system from one base station - which is presently active - to a new base station. In general handover is applied when a user moves through the coverage area of a cellular network and crosses cell boundaries. The handover between wireless cells of the same type (in terms of coverage, data rate and mobility) is often referred to as horizontal handover, whereas the handover between wireless cells of different type is characterized as vertical handover. 

In the Internet Protocol (IP) suite the IP address of an end system identifies a host uniquely and also the IP subnet to which the host is attached. Therefore the meaning of the IP address is twofold: end point identification and location identification. When a host changes its point of attachment the IP address must be modified in order to route packets to the mobile's new subnetwork. Unfortunately, ongoing TCP connections break since the IP address is part of the TCP connection identifier and used at TCP connection setup.

Traditional IP-based mobility approaches, such as IETF Mobile IP, were designed with respect to horizontal handover. Thus, the vertical handover and other new services and network architectures pose new requirements on handover design. In our research on handover in All-IP mobile networks we concentrate on these requirements and try to develop new solutions for the problems and challenges arising out of them.

2. Main Research Interests

Our main research interests in this area can be structured into the following categories: 
  • Evaluation of Existing Approaches 
  • Multicast-Based Handover 
  • Security Issues of Handover in Roaming Scenarios 
The following sections give more informations about these topics.

2.1. Evaluation of Existing Approaches

Recent reserach-oriented handover approaches will be evaluated, such as IETF Mobile IPv4, Exensions of IETF Mobile IPv4 (Hierarchical Foreign Agents, Extensions by MosquitoNet), IETF Mobile IPv6, Multicast-based handover, HAWAII , Cellular IP, Mobile People Architecture, ICEBERG, Extended SIP Mobility. A performance evaluation will be performed for selected approaches. This approaches will be compared with respected to handover-specificic performance metrics, such as handover latency, packet loss and others. The performance evaluation will be done by measurement in a testbed (see below) and by simulations, respectively.

2.2. Multicast-Based Handover

In order to solve the IP mobility problems two main approaches can be identified: Address Translation and Indirect Routing (e.g. IETF Mobile IP) or Location-Independent Addressing and Routing. Interestingly, there is already an approach which supports addressing and routing independently of the host's location - multicast. Its ability is similar to the requirement of mobility support, though in a different context. MOMBASA. MOMBASA stands for 'MObility support - a Multicast-BASed Approach'. MOMBASA intends to utilise multicast in order to support network-level mobility.  In relation to the classical approach of Address Translation and Indirect Routing, MOMBASA has three main advantages: 
  • Rerouting for handover is done in a network node where the path to the old and from the new base station diverge (and not in a software agent in the mobile's home network according to the Mobile IP approach). 
  • No "handover-specific" signalling and infrastructure is required, instead multicast is reused for mobility purposes. 
  • MOMBASA minimises the handover latency. In the utmost case packets are distributed in advance to potential new base stations which buffer the packets. In this case the usage of multicast provides the efficient distribution of data to multiple base stations. 
Even for vertical handover the usage of predictive mechanisms, such as registration in advance and predictive data distribution, decreases the handover latency and packet losses. The realization of these mechanisms with multicast is very natural. The today's IP multicast protocols fit well for 'broadcast-like' applications although there are a number of open issues to solve. But handover has requirements similar to 'narrow-cast' applications. Such 'narrow-cast' multicast protocols are matter of research  and it is expected that the future Internet will provide an optimised Multicast. MOMBASA works out the requirements of handover on future multicast protocols and investigates alternate multicast protocols (e.g. CMAP/CMNP, Small Group Multicast, EXPRESS) for their application to host mobility.

2.3. Security Issues of Handover in Roaming Scenarios

As the mechanisms and protocols developed to realize mobile communications also open up new ways for attacking ongoing communications as well as subscriber fraud, specific attention needs to be paid to the security issues of handover design. The most important security service in this respect is the authentication of the entities participating in and the messages exchanged during a communication. Strongly related to authentication is the key management whose task it is to ensure, that the entities performing and verifying an authentication exchange (and only they) possess the neccessary keys for this. 

Our current research topics in this context are:

  • Security Evaluation and Design of an Authentication and Key Management Architecture for Handover in Roaming Scenarios 
  • Performance Issues of Authentication and Key Management 

3. Experimental Testbed

For evaluation and comparison of (selected) existing approaches and the mechanisms developed by our group we use - besides simulation - an experimental testbed as depicted in the following figure: 

Its main components are:

  • a subnetwork unaware of mobility mechanisms where the corresponding host is located, 
  • a WAN-simulator, basically emulating the traffic- and transport-characteristics of the Internet, 
  • a Gateway router to the visited domain, 
  • some routers inside the visited domain, 
  • a cell switch, to ..., 
  • some base stations with different technologies (IEEE 802.11, IrDA), and of course 
  • some mobile hosts. 


4. References

[FAWW00a] A. Festag, T. Assimakopoulos, L. Westerhoff, A. Wolisz. Rerouting for Handover in Mobile Networks with Connection-Oriented Backbones: An Experimenatal Testbed. Proceedings of the IEEE Conference on High Performance Switching and Routing ICATM'2000, pp. 491-499, June 2000, Heidelberg, Germany
[FKS00a] A Festag, H. Karl, G. Schäfer. Current Developments and Trends in Handover Design for All-IP Wireless Networks. TKN Technical Report TKN-00-05, TU Berlin, August 2000.
[FW00a] A. Festag, A. Wolisz. MOMBASA: Mobility Support - A Multicast-based Approach. Proceedings of European Wireless 2000 together with ECRR 2000, September 2000, Dresden, Germany.

5. Contact

For more information please contact one of the following people working in this area: 

Questions? Contact webmaster.

Contents subject to change. All rights reserved.
Mit dem Urteil vom 12. Mai 1998- 312 O 85/98- "Haftung für Links" hat das Landgericht Hamburg entschieden, daß man durch die Anbringung eines Links, die Inhalte der gelinkten Seite ggf. mit zu verantworten hat. Dies kann nur dadurch verhindert werden, daß man sich ausdrücklich von diesen Inhalten distanziert.
"Hiermit distanzieren wir uns ausdrücklich von allen Inhalten aller extern gelinkten Seiten auf unserem Server und machen uns diese Inhalte nicht zu eigen. Diese Erklärung gilt für alle auf unserem Server angebrachten externen Links."