Mobility is a term used to define universal access to various tools and applications that a person relies on to increase the person's productivity. The point of importance here is that these tools and applications are to be made available regardless of the user's location or the device currently in use. A person can access messages (voice, email, fax, etc) from one mailbox, dial one number to reach associates whether at the office or somewhere else and enjoy the same communications features wherever he/she is working; might it be on the road, at home, at office or anywhere else. Mobility is a key feature of future 4G networks as relating to PAN (Personal Area Networks) and other kinds of networks. This makes security a vital concern for implementation of mobility since the user will be using the network for all his/her communications which will contain sensitive personalized information.

One point must be noted here. Mobility itself is not a service. However, it enhances the availability of other services. It can be added to any network system.

In 'Internetwork Mobility - The CDPD Approach' by Mark S. Taylor, William Waung and Mohsen Banan, the authors define mobility as:

Mobility means different things to different people. Some people are quite happy being able to get around town. Others view the world in terms of time distance-four hours from Chicago to San Francisco by airline, perhaps. Obviously, range of motion is an important aspect of mobility.

Another factor in mobility is ease of access. What might be considered mobile in one context is quite immobile in another. Certainly the pioneers crossing the North American continent in ox-drawn wagons covered the same distance as the airliner from Chicago to San Francisco. But we would today hardly consider these pioneers to have had much mobility.

A more pertinent example of mobility is the ever decreasing size of cellular telephones. What was once considered a "mobile phone" had to be transported in a vehicle. A major step forward was the "transportable" phone, which freed the user from their vehicle but weighed in at about twenty pounds, still huge by today's standards. With the advent of "brick" phones in the mid-1980's came the era of "portable" phones. This continuing decrease in size and weight of handsets has greatly increased the mobility of cellular subscribers.

In this book on mobile data communications we define mobility as the ability to send and receive communications anytime anywhere. Mobility means that both source and destination devices, applications and people are free of the constraints imposed by physical location. Access to an Ethernet port, for example, need not limit one's ability to send and receive data in a mobile WAN environment any more than access to a landline phone currently limits one's ability to place a voice call in an area covered by cellular service.

There are two important aspects to mobile communications:

1. Mobile network access

2. Mobility management

Keeping the question in mind, I shall go directly to explaining mobility management. Mobility management is to efficiently send and receive communications once you are connected to the network. Sometimes, this is also called location management which gives a much more clearer picture of what the term means. The implications of mobility management are that the person using mobility enabled services should be contactable regardless of where ever he or she is. What this means is that the user is track able so that his data can be successfully sent to his/her device. This can be a challenge for mobile communications depending on the infrastructure being used and the spread of base stations across any particular region. Groups of base stations form a location or routing area over which Temporary Mobile Subscriber Identity (TMSI) or International Mobile Subscriber Identity (IMSI) can be used to obtain the user location. This has critical effects on mobile data applications which may be utilizing mobility enabled services. Therefore, mobility management is a major feature of mobile communication networks like GSM, UMTS, CDMA, etc. This is critical for features like roaming which need mobility management to work properly.

Mobility can be classified into two broad categories:

1. The category of the user

   Mobility benefits different kinds of users. These users can be a person, hardware or software. These user types lead to various types of mobility:

   1. Personal mobility: Personal mobility allows a human being to access or to be accessed by the network independently of where the access point and terminal used are located in the network and maintaining all services. An example can be a person who wants to be contacted on his stationary PC in the office during the working hours and by an IP-phone at home the rest of the day.

   2. Terminal mobility: Terminal mobility permits a terminal to change location while maintaining all the services. Terminal mobility can be provided at different layers and thus, different approaches exist. Mobile IP is an example of how terminal mobility can be achieved on the network layer.

   3. Component mobility: It allows components e.g. earphone, microphone, etc. of a terminal to change location while maintaining all services. The components form a PAN (Personal Area Network).

   4. Application mobility: Application mobility allows a software process to be relocated from one machine to another or even moved between machines while processing.

   5. Session mobility: Session mobility is an additional feature to those above. Session mobility is ensuring that active sessions are not disrupted while terminals, persons or applications are moving or being relocated.  However, sessions may be brought to a well-defined halt state in order to be resumed later. An example of session mobility can be handover in cellular systems, permitting calls to continue while the mobile terminal moves from one cell to another. In a GSM network, the session is maintained without interruption, i.e. without loss of information during handover.

   6. Role mobility: Role mobility is a relatively new type of mobility. Usually, each individual has multiple roles, e.g. employee of a company, head of a family, etc. Furthermore, an individual may have several job positions in different companies and hence different roles. For each role, there is defined a set of services with distinct preferences, rights and limitations. Role mobility aims to assist the user to move from one role to another easily and smoothly.

    

2. The availability of services

   Services can be either available all the time, partially available or unavailable. There is no mobility for unavailable services so I will not discuss it. Considering the always available and partially available services, there are two types of mobility:

1. Continuous mobility: It aims to provide the user continuous services while the user is moving. A common example of this can be the cellular networks. It makes use of session mobility for this. Global continuous mobility is an aim of the 4G networks where various kinds of networks can interact with each other and provide the user with continuous mobility. The picture below shows the arrangement of cells in a typical cellular network providing continuous mobility:

2. Discrete mobility: It aims to provide availability of services within certain areas or access points only. The services are not available while moving from one area to another. Session mobility is used in discrete mobility too to ensure that sessions are properly halted and maintained while the terminal is being shifted to another area. The picture below shows the arrangement of access points for discrete mobility:

The two classifications of mobility shown above cannot exist independently. They correlate orthogonally:

1. Personal continuous mobility: This is not yet supported on a global scale thought its support mechanism (Session Initiation Protocol) exists. It allows a person to move freely between different domains and terminals and remain available for the calling parts without additional notification of them on the user’s current IP address. This means that information about a certain user is kept on a server and is updated each time the user’s IP address changes. As a result, anyone, who wants to contact this user, contacts this server first and the server is responsible for finding the called part. After both parts have found each other, a session between them is initiated and information exchange can proceed.

2. Personal discrete mobility: This is widely implemented today. An example can be that of a LAN where each user is given his or her own login credentials (user name and password). Multiple users using their own login credentials can initiate sessions with each other.

There is still a lack of session mobility in personal continuous and personal discrete mobility. Due to this, a change of terminal requires re-initiation of the session.

3. Terminal continuous mobility: This is not yet supported on a global scale though the mechanism exists in the form of Mobile Internet Protocol which allows one terminal to travel between different domains and each time obtain new IP addresses and still remain available to the counter parts without additional notification of them on the terminal’s current IP address. Mobile IP supports session mobility as well. Due to the additional units, Home Agent and Foreign Agent, which are involved in the communication process, a terminal is able to change its IP address and keep the ongoing session.

4. Terminal discrete mobility: This is in wide use today in WLANs. When a terminal is connected to a WLAN it can be moved freely inside the coverage area of this WLAN. The terminal can even change its point of attachment as long as it is on the same subnet as the previous point of attachment. In this case the session mobility is supported on the link level and the ongoing sessions do not suffer from this handover.

5. Application continuous mobility: Application continuous mobility has not been achieved on a global scale. As for personal and terminal continuous mobility a technique for it exists. It is known as a Mobility Agent. Mobility Agent is a program that executes on behalf of a person or an organization, which in addition may migrate from host to host during the execution process.

6. Application discrete mobility: A java applet residing on one host, which is downloaded to another host in order to be executed there can be an example of this kind of mobility. However, unlike Mobile Agent, it cannot migrate forward to another host.

I don't know much about the OSI layers or the TCP/IP protocol. They have been mentioned during the ICT course (1st semester) but weren't explained due to the shortage of time at the end of semester. I have tried to understand them by myself from online sites but instead of clarifying the concepts, I have gotten quite a bit confused. In the questions below, I have tried to answer the best I can, to my own understanding.

Can tutorials for this be arranged under some faculty member so that we can have clear concepts regarding these important topics in network communications? I am sure this will benefit other K World members too.

Today, mobile computing is an area experiencing high growth. Advances are being made in miniaturization technology resulting in increased mobility. TCP/IP on the other hand has been in a constant state of evolution since its inception more than three decades ago. TCP/IP works well for non-mobile devices like a desktop computer since it is stationary. Unfortunately, the implementation of TCP/IP does not supports mobility. The reason for this is the IP addressing and routing function.

IP addresses are fundamentally divided into two portions:

• A network identifier which specifies which network a host is on, and

• A host identifier which uniquely specifies hosts within a network.

This structure of the IP addresses is fundamental to the datagram routing. Devices use the network identifier to determine whether the recipient is on a local or remote network and routers use it to determine how to route the datagram. This results in the IP address being tightly tied to the network where the device is located. For mobile devices, this spells trouble since when the device travels away from it's fixed location, the system of routing based on IP address will not be able to locate the device for sending the datagram.

The solution, though not asked for in the question, lies in forwarding the datagram from the fixed IP address to the current IP address of the device. However, this is an inefficient method since the data has to travel twice through the inter-networks. The device needs to have two addresses:

• Home address: This is the permanent IP address assigned to the device.

• Care-of address: This is a temporary address used by the device while away from 'home'. It is a 32 bit IP address and is used only by Mobile IP for forwarding IP datagrams to the device. The higher OSI layers don't use it. There are two types of care-of addresses:

  • Foreign Agent care-of address: A foreign agent is involved in the delivery between the home network and the device. The device has no distinct IP address.

  • Co-Located care-of address: This kind of address is assigned directly to the device. It may be assigned manually or through DHCP. The traffic from the home agent is forwarded directly to the device.

Co-located care-of addresses have the advantage of flexibility, but require each device to have a unique IP address on the remote network. Foreign agent care-of addresses have the chief advantage of allowing many mobile devices on a foreign network without each requiring a distinct IP address.

While searching for an answer to this question, I wasn't able to find much information. Therefore, I will speak generally here.

Mobility for all OSI layers means that each OSI layer can support the requirements of mobility. This means that datagrams must be able to travel layer by layer without any hindrance so that mobility features and services can be supported.

The OSI reference model consists of 7 layers. Sometimes in some protocols, multiple layers are combined to form one layer. One such example is the TCP/IP protocol. The layers are also divided into higher and lower layers. The lower layers; namely the physical, data link and network layers deal with raw data from the hardware. The higher layers consist of transport, session, presentation and application layers. The lower layers can not usually hide the effects of mobility from the upper layers. Following are the layers and which services that they provide are affected by mobility:

On the physical layer, mobility affects data transmission and reception, and the topology and physical network design.

On the data link layer, mobility affects Logical Link Control (LLC), Media Access Control (MAC) and addressing.

On the network layer, mobility affects logical addressing (the layer three address), routing and datagram encapsulation.

On the transport layer, mobility affects process level addressing, multiplexing, demultiplexing and connection establishment, management and termination.

On the session, presentation and application layers, mobility affects translation and specific application, if any.

I am assuming that both; continuous and discrete mobility are affected in each case. Here in the table below, I am combining the session, presentation and application layers:

From the table, we can see that the OSI layers affect all mobility types which were discussed above in a previous question.

To quite an extent, the answer to this question has been given in question number five since the services offered by each layer are affected. The physical layer can be called the 'wire'. Since there may or may not be a wire used for communication, mobility will affect the physical layer. The data link layer encodes and decodes the data packet which includes the address. This address might be affected by mobility depending on where the user travels to with the device. In the network and transport layers, the switching, routing and the actual transfer of data between end systems is affected. In the session, presentation and application layers, the connection establishment, management and termination are affected alongside the various data formats. All of this is because when considering the concept of mobility purely, the device has no fixed address. However, using Mobile IP resolves these issues as discussed in another question before.

For the purpose of mobility, layers 3 and 4 are most affected. Due to this I won't discuss layers 1, 2, 5, 6 and 7. Information for this answer has been taken from a seminar on internetworking by Mika Ratola. Quoting from the seminar:

According to the OSI-model, the network layers main functionalities concern the controlling of the subnet. Key issues here are routing of packets, congestion control, and the interconnection of different networks. On the other hand, the transport layers main function is to accept data from the layer above, split it up into smaller units if needed, pass these to the network layer, and ensure that the pieces arrive correctly at the other end. Related issues here are handling multiple connections, multiplexing, flow control, and creating an error-free point-to-point channel that delivers data in the order it was sent.

Three solutions being considered are usually in the form of a separate mobility layer:

• Mobile IPv6 as layer 3,
• Host Identity Payload (HIP) between the 3rd (Network) and the 4th (Transport) layers,
• Stream Control Transmission protocol (SCTP) as a protocol for the transport layer.

More details about each solution can be read on the linked seminar PDF file above. Another article by the name of "At What Layer Does Mobility Belong" by Wesley M. Eddy also highlights some possibilities of mobility at the network, transport and/or session layers.

Using the solutions above, the outcome from the most needed layers (Network, Transport) will experience some increase in size due to extra data. However, this will enable the support for mobility and services which might utilize it.

References:

http://www.faqs.org/rfcs/rfc3753html

http://www.ambient-networks.org/publications/Appendix%20of%20D1_2%20public.pdf

Ramjee Parasad and Rasmus L. Olsen. "The Unpredictable Future: Personal Networks Paving Towards 4G"

http://www.tcpipguide.com/index.htm

Back to K World home page

Copyright © 2006 Asad Asif - All rights reserved.