Back to Main Page This information is provided by Johnson Consulting

Seamless Roaming

  1. What is Roaming?
  2. What is "Full Handover"?
  3. Interconnection and Roaming
  4. What´s a LAN?
  5. What´s a WLAN?
  6. WLAN and WiFi, what´s the difference?
  7. How 3G Functions
  8. Why integrate WLAN and 3G technologies?
  9. Definition of Seamless Roaming
  10. Seamless Roaming Technology
  11. Overview of Mobile IP operation

Figure 1


In our "brave new world", mobile computerized units of various kinds (not only cellular phones) will need to send data packets back and forth, if they are to use the services being offered. These data packets are quite sensitive to momentary interruptions at hand-over time, which is a challenge that needs to be met. To understand "seamless roaming", let us first recall what "roaming" is. It is a, by now, rather old technique, whereby a travelling mobile phone transfer its connection from one tranceiver station to another. The illustration at right will suffice to give you an idea.

What is Roaming?

Let´s say that the driver is talking into his cellular phone while driving (which is against the law in some countries!). The signal strenght is then constantly measured by the stationary station (A in the figure). Actually, this signal measuring goes on all the time, as long as the phone is "ON".

When you dial a number, station A "knows" that it is its responsibility to reply. Likewise, when you receive a call to your mobile unit, the cellular phone network "knows" to which station it should relay the call. If the signal strength should fall under a certain level, or the noise factor is too high for good reception, then the station will send a signal to nearby stations to contact this telephone and measure resulting signal strength. Hopefully, at least one of them (B in the figure) will report a satisfactory signal strength, whereupon the two stations (A and B) agree on a so-called "handover", i.e. the communication is handed over to station B.

For cellular phones, this handover is rarely noticed; one might sometimes hear a "click" or notice a change in sound quality. No big deal.

But when it comes to data transfer, this handover procedure can be quite disruptive. data packets might get corrupted and have to be resent. For this reason, Seamless Roaming becomes interesting. As you might expect, it is a technique to handle this handover so smoothly that there is no disruption in data transfer. The Bluetooth protocol has dealth with this problem in an efficient way, but Bluetooth is not really meant for cellular phones.

What is "Full Handover"?

Handover, as noted above, is when one station in a mobile phone network agrees to take over communication with a mobile unit from another station. More broadly speaking, the term applies to any situation where an on-going session with a mobile unit is transferred from one stationary unit to another. If this technology is not implemented in a wireless network, the mobile unit will find that the session is broken off when contact with the stationary unit is lost. And one of the parties will have to establish a new session with the same stationary unit, or with another such unit.

Now, "Full Handover" is more ambitious than a normal handover, although not quite as ambitious as the subject matter here; "Seamless Handover". Originally, the GSM-standard did not foresee the need for integration with other kinds of wireless networks. And so, with UMTS (the Europen "version" of 3G) complementary functions needed to be implemented, and these affects all the three key components of the 3G-networks; "RBS" (The Radio Base Station), "RNC" (The Radio Network Controller) and "RANOS" (The Radio Network supervisory system). RNC is the component responsible for the mobile unit being connected to the "right" base station (i.e. the station with the best reception quality at the moment).

The aim of "Full Handover" is to ensure that not only different wireless networks can hand over mobile units in between them, but also that base stations from different manufacturers can smoothly handle this transfer of mobile units in between them. This requires that base stations of different makes, in different networks, can communicate directly in between them, and mutually agree on how to treat the mobile unit being subject to handover.

Interconnection and Roaming

The future is almost here, with wireless LANs (WLANs) and third-generation mobile phones (3G) with a bandwith of 2 Megabit/second.

Full hand-over; continually switching base stations and, occasionally, even networks.

Naturally, we will want to:

  1. Use laptop computers and similar devices to communicate with our company´s WLAN.

  2. Use the same laptop for 3G connections (using the local technology for this; "CDMA 2000" in USA and "UMTS" in Europe).

  3. Use these connections to transfer data.

  4. Seamlessly switch from a WLAN-connection to to the 3G network when we get out of range of the WLAN, without loss of data (and back again, when we are within reach of the WLAN).

Items 1, 2 and 3 should not be too complicated, considering what today´s technology can do. But that fourth option, is that really possible? Yes, there are proposals for how this could be done, and we will look at that possibility here.

What´s a LAN?

Example of LAN interconnections

A "Local Area Network" (LAN) is simply a group of computers, having varying functions and networking with each other through cables. They need not be in the same physical location. With a technique known as Virtual Private Network (VPN), local networks can inter-connect through the Internet and form larger LANs (see illustration below). Computers on LANs can use IP-addressing to find each other, and if they are protected by properly configuerad firewalls, they need not be concerned as to wether these same IP-addresses exist somewhere else on the Internet.

What´s a WLAN?

Example of wireless LAN interconnections

A WLAN is a Wireless LAN. The mode of operation is the same as for a LAN, with the notable difference that the communication carriers are not cables but radio waves. WLANs usually have access points to normal LANs. Topologically, these accesspoints function as routers, handling the LAN and the WLAN as two different networks. An accesspoint can be an ordinary PC, equipped with WLAN interface card and antenna, as well as the ordinary LAN card.

The principle of forming larger LANs using VPN-channels through the Internet

Figure 2

WLAN and WiFi, what´s the difference?

Wireless LAN-technology (WLAN) is, like Bluetooth, based on the IEEE 802.11 communications standard and they use the same frequencies for carriers. But the IEEE 802.11 standard is subdivided into various protocols, meant for different purposes. WLANs use ordinary LAN protocols for communication. "Wi-Fi" is the name of the technology based on 802.11b, which is one of the sub-sets of IEEE 802.11. Wi-Fi is an open-standard technology that enables wireless connectivity between laptops and ordinary local area networks (LANs). Today's Wi-Fi products, which transmit in the unlicensed spectrum at 2.5 GHz, are capable of speeds of up to 11 Mbps, which is about seven times faster than a typical T1 connection. Wi-Fi´s pervasiveness and its ability to provide access at almost every physical destination in the world makes it a challenge even for cellular phones and the need for 3G services.

The WiFi technology is, with its use of unlicensed radio frequency spectrum, relatively simple deployment and very low equipment cost, sometimes regarded as a competitor to the 3G technologies. The key difference that makes these technologies complementary, rather than competitive: is that WiFi networks have a more limited range than 3G networks.

Initially, WiFi networks were designed to extend LAN networks without the cabling of fixed LANs, which is why they’re also known as wireless LANs or WLANs. Now, however, WiFi is used for more than mobile high-speed data on company premises. Nowadays, Wireless Internet Service Providers (WISPs), are using the WLAN standard IEEE 802.11b to extend broadband access to public areas, such as airports, hotels, conference centers and cafés. These access points to the Internet are called “hot spots.”

How 3G Functions

The so-called third generation (3G) of mobile phone communication has a bandwith of 2 Mbit/second. The two dominating technologies that uses 3 G is "UMTS" in Europe and "CDMA 2000" in USA. The 3G systems use a "forest" of antennas in cities and wherever people live, with the ambition of reaching everywhere that there are people. 3G offers a truly wide range of services, somewhat improving upon the two previous versions of mobile phone systems:

  • Permanently connected to the Internet
  • Ability to send and receive pictures and short videos, as well as SMS-messages
  • Access to do Internet-based services, such as banking
  • Synchronisation of address registers and calenders
  • Bluetooth
  • e-mail
  • Positioning and navigation
  • Video-based conferenses
Now, isn´t that something?

Why integrate WLAN and 3G technologies?

By making use of the basic character of today’s mobile high-speed data technologies, companies can take full advantage of the opportunities these technologies present, chiefly to increase productivity and cut costs. This integrated approach offers expanded coverage, with faster, more convenient access to data from a broader range of locations than up to now has been possible. It also presents users with an important opportunity to upgrade business processes, making them more cost-effective and competitive. In short, the investment might well be worth it!

Despite the expansion in usage, WiFi’s limited range makes it impractical as a technical and economic alternative to 3G technologies for ubiquitous wide area coverage. As a result, businesses can gain greatest advantage from mobile data solutions that seamlessly integrate these two types of mobile technology—just as they gain more power and convenience by integrating WLANs and WANs (= Wide Area Network).

By integrating WiFi networks with 3G resources, telecommunications operators can extend mobile high-speed access in more locations and faster than before. There already exist IEEE 802.11b-enabled laptops and personal digital assistants (PDAs) which can be used in WiFi "hot spots". Integrated 3G mobile solutions leverage these existing locations, and extend them by giving end users the ability to roam across the wider area, supported by 3G networks.

So, consider the advantages of this: General access to intranet resources at all times! While on company premises, employees can use the corporate WLAN for mobile data connectivity. At home or on the road, they can use wide-area 3G coverage to quickly access the same resources. As a result, employees can save time and increase their productivity.

Principles of roaming between 3G-networks and WLAN

Figure 3

Definition of Seamless Roaming

We mentioned above that data transfer is considerably more sensitive to disruptive handover-procedures. For this, we need seamless roaming (which is also sometimes referred to as “service continuity”). How can this be achieved?

For data connections, this phrase means that a network’s end users can move easily between WiFi/WLAN technology and 3G technology without really noticing the effects on their data connection. What’s required to achieve service continuity depends on the type of applications an end user is running. Most data applications are not “real time” in nature. For these applications, service continuity can be defined as “maintaining the session as the user moves.”

This can be attained in a number of ways. Lucent Technologies have opted to use "Mobile IP", an IETF protocol defined in RFC 3344. This protocol enables mobility at the network layer, and allows a mobile unit, such as an organizer, laptop or PDA, to change its location without the need to restart applications or terminate and re-establish an ongoing connection.

Seamless Roaming Technology

In order to use the Mobile IP protocol, end users must install an appropriate mobile “client” onto the data devices they will use. Then they log into this software using their chosen passwords. After activating a session, these end users will be able to use the same data transfer applications as are used over the Internet, such as Web browsing, FTP and streaming media sessions.

The users will also be able to move from one area of coverage to another, or to move to a different system, with no noticeable change in service other than a slight momentary pause. For example, a user could walk outside the company headquarters leaving the coverage of the Corporate WLAN, continue the session on the mobile network as he drives across an area with 3G coverage, then resume on WLAN as a stop off at a place offering WiFi access. And all this without interrupting the data session.

Figure 3 above shows Wireless LAN/3G mobile integration using Mobile IP with additional network elements in the mobile service provider’s network. 3G architectures remain unchanged by the introduction of WiFi/WLAN access. When using Mobile IP, the integrated WLAN solution includes two additional elements in mobile operator’s network: a Mobile IP Home Agent and a Foreign Agent.
  • Mobile IP Home Agent (HA): A router located on a mobile node’s home network (in this case the network operator). It maintains the current location for each departed mobile node (either in the WLAN/WiFi hotspot or the 3G network) and tunnels the received packets for the mobile node to its current location.

  • Mobile IP Foreign Agent (FA): This router is located on a mobile node’s visited network (either WLAN/WiFi hotspot or 3G network). It supports the registration of the mobile node, upon attachment, and enables the mobility process by co-operating with the Home Agent. The Foreign Agent extracts packets intended for the mobile node by de-capsulation of the packets tunneled from the Home Agent; packets sent by the mobile node are transmitted via the Foreign Agent.

Now, the Home Agent is responsible for intercepting all IP packets sent to the mobile node at its home network address. When an IP packet sent to the relevant mobile station arrives from the Internet, the Home Agent encapsulates it in an IP packet and directs it to the care-of address of the Foreign Agent. The Foreign Agent receives the tunneled packet, removes the encapsulation, matches the packet’s destination address to that of the registered mobile node and forwards it to its final destination through the appropriate interface.

Any packets to be transmitted by the mobile node are normally directed to the destination through the use of standard IP routing procedures, and they do not have to involve the Home Agent. However, the Mobile IP protocol also includes the possibility of tunneling these packets all the way to the Home Agent, also known as reverse tunneling. This method of operation can be valuable when the network is not able to route those packets; for example, because they use some sort of source IP address processing or because the home network uses a private addressing scheme that is not routable through the public network.

There can be short delays at hand-over time with this complicated routing of datapackats. But this technology (with or without Foreign Agents) ensures that no packets get lost, and that corrupted datapackets still get resent, should they get corrupted along the way. That assurance is part of the TCP-protocol, and that protocol supervises the traffic throughout all these networks.

In Lucent’s solution, there are two types of Foreign Agents:
  • A Foreign Agent in each WLAN hot spot with support for 802 protocols.

  • Foreign Agent functionality in the core of the 3G mobile network.
The use of a Foreign Agent is, however, optional. The Mobile IP protocol include a mode of operation which is not included in the Internet-version of IP (not even in IP version 6). It is known as “co-located IP address”, and it enables the Mobile IP client software to perform the equivalent role of Foreign Agent. This mode of operation is adopted when it is not possible to have Foreign Agent functionality, such as an existing PDSN/GGSN or a roaming WLAN network that doesn’t support it.

Overview of Mobile IP operation

The mobile unit (laptop computer, PDA or other end-user device) uses the Agent Solicitation message to request the address and other information of the Foreign Agent providing Mobile IP services. The Foreign Agent replies by transmitting an Agent Advertisement. The mobile node obtains its care-of address on the visited network through the Agent Advertisement—and registers this address with its Home Agent in its home network via the Foreign Agent. Care-of address is associated with a foreign network visited by the mobile node, and it changes as the node moves from one foreign network to another.

The Foreign Agent then informs the Home Agent of the current node location by means of the care-of-address. It also serves as the default router for the mobile node when it is connected to this foreign network. The mobile node changes its location and points-of-connection to the network without interruption to the higher layer applications, while maintaining its permanent IP address.


Last Updated: 2007-01-02
Author: Ove Johnsson