Interworking System Architectures

By connecting the IEEE 802.11 WLAN to a UMTS CN as a complementary
radio access network, a second form of mobile packet data services is
provided by this heterogeneous IP-based system. Figure 2.12 represents the five interconnection points between WLAN
and UMTS. These interconnection architectures involve minimum changes
to the existing standards and technologies, especially for MAC and PHY
layer to ensure that existing standards and networks continue to function as
before. The first two interconnections in Figure 2.12 will always have interaction
between WLAN access point (AP) and the packet-switched part of the
UMTS CN [8]. This means that the gateway to the IEEE 802.11 WLAN
network is attached to the packet-switched domain. This interconnection is
possible through the 3G serving GPRS supporting node (3G-SGSN) entity
and gateway GPRS supporting node (GGSN) entity, which are the elements of
the UMTS packet switched core network (PS CN). In both cases the WLAN
network appears to be a UMTS cell or routing area, respectively. The UMTS
network will be a master network and the IEEE 802.11 WLAN network will
be a slave network. This means that the mobility management and security
will be handled by UMTS network, and the WLAN network will be seen as
one of its own cells or routing areas. This may require dual mode personal
computer memory card international association (PCMCIA) cards to access
two different physical layers. In addition, all traffic will first reach the UMTS 3G-SGSN or 3G-GGSN before reaching their final destinations, even if the
final destination were to be in the WLAN home network.
In the third interconnection, the virtual access point (VAP) reverses the
roles played by the UMTS and WLAN in the first two architectures. This is
called a tight coupling because there is always interaction between both networks.
Here, the IEEE 802.11 WLAN is a master network, and the UMTS
is the slave network. Mobility management is according to the WLAN and
Interaccess Point Protocol (IAPP) is the protocol that is specified for this
management.
In the fourth interconnection architecture, a mobility gateway/mobile
proxy (MG) is employed in between the UMTS and IEEE 802.11 WLAN
networks. These are both peer-to-peer networks. The MG is a proxy that is
implemented on either the UMTS or the WLAN sides and will handle the
mobility and routing.
The fifth interconnection architecture is based on mobile IP protocol.
This is called no coupling, and both networks are peers. Mobile IP handles
the mobility management. A home agent/foreign agent (HA/FA) entity is
involved with this architecture, whereby the IP layer becomes aware of the
agent advertisements of the mobility agent (HA/FA), and would do a binding
update in certain lifetime [17].