<<

. 86
( 132 .)



>>

RLC
A2SIG A2SIG


Figure 6.137 Circuit call release
6.22 UMTS CALL LIFE CYCLE 413


6.22.7 Packet core connection
After the call, the user decides to check his email and must establish a packet core
connection. Once again, the higher-layer signalling is the same as for GPRS, as described
in Chapter 4. Another observation that should be made is that the mechanisms for estab-
lishing radio bearers are much the same as for a circuit call; only the QoS pro¬le of the
connection that is established differs.

1. Once again, the signalling connection is established (Figure 6.138).
2. The initial message from the UE is the GPRS attach request.
3. Once the authentication and security procedures have successfully completed, the
SGSN will reply with a GPRS attach accept.
4. The common ID update is performed.
5. To transfer data, the UE must establish a PDP context. This is done through the
activate PDP context request. There will also be a create PDP context/create PDP
context response between the SGSN and the GGSN, during which the UE is
allocated an IP address which is then bound to its IMSI.

UE BTS SRNC 3G SGSN

Signalling Connection Establishment 1
Initial Direct Transfer
RRC RRC
GPRS Attach Request 2
initial UE message
RANAP RANAP
GPRS Attach Request
Direct Transfer
RANAP RANAP
GPRS Attach Accept 3
Downlink Direct Transfer
RRC RRC
GPRS Attach Accept
Common ID Update
RANAP 4
RANAP
Uplink Direct Transfer
RRC RRC
Activate PDP Context Req 5
Direct Transfer
RANAP RANAP
Activate PDP Context Req
RAB Assignment Req
RANAP RANAP 6
RL Reconfig Request
NBAP NBAP
7
RL Reconfig Response
NBAP NBAP
ERQ
A2SIG A2SIG
8
ECF
A2SIG A2SIG
Radio Bearer Setup
RRC RRC
9
Radio Bearer Setup Complete
RRC RRC
RAB Assignment
RANAP RANAP 10
Response
Direct Transfer
RANAP RANAP
11
Downlink Direct Transfer Activate PDP Context Accept
RRC RRC
Activate PDP Context Accept
Packet Connection Established 12



Figure 6.138 Packet connection establishment
414 UNIVERSAL MOBILE TELECOMMUNICATIONS SYSTEM


6. The SGSN will now use the RAB assignment request to set up the bearer. Typically,
one RAB is established, and the QoS of it at this point is dictated by the service
pro¬le for the user. For example, the user may have subscribed for a background
service with a maximum data rate of 64 kbps. Notice that there is no AAL2
signalling over the Iu interface following since this is an IP connection. Instead, the
RAB assignment request will contain the IP address of the SGSN and the
GTP TEID.
7. Once again, the radio link is recon¬gured for this data transfer. Notice that since
this is data, time is not critical, so a radio link recon¬guration request/response is
used, instead of the synchronized radio link recon¬guration prepare which was used
for the circuit call.
8. An AAL2 transport bearer is also established.
9. Now, a new data radio bearer is established between the UE and the SRNC with the
selected QoS pro¬le. The QoS pro¬le here can be downgraded by the SRNC from
that requested by the SGSN, due perhaps to resource limitations within UTRAN.
10. Once completed, the SRNC replies to the SGSN with the RAB assignment response.
This will contain the RNC IP address and the GTP TEID to complete the
information about the other end of the tunnel.
11. Now the SRNC may inform the UE that the connection is ready with the activate
PDP context accept.
12. The packet connection is now established, and the user is free to send and receive
data through their permitted access points.

The nature of this packet connection is also different from the circuit side. With a phone
call, there is a very clearly de¬ned start and ¬nish. However, with packet data, there may
be long idle periods where the UE is ˜connected™ to the network, but is not transferring
any data. This is the ˜always connected™ aspect of GPRS. What it actually means is that
the SGSN holds the PDP context of the UE. In the above case, consider that the user
has ¬nished reading the email but remains connected to the network. After some time
period of no data transfer, the SRNC will most likely request that the Iu connection be
released using the Iu release request message, stating the reason as user inactivity. Once
the UE sends or receives data, the UTRAN connection can be quickly re-established. For
the packet core, a paging message indicates downlink packet data for the user.
Figure 6.139 shows a trace ¬le captured on the Iu-PS interface, where R1 and R2
represent the RNC and SGSN, respectively. Here the GPRS attach and PDP context
activation procedures are shown, followed by IP data transfer. The last exchange is a
˜ping™ on the interface.



6.23 CDMA2000

The original work for this system started out as TIA TR45.5 and has been further
developed by 3GPP2 as an integrated part of the IMT2000 speci¬cation suite under
6.23 CDMA2000 415




GPRS attach



PDP context activation




IP traffic




IP ping
echo/reply




Figure 6.139 Sample Iu-PS trace packet connection establishment. Reproduced by per-
mission of NetHawk Oyj


the IMT-MC (multicarrier). The CDMA2000 system leverages on the 2G IS-95 (com-
monly known as cdmaOne) system and does not require a major overhaul of the existing
system. CDMA2000 is backward compatible with IS-95 and as such allows operators to
upgrade the 2G system to 3G in stages. Upgrades are required to the BTS and BSC; as
well as this a packet data server network (PDSN) is also required. Because of the back-
ward compatibility, existing IS-95 subscribers, will be able to use their mobile devices
on the new network and CDMA2000 devices will work on IS-95 networks, and handover
between these will be seamless. Services such as those required for voice, data, SMS as
well as over-the-air provisioning and activation are also supported by CDMA2000. There
are a number of variations to the CDMA2000 system and these will be described in the
following sections. CDMA2000 is designed to work with two spreading rates referred to
as rates 1 and 3: rate 1 is 1.2288 Mcps and rate 3 is 3.6864 Mcps.



6.23.1 History of cellular in the USA
The original ¬rst generation (1G) telecommunication system established on the cellular
concept was the advanced mobile phone system (AMPS) which was developed by Bell
labs in 1947. This system was based on an analogue rather than a digital technology and
was adopted by many countries throughout the world. In the early 1980s the AT&T/Bell
416 UNIVERSAL MOBILE TELECOMMUNICATIONS SYSTEM


monopoly was broken up under the modi¬cation of ¬nal judgement (MFJ) and the AMPS
technology was given to a number of regional Bell operating companies (RBOCs). To
introduce competition, the frequency bands available in the US were separated into A and
B frequency bands. The A bands were licensed to nonwire operators, i.e. new cellular
operators who did not have any association with the local ¬xed-line carriers, and the
B band was licensed to the wireline companies, which essentially comprised the local
RBOCs. These frequencies were licensed for operation with both analogue and digital
systems. In the early 1990s the US government auctioned six new nationwide licences,
which were referred to as the Personal Communication Service (PCS) A“F band carriers
(Figure 6.140) and resided within the 1900 MHz band. The PCS bands were licensed to
be used by digital systems only such as TDMA, CDMA and GSM.
It can be seen from this diagram that these frequencies overlap with those that
were set aside by the ITU for IMT2000 (1920“1980 MHz and 2110“2170 MHz). The
spectral deployment is primarily within the cellular and PCS bands of North America:
824“849 MHz in the reverse link, 869“894 MHz in the forward link (the same as for
South Korea) and 1850“1910 MHz reverse link and 1930“1990 MHz for the forward
link. These are almost the same frequencies that are allocated in South Korea and Japan.
There are two digital access systems that have wide-scale availability in the US. These
are TDMA and CDMA. Both of these systems work with the ANSI-41 (formerly IS-41)
core network, which is also implemented in the AMPS system. However, the market there
is no longer so clear cut as GSM is also now making inroads into the US in the 1900 MHz
band. As discussed in Chapter 2, digital systems have a number of advantages over
analogue systems and thus many of the AMPS networks in the US have been replaced by
TDMA, CDMA or GSM systems. It should be noted that whereas GSM/UMTS refer to the
uplink and downlink, in CDMA these are termed reverse and forward link, respectively.
This situation in the US may be set to change. At the time of writing, the US regu-
lator, the Federal Communications Commission (FCC), has made a radical reform of its
spectrum allocation policy, moving to a market system, where spectrum can be refarmed.
This has been prompted by the perceived failure of the traditional approach of regulated
designation, where some allocated frequencies are severely underutilized, but cannot be
used by others. Other related developments are the increase in availability of unlicensed
technologies such as WiFi, and the potential of software de¬ned radio (SDR), where the
air interface and frequency used is merely an application at the transceiver. This would
allow the software to dynamically utilize frequency not currently occupied. The FCC has



MSS
IMT-2000


PCS Band
F

A D B EF C A D B EF C



frequency (MHz)
1850 1900 1950 2000

Figure 6.140 US PCS bands
6.23 CDMA2000 417


de¬ned interference levels as a real-time gauge of frequency usage, noting that interference
rejection is already a key component of digital wireless communication systems.


6.23.2 The TDMA system
The TDMA system is a digital system which was initially introduced to work within the
same cellular spectrum as the analogue AMPS system but to give increased ef¬ciency.
The ¬rst of these was the IS-54 standard, which was a dual-mode system working with
the AMPS attributes but allowing user data to be sent over digital traf¬c channels. The
introduction of digital signal processor (DSP) techniques made it three times as ef¬cient
as the old AMPS system since three subscribers could share the same 30 kHz frequency
band simultaneously. Digital control channels were added in the IS-136 standard and this
gave further ef¬ciencies, extending mobile device battery life through the sleep facility
and also introducing SMS. In October 1996 a revised version of IS-136 was published
for use in the 1900 MHz PCS band.


6.23.3 The CDMA system
The original CDMA system, which is referred to as cdmaOne or IS-95, comprises a
1.25 MHz channel and uses a chip rate of 1.2288 Mcps. As well as voice, IS-95A will
support data up to 9.6/14.4 kbps; for packet data transfer, IS-95B requires the addition of
an Interworking Unit and can support data in excess of 64 kbps. It can be seen as a 2.5G
solution similar to GPRS.


6.23.4 Evolution path
The path to higher speed data via the CDMA2000 route consists of three evolutionary
steps, as illustrated in Figure 6.141.


IS-41 Core Network


CDMA2000
IS-95A IS-95B
1XEV-DV

<<

. 86
( 132 .)



>>