<<

. 62
( 132 .)



>>

For the secondary CCPCH (S-CCPCH), it is offset from the P-CCPCH by a multiple
of 256 chips. This carries the FACH and PCH channels. The SF and code used for
this channel are variable, and the UE will receive the con¬guration information for this
channel on the broadcast channel. For user paging, WCDMA uses a second channel known
as the paging indication channel (PICH). This channel is offset from the S-CCPCH by
„PICH = 7680 chips. Figure 6.25 shows the relative timings, and the arrow indicates the
S-CCPCH frame associated with the PICH frame.
The PICH is a ¬xed rate SF = 256 channel of 30 kbps. This means that each radio
frame contains 300 bits. The ¬rst 288 bits of these are paging indicators, and transmission
is switched off for the last 12 bits. These 288 bits contain N paging indicators representing
different paging groups, where N can be 18, 36, 72 or 144. This means, for example, if
there are 72 paging indicators, then each indicator is 4 bits (Figure 6.26).
When a UE connects to the network, it can calculate which paging group it belongs to,
based on its international mobile subscriber identity (IMSI) number. It will then monitor
the PICH, and if its paging group indicators are 1, it will decode the associated frame of
the S-CCPCH for speci¬c paging information. Since the UE knows the period when it
needs to check the PICH, it can use discontinuous reception (DRX) in the interim, thus
saving battery power.
The remaining common physical channels are discussed in Section 6.9.
6.8 WCDMA PHYSICAL LAYER 297


10 ms 10 ms 10 ms



CPICH



P-CCPCH P-CCPCH



PICH PICH PICH



S-CCPCH S-CCPCH S-CCPCH
„PICH




Figure 6.25 Common physical channel timings


10 ms radio frame
288 bits 12 bits

b0 b1 b 2 b3 b286 b287
PICH Tx off


one paging group
if N=72


Figure 6.26 PICH structure


6.8.7 Dedicated physical channels

Dedicated channels (DCH) are transported at the physical layer using the dedicated phys-
ical data channel (DPDCH). This physical layer channel contains one or more transport
channels, as formed by a CCTrCH. This physical layer channel has a control channel
associated with it, the dedicated physical control channel (DPCCH), which contains rel-
evant information with regard to the physical transmission of data. The structure of both
is shown in Figure 6.27.
In the uplink these two channels (DPDCH and DPCCH) are separated using different
channelization codes. They are, however, in phase/quadrative (I/Q) multiplexed together.
There may be a number of DPDCHs on each radio link associated with one DPCCH. An
example of this would be where higher data rates are achieved using multiple codes, and
hence multiple DPDCHs. The DPDCH can have a variety of SFs associated with it and
this is determined by the actual data rate required by the subscriber in each particular
298 UNIVERSAL MOBILE TELECOMMUNICATIONS SYSTEM


10 ms
Uplink
Frame Frame Frame
N N+1 N+x


Slot Slot Slot Slot Slot Slot Slot Slot Slot Slot Slot Slot Slot Slot Slot
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
15 slots per frame (1.5kHz)




DPDCH

DPCCH

Pilot TFCI FBI TPC
2560 chips every slot, therefore 3,840,000 per second

Figure 6.27 Uplink radio frame format

connection. However, in the uplink, the DPCCH always uses the lowest data rate, SF =
256. This equates to a channel symbol rate of 15 ksps, or 10 bits per slot.
The DPCCH is used to carry control information generated at layer 1. This control
information consists of a number of different ¬elds, some of which may or may not be
present. The number of chips for each of these ¬elds is not ¬xed and can vary. There are
a number of de¬ned slot formats, and the one being used is con¬gured by higher layers.
For example, slot format 0 is de¬ned as shown in Figure 6.28.
The ¬elds are de¬ned as follows:

• The pilot bits are prede¬ned, and are therefore known by both the UE and the BTS.
These bits are used for channel estimation in a similar manner to the use of a training
sequence in a GSM burst.
• The transmit power control (TPC) is used to request increased or decreased power on
the downlink channels. This is actually only one bit, where 0 indicates that a power
down is required, and 1 a power up. If there is more than one bit allocated for TPC
in the DPCCH, then the bits are repeated. For the example con¬guration of slot 0, a
power up command would be represented as TPC = 11. Repetition is required since
the air interface is notoriously bad and the TPC is required to be received successfully
while keeping the power level as low as possible.

6 bits 2 bits 2 bits


Pilot TFCI TPC

Slot Format 0

Figure 6.28 DPCCH slot format 0
6.8 WCDMA PHYSICAL LAYER 299


• The transport format combination indicator (TFCI) is used to inform the BTS of the
format of the speci¬c transport channels mapped into the current DPDCH. Transport
formats are de¬ned by higher layers during radio link establishment. The TFCI ¬eld is
not required for ¬xed-rate services, and is only used for variable-rate services where
there are a number of possible rates.
• The feedback indicator (FBI) bits are used to support diversity techniques in the BTS
which require feedback from the mobile device. These are explained in Section 6.11.

The coding of the TFCI bits into slots is performed as follows. One TFCI is needed to
describe the format of the DPDCH in each radio frame. If it is less than 10 bits long, the
TFCI is padded out to 10 bits long by adding zeroes at the most signi¬cant bit (MSB), and
then coded using a Reed“Muller code to provide error protection. This generates a 32-bit
coded TFCI. In the above example of slot format 0, the ¬rst 30 bits of this are transferred
across each of the 15 slots of the radio frame, starting from the lowest index bit. This
means slot 0 will contain bits 0 and 1 of the coded TFCI, etc. The most signi¬cant two
bits are ignored.
In the downlink the two channels (DPDCH and DPCCH) are combined using TDM
and transmitted under the same channelization code. This combined channel is referred
to as the downlink dedicated physical channel (downlink DPCH) (see Figure 6.29). As
with the uplink, the DPCCH is used to carry control information generated at layer 1.
The control information consists of the same ¬elds as the uplink with the exception of
the FBI ¬eld. This is not required in the downlink since transmit diversity is not used on
the mobile device. In the uplink, the DPDCH and DPCCH were I/Q multiplexed together.
However in the downlink instead, the single downlink DPCH goes through a serial to
parallel conversion and the odd and even bits are then I/Q multiplexed. This results in a
doubling of the channel bit rate.
Once again, the format of the slot to be used is de¬ned by higher-layer signalling at
channel establishment. An example of slot format 9 is shown in Figure 6.30. For this

10 ms
Downlink
Frame Frame Frame
N N+1 N+x


Slot Slot Slot Slot Slot Slot Slot Slot Slot Slot Slot Slot Slot Slot Slot
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
15 slots per frame (1.5kHz)




Data part 1 TPC TFCI Data part 2 Pilot

DPDCH DPCCH DPDCH DPCCH

2560 chips every slot, therefore 3,840,000 per second

Figure 6.29 Downlink radio frame format
300 UNIVERSAL MOBILE TELECOMMUNICATIONS SYSTEM


6 bits 2 2 26 bits 4




TFCI
TPC
Data 1 Data 2 Pilot

Slot Format 9

Figure 6.30 Downlink DDPCH slot format 9


format, SF = 128 is used, which provides a symbol rate of 30 ksps. However, because of
the I/Q modulation used on the downlink, this equates to a channel bit rate of 60 kbps.
Therefore there are 40 bits available per slot.
The reasoning behind time multiplexing the DPDCH and DPCCH in the downlink and
I/Q multiplexing in the uplink is to avoid uplink interference. If the two uplink channels
were time multiplexed together, then there would be situations where there was no data
transfer, but the control channel still contained power control information. In this case,
since power control is performed in each slot, this would generate a 1.5 kHz signal,
which is located in the centre of the telephony band and causes unacceptable interference.
Therefore the I/Q multiplexing method is employed. This 1.5 kHz signal does not present
a problem in the downlink since there are many channels transmitting simultaneously.



6.9 INITIAL CONNECTION TO NETWORK

After the user switches on their mobile device, they must establish a connection to the
network, principally to perform a location update to advise the network of their existence
and location. Before a user can contact the core network to do this, the UE must establish
a signalling connection to the RNC as described in Section 6.16. However, prior to that
there are a number of steps that the device must go through, as summarized in Figure 6.31.



6.9.1 Synchronization procedures
As mentioned previously, within the WCDMA cell each user shares the same wide fre-
quency band of 5 MHz. The frequency reuse within adjacent cells is 1, which means
that the adjacent cells may also be using the same frequency range provided that they are
within the same system, i.e. belonging to the same mobile operator. When a user switches
on their mobile device it needs to connect to one of the base stations in the coverage
area. Since according to the ITU-T frequency allocation for 3G, the maximum available
bandwidth is 60 MHz, that means that there are at most 12 channels in use in a given
country. Therefore the frequency search procedures are fairly trivial.
To ascertain the necessary parameters for connection to the network, the UE must
access the BCCH, where all of this information is transmitted. Before the station can
decode information on the BCCH, it must ¬rst synchronize with the system. It does this
with the aid of the synchronization channel (SCH). The SCH is actually made up of two
6.9 INITIAL CONNECTION TO NETWORK 301



UE BTS RNC CN


Switch on
SCH-1/SCH-2
perform synchronisation

P-CCPCH
decode broadcast info

PRACH
initial access attempt

AICH
successful access
CCCH/RACH/PRACH
RRC request to establish signalling link

RACH/FACH/DCH
RRC connection procedures

Direct Transfer Messages
L3 Location Update


Figure 6.31 Initial network connection

separate channels known as the primary SCH and secondary SCH. Connecting to the BTS
is known as ˜camping on the cell™. The BCCH is carried in the P-CCPCH, which has a
universally ¬xed channelization code and is always coded using the primary scrambling
code for the cell. Therefore to ˜camp on a cell™ the UE has to initially determine the
downlink scrambling code for that cell, and it also has to synchronize with the frame

<<

. 62
( 132 .)



>>