Uplink Spreading, Scrambling, and Modulation

A physical channel is what carries the actual user data or control information
over the air interface. A physical channel can be considered a combination
of frequency, scrambling code, and channelization code, and in the
uplink, as we shall describe later, the relative phase is also significant. For example, if a given user is transmitting user data and control information,
then the user data stream will be carried on one physical channel and the
control information will be on a different physical channel.

A number of different physical channels are used in the uplink, with a
given type of channel selected according to what the user equipment (UE)
is attempting to do—such as simply request access to the network, send
just a single burst of data, or send a stream of data. These channels are
described in further detail later in this chapter. For now, let us focus on the
situation where a user is transmitting a stream of data, which would happen in a voice conversation. In such a situation, the terminal will
normally use at least two physical channels—a Dedicated Physical Data
Channel (DPDCH) and a Dedicated Physical Control Channel (DPCCH).
The DPDCH carries the user data and the DPCCH carries control information.
Depending on the amount of data to be sent, a single user can use
just a single DPDCH, which will support up to 480 Kbps of user data or as
many as six DPDCHs, which will support up to 2.3 Mbps of user data.

A DPDCH can have a variable spreading factor. This simply means that
the user bit rate does not have to be fixed to a specific value. The spreading
factor for a DPDCH can be 4, 8, 16, 32, 64, 128, or 256. These correspond to
DPDCH bit rates of 15 Kbps (3.84
106/256  15
103) and up to 960 Kbps
(3.84
106/4  960
103). Of course, these are not the actual user data
rates, because a significant amount of coding overhead is included in the
DPDCH to support forward error correction. In general, the user data rate
is approximately half (or less) of the DPDCH rate. Thus, for example, a
DPDCH operating at a spreading rate of 4 will carry data at a rate of 960
Kbps. Of this, however, only about 480 Kbps will correspond to usable data.
The rest is consumed by additional coding required for error correction. If a
single user wants to transmit user data at a rate greater than 480 Kbps,
then multiple DPDCHs can be used (up to a maximum of 6).

Figure 6-2 shows how multiple DPDCHs are handled. Also shown is the
DPCCH, which is also sent whenever one or more DPDCHs are sent. The
channelization codes (Cd,1 to Cd,6) represent the channelization codes
applied to each of the six DPDCHs. The channelization code applied to the
DPCCH is represented as Cc. Each of the DPDCHs is spread to the chip rate
by a channelization code. DPDCHs 1, 3, and 5 are channelized and weighted
by a gain factor bd. These DPDCHs are on the so-called I (in-phase) branch.
DPDCHs 2, 4, and 6 plus the DPCCH are on the so-called Q (quadrature)
branch. These are also channelized. These spread DPDCHs are also
weighted by the gain factor bd, whereas the spread DPCCH is weighted by
the gain factor bc. The two gain factors are specified as 4-bit words that represent
steps from 0 to 1. Thus, 0000  off, 0001  1/15, 0010  2/15, and
1111  15/15  1. At any given instant, one of the two gain factors has the
value of 1 (binary 1111  15/15  1).

Mathematically, the spread signals on the Q branch are treated as a
stream of imaginary bits. These are summed with the stream of real bits on
the I branch to provide a stream of complex-valued chips at the chip rate.
This stream of complex-valued chips is then subjected to a complex-valued
scrambling code, which is aligned with the beginning of a radio frame.