4.8 Coding for E‑DCH
25.2223GPPMultiplexing and channel coding (TDD)Release 17TS
Figure 22 shows the processing structure for the E‑DCH transport channel mapped onto a separate CCTrCH. Data arrives to the coding unit in form of a maximum of one transport block once every transmission time interval (TTI). A 10ms TTI is used for E-DCH for 3.84Mcps and 7.68Mcps TDD whilst for 1.28Mcps TDD, a TTI of 5ms will be used.
For 1.28Mcps TDD multi-carrier E-DCH transmission, a number of transport blocks may arrive at the coding unit in one TTI, where the number of the transport blocks equals to the number of the scheduled E-DCH carriers. Each transport block for each scheduled E-DCH carrier shall be coded separately. The following coding steps for E-DCH on one carrier can be identified:
– add CRC to each transport block (see subclause 4.8.1);
– code block segmentation (see subclause 4.8.2);
– channel coding (see subclause 4.8.3);
– hybrid ARQ (see subclause 4.8.4);
– bit scrambling (see subclause 4.8.5);
– interleaving for E-DCH (see subclause 4.8.6);
– constellation re-arrangement for 16QAM (see subclause 4.8.7);
– mapping to physical channels (see subclause 4.8.8).
Figure 22. Coding chain for E-DCH
In the following the number of transport blocks is always one. When referencing non E-DCH formulae which are used in correspondence with E-DCH formulae the convention is used that transport block subscripts may be omitted (e.g. Xi when i is always 1 may be written X).
4.8.1 CRC attachment for E-DCH
CRC attachment for the E-DCH transport channel shall be performed according to the general method described in 4.2.1 above with the following specific parameters.
The CRC length shall always be L1=24 bits.
4.8.2 Code block segmentation for E-DCH
Code block segmentation for the E-DCH transport channel shall be performed according to the general method described in 4.2.2.2 with the following specific parameters.
There is a maximum of one transport block. The bits input to the block are mapped to the bits directly. It follows that Xi = Bi. Note that the bits x referenced here refer only to the internals of the code block segmentation function. The output bits from the code block segmentation function are oir1, oir2, oir3, …, oirK.
The value of Z = 5114 for turbo coding shall be used.
4.8.3 Channel coding for E-DCH
Channel coding for the E-DCH transport channel shall be performed according to the general method described in section 4.2.3 above with the following specific parameters.
There is a maximum of one transport block, i=1. The rate 1/3 turbo coding shall be used.
4.8.4 Physical layer HARQ functionality and rate matching for E-DCH
The hybrid ARQ functionality matches the number of bits at the output of the channel coder to the total number of bits of the E-PUCH set to which the E-DCH transport channel is mapped. The hybrid ARQ functionality is controlled by the redundancy version (RV) parameters.
Figure 23: E‑DCH hybrid ARQ functionality
4.8.4.1 Determination of SF, modulation and number of physical channels
The SF, modulation type and number of E-PUCHs in the E-PUCH set is determined by higher layers (see [15]). These correspond to a value of Ne,data,j.
4.8.4.2 HARQ bit separation
The HARQ bit separation function shall be performed in the same way as bit separation for turbo encoded TrCHs with puncturing in 4.2.7.2.1 above.
4.8.4.3 HARQ Rate Matching Stage
The hybrid ARQ rate matching for the E-DCH transport channel shall be done with the general method described in 4.2.7.3 with the following specific parameters.
The parameters of the rate matching stage depend on the value of the RV parameters s and r. The s and r combinations corresponding to each RV allowed for the E-DCH are listed in table 22 below.
Table 22: RV for E-DCH
|
E-DCH RV Index |
s |
r |
|
0 |
1 |
0 |
|
1 |
0 |
0 |
|
2 |
1 |
1 |
|
3 |
0 |
1 |
The parameter eplus, eminus and eini are calculated with the general method described in 4.5.4.3 above. The following parameters are used as input:
Nsys = Np1 = Np2 = Ne,j/3
Ndata = Ne,data,j
rmax = 2 (for both QPSK and 16-QAM)
4.8.4.4 HARQ bit collection
HARQ bit collection for E-DCH shall be performed according to the general method described for HS-DSCH in subclause 4.5.4.4.
4.8.5 Bit scrambling
The bit scrambling for E-DCH shall be performed in accordance with the general method described in subclause 4.2.9.
4.8.6 Interleaving for E-DCH
Interlevaing for E-DCH shall be performed in accordance with the general method described for HS-DSCH in subclause 4.5.6.
4.8.7 Constellation re-arrangement for 16 QAM
Constellation rearrangement shall be performed in the case of 16-QAM in accordance with the general method described for HS-DSCH in subclause 4.5.7. For QPSK this function is transparent.
For 3.84Mcps and 7.68Mcps, the constellation version parameter b is associated with the E-DCH RV index as shown in table 23 below.
Table 23: Mapping of RV to constellation rearrangement parameter b for E-DCH (3.84Mcps and 7.68Mcps options)
|
E-DCH RV Index |
b |
|
0 |
0 |
|
1 |
1 |
|
2 |
2 |
|
3 |
3 |
For 1.28Mcps option, the constellation version parameter b is associated with the retransmission sequence number (RSN). The mapping between RSN and b parameters for constellation re-arrangement is listed in table 25A in subclause 4.9.2.1.2.
4.8.8 Physical channel mapping for E-DCH
The E-PUCH is defined in [7]. The bits input to the physical channel mapping are denoted by r1, r2, …, rR, where R= Ne,data,j and is the number of physical channel data bits to be transmitted in the current TTI. These bits are mapped to the physical channel bits, {wt,k : t = 1, 2, …, T; and k = 1, 2, …, Ut}, where t is the timeslot index, T is the number of timeslots in the allocation message, k is the physical channel bit index and Ut is the number of bits in the E-PUCH physical channel in timeslot t. The timeslot index, t, increases with increasing timeslot number and the physical channel bit index, k, increases with increasing physical channel bit position in time.
The bits r1, r2, …, rR shall be mapped to the physical channel bits wt,k according to the following rule :
for k = 1, 2, …, U1
for k = 1, 2, …, U2
…
for k = 1, 2, …, UT