Content for TR 38.835 Word version: 18.0.1
1…
4…
5…
6
A
B…
B.1.2
B.1.3
B.1.4
B.1.5
B.1.6
B.1.7
B.1.8
B.1.9
B.1.10
B.2…
B.2.2
B.2.3
B.2.4
B.2.5
B.2.6
B.2.7
B.2.8
B.2.9
B.2.10
B.2.11
B.2.12
B.2.13
B.2.14
B.2.15
B.2.16
B.2.17
B.2.18
C…
B.1.3 Enhanced CQI for CBG-based transmissions p. 23
This clause captures the capacity performance evaluation results of enhanced CQI (eCQI) for code block group (CBG)-based transmissions. The use of CBG-based transmissions allows avoiding a re-transmission of the full transport block in case the receiver fails to correctly decode the full TB. A large size of one video frame as per TR 38.838, results in a large TB(s) size to convey a video frame.
The performance of the current legacy link adaptation with TB-based transmission legacy CQI reporting (scheme 3.1 in Table B.1.3-1) has been compared against CBG-based transmission with eCQI (scheme 3.2 in Table B.1.3-1). The legacy CQI corresponds to the highest supported MCS while not exceeding a 10% BLER target for the TBs.
Scheme 3.2:
eCQI is the scheme where UE estimates the highest supported MCS (expressed via a CQI index), assuming that downlink transmissions occupy a set of downlink physical resource blocks termed the CSI reference resource with M code block groups, while the error probability of at most N failed code block groups does not exceed P. Parameters M, N, and P may be configured by the network, or fixed to values that are attractive for XR services. The configuration of the UE to use eCQI may be conducted with RRC signaling (note that configuration of current CQI schemes for a UE to use is also via RRC). The reporting of the eCQI can be in the form of an eCQI index (from current CQI tables [TS 38.214 - Tables in 5.2.2.1] or enhanced CQI tables) that will guide the gNB to choose the supported modulation scheme, effective code rate, and overall efficiency that for its PDSCH transmissions. For the CBG-based transmission with eCQI, results are shown for parameter settings M=8 assuming either N=2 or N=4 and P= 50%.
The performance results are reported in Table B.1.3-1 in terms of the ratio of satisfied users.
| Source | Tdoc Source | Scheme | TDD format | SU/MU-MIMO | Data rate (Mbps) | PDB (ms) | Capacity (UEs/cell) | C1=floor (Capacity) | % of satisfied UEs when #UEs/cell = C1 | Notes |
|---|---|---|---|---|---|---|---|---|---|---|
| Source [Nokia] | R1-2209536 | 3.2* | DDDSU | SU-MIMO | 30 | 10 | 6.35 | 6 | 99% | Note 1,2 |
| 15 | 7.31 | 7 | 99% | |||||||
| Source [Nokia] | R1-2209536 | 3.2* | DDDSU | SU-MIMO | 45 | 10 | 4.15 | 4 | 95% | Note 1,2 |
| 15 | 5.12 | 5 | 95% | |||||||
| Source [Nokia] | R1-2209536 | 3.2** | DDDSU | SU-MIMO | 30 | 10 | 6.22 | 6 | 97% | Note 1,2 |
| 15 | 7.31 | 7 | 99% | |||||||
| Source [Nokia] | R1-2209536 | 3.2** | DDDSU | SU-MIMO | 45 | 10 | 4.10 | 4 | 93% | Note 1,2 |
| 15 | 4.83 | 4 | 100% | |||||||
| Source [Nokia] | R1-2209536 | 3.1 | DDDSU | SU-MIMO | 30 | 10 | 5.45 | 5 | 95% | Note 1 |
| 15 | 6.31 | 6 | 98% | |||||||
| Source [Nokia] | R1-2209536 | 3.1 | DDDSU | SU-MIMO | 45 | 10 | 3.35 | 3 | 99% | Note 1 |
| 15 | 4.18 | 4 | 94% | |||||||
|
NOTE 1:
BS antenna parameters: 32TxRUs, (M, N, P, Mg, Ng; Mp, Np) = (4,4,2,1,1,4,4)
NOTE 2:
non-iid CBG errors in a TB are assumed
*
At most N failed code block groups, N= 4
**
At most N failed code block groups, N= 2
|
||||||||||
Based on the evaluation results in Table B.1.3-1, the following observations can be made:
- For FR1, InH, DL, with 100MHz bandwidth for VR/AR single-stream traffic model, 30Mbps, 10ms PDB, 60 FPS, with SU-MIMO and 32TxRU, it is observed from Source [Nokia] that the capacity is increased from 5.45 UEs per cell with legacy CQI to 6.35 UEs per cell with eCQI, where the error probability of at most N=4 failed code block groups out of 8 CBGs does not exceed P = 50% (capacity gain is 17%). For N=2, the results show similar trend.
- For FR1, InH, DL, with 100MHz bandwidth for Cloud Gaming single-stream traffic model, 30Mbps, 15ms PDB, 60 FPS, with SU-MIMO and 32TxRU, it is observed from Source [Nokia] that the capacity is increased from 6.31 UEs per cell with legacy CQI to 7.31 users per cell with eCQI, where the error probability of at most N=4 failed code block groups out of 8 CBGs does not exceed P = 50% (capacity gain is 16%). For N=2, the results show similar trend.
- For FR1, InH, DL, with 100MHz bandwidth for VR/AR single-stream traffic model, 45Mbps, 10ms PDB, 60 FPS, with SU-MIMO and 32TxRU, it is observed from Source [Nokia] that the capacity is increased from 3.35 UEs per cell with legacy CQI to 4.15 UEs per cell with eCQI, where the error probability of at most N=4 failed code block groups out of 8 CBGs does not exceed P = 50% (capacity gain is 24%). For N=2, the results show similar trend.
- For FR1, InH, DL, with 100MHz bandwidth for Cloud Gaming single-stream traffic model, 45Mbps, 15ms PDB, 60 FPS, with SU-MIMO and 32TxRU, it is observed from Source [Nokia] that the capacity is increased from 4.18 UEs per cell with legacy CQ to 5.12 UEs per cell with eCQI, where the error probability of at most N=4 failed code block groups out of 8 CBGs does not exceed P = 50% (capacity gain is 22%). For N=2, the results show similar trend.
