maki/openairinterface5g
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(ldpc_aal): Documentation

Browse Source 
The tuning of the BLER is necessary for Intel ACC, not for the T2.
This commit is contained in:
Romain Beurdouche
2025-09-04 08:31:10 +00:00
parent 6647af4116
commit 79c1133736
1 changed files with 4 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
doc/LDPC_OFFLOAD_SETUP.md
+4 -2
View File
@@ -301,10 +301,12 @@ sudo ./nr-softmodem -O ~/gnb.conf --loader.ldpc.shlibversion _aal --nrLDPC_codin
## Potential Low Throughput ## Potential Low Throughput
The current implementation has been tested to work in an end-to-end setup and is functional. The current implementation has been tested to work in an end-to-end setup and is functional.
However, there are still opportunities for optimization, particularly in LDPC decoding performance, which is an area of ongoing improvement. However, depending on the accelerator in use,
there are still opportunities for optimization, particularly in LDPC decoding performance, which is an area of ongoing improvement.
As such, downlink/uplink throughput may be suboptimal with the default configurations, but enhancements are actively being explored. As such, downlink/uplink throughput may be suboptimal with the default configurations, but enhancements are actively being explored.
To achieve better E2E performance with the current implementation, we recommend the following adjustments: For example, to achieve better E2E performance with the current implementation with Intel ACC 100 and 200 (vRAN Boost),
we recommend the following adjustments:
1. Increasing the number of LDPC decoding iterations of the L1, e.g., `max_ldpc_iterations` to 200. 1. Increasing the number of LDPC decoding iterations of the L1, e.g., `max_ldpc_iterations` to 200.
2. Increasing the BLER targets of the MAC scheduler. 2. Increasing the BLER targets of the MAC scheduler.