nrORANBlockCompress

O-RAN fronthaul block compression

Since R2022b

Syntax

[cGrid,cParam] = nrORANBlockCompress(grid,method,cIQWidth,IQWidth)

[cGrid,cParam] = nrORANBlockCompress(grid,method,cIQWidth)

Description

[cGrid,cParam] = nrORANBlockCompress(grid,method,cIQWidth,IQWidth) performs open radio access network (O-RAN) fronthaul block compression on the user plane (U-Plane) IQ data, grid. The input method specifies the compression method. The function implements the block floating point (BFP), block scaling, and μ-law compression methods, as defined in O-RAN.WG4.CUS TS Annexes A.1.1, A.2.1, and A.3.1, respectively [1]. The function returns the compressed grid, cGrid, and the compression parameter, cParam. Inputs IQWidth and cIQWidth specify the bit width of the IQ samples before and after compression, respectively.

[cGrid,cParam] = nrORANBlockCompress(grid,method,cIQWidth) performs only BFP or μ-law compression.

Examples

Perform O-RAN Fronthaul Block Compression and Decompression

Generate an NR downlink resource grid to perform O-RAN fronthaul block compression.

cfg = nrDLCarrierConfig;
[waveform,info] = nrWaveformGenerator(cfg);
grid = info.ResourceGrids.ResourceGridBWP;

Scale the bit width of the IQ samples in the resource grid to 16 bits.

IQWidth = 16;
peak = max(abs([real(grid(:)); imag(grid(:))]));
scaleFactor = peak / (2^(IQWidth-1)-1);
scaledGrid = round(grid / scaleFactor);

Specify the bit width for the compressed IQ samples.

cIQWidth = 9;

Specify the compression method.

method = 'BFP';

Perform O-RAN fronthaul block compression.

[cGrid,cParam] = nrORANBlockCompress(scaledGrid,method,cIQWidth);

Perform O-RAN fronthaul block decompression.

dGrid = nrORANBlockDecompress(cGrid,cParam,method);

Descale the decompressed grid using the scaling factor from before the compression.

descaledGrid = dGrid * scaleFactor;

Input Arguments

grid — Resource grid of U-Plane IQ data

K-by-L-by-P complex array

Resource grid of U-Plane IQ data, specified as a K-by-L-by-P complex array. K is the number of subcarriers, a multiple of 12, where 12 corresponds to the resource elements (REs) in a physical resource block (PRB). L is the number of OFDM symbols. P is the number of antennas. The input IQWidth specifies the bit width of the IQ samples in the grid.

Data Types: double | single
Complex Number Support: Yes

method — O-RAN fronthaul compression method

'BFP' | 'blockScaling' | 'muLaw'

O-RAN fronthaul compression method, specified as one of these values.

• 'BFP' — Specify this value for BFP compression, as defined in O-RAN.WG4.CUS TS Annex A.1.1.

• 'blockScaling' — Specify this value for block scaling compression, as defined in O-RAN.WG4.CUS TS Annex A.2.1.

• 'muLaw' — Specify this value for μ-law compression, as defined in O-RAN.WG4.CUS TS Annex A.3.1.

Data Types: string | char

cIQWidth — Bit width of compressed IQ samples

integer from 1 to 16 | K/12-by-L-by-P integer array

Bit width of the compressed IQ samples (in the output cGrid), including the sign bit, specified as one of these values.

• Integer from 1 to 16 — This value specifies the bit width for all compressed IQ samples.

• K/12-by-L-by-P integer array with values from 1 to 16 — The vector elements specify the bit width of the compressed IQ samples within a PRB.

Data Types: double

IQWidth — Bit width of IQ samples before compression

integer from 1 to 32

Bit width of the IQ samples before compression (in the input grid), specified as an integer from 1 to 32. For μ-law compression, this input must be 16. For BFP compression, the function ignores this input.

Data Types: double

Output Arguments

cGrid — Compressed resource grid of U-Plane IQ data

K-by-L-by-P complex array

Compressed resource grid of U-Plane IQ data, returned as a K-by-L-by-P complex array. The input cIQWidth specifies the bit width of the IQ samples in the compressed grid.

Data Types: double | single
Complex Number Support: Yes

cParam — Compression parameter

K/12-by-L-by-P integer array

Compression parameter, returned as a K/12-by-L-by-P integer array.

• For BFP compression, cParam is the common exponent applied per compressed PRB, with integer values from 0 to IQWidth – 1.

• For block scaling compression, cParam is the common scale factor applied per compressed PRB, with integer values from 1 to 128.

• For μ-law compression, cParam is the common shift applied per compressed PRB, with integer values from 0 to 7.

Data Types: double

References

[1] O-RAN.WG4.CUS TS. “O-RAN Fronthaul Working Group; Control, User and Synchronization Plane Specification” Open Radio Access Network Alliance Technical Specification.

Extended Capabilities

C/C++ Code Generation
Generate C and C++ code using MATLAB® Coder™.

Version History

Introduced in R2022b