Visible to Intel only — GUID: GUID-270337C4-295C-47DD-AF9E-A56964552232
Visible to Intel only — GUID: GUID-270337C4-295C-47DD-AF9E-A56964552232
MimoMMSE
DEPRECATED. Implements the MIMO MMSE estimator algorithm.
Syntax
IppStatus ippsMimoMMSE_1X2_16sc(Ipp16sc* pSrcH[2], int srcHStride2, int srcHStride1, int srcHStride0, Ipp16sc* pSrcY[4][12], int Sigma2, IppFourSymb* pDstX, int dstXStride1, int dstXStride0, int numSymb, int numSC, int SINRIdx, Ipp32f* pDstSINR, int scaleFactor);
IppStatus ippsMimoMMSE_2X2_16sc(Ipp16sc* pSrcH[2], int srcHStride2, int srcHStride1, int srcHStride0, Ipp16sc* pSrcY[4][12], int Sigma2, IppFourSymb* pDstX, int dstXStride1, int dstXStride0, int numSymb, int numSC, int SINRIdx, Ipp32f* pDstSINR, int scaleFactor);
IppStatus ippsMimoMMSE_1X4_16sc(Ipp16sc* pSrcH[2], int srcHStride2, int srcHStride1, int srcHStride0, Ipp16sc* pSrcY[4][12], int Sigma2, IppFourSymb* pDstX, int dstXStride1, int dstXStride0, int numSymb, int numSC, int SINRIdx, Ipp32f* pDstSINR, int scaleFactor);
IppStatus ippsMimoMMSE_2X4_16sc(Ipp16sc* pSrcH[2], int srcHStride2, int srcHStride1, int srcHStride0, Ipp16sc* pSrcY[4][12], int Sigma2, IppFourSymb* pDstX, int dstXStride1, int dstXStride0, int numSymb, int numSC, int SINRIdx, Ipp32f* pDstSINR, int scaleFactor);
Include Files
ippe.h
Parameters
pSrcH |
Pointer to line 2 of the H matrix. |
srcHStride2 |
Stride between H matrices (H[symb0] and (H[symb1]). |
srcHStride1 |
Stride between rows of the H matrix (h00 and h10) |
srcHStride0 |
Stride between elements of the row (h00 and h01). |
pSrcY |
Array of pointers to the RX signal Y. The maximum size is four TX antennas and 12 symbols. |
Sigma2 |
Noise power. |
numSymb |
Number of symbols. |
numSC |
Number of subcarriers. |
pDstX |
Pointer to the estimated TX signal grouped by four symbols (quads). |
dstXStride1 |
Stride between TX signals (X[ant0] and X[ant1]). |
dstXStride0 |
Stride between quads inside one antenna. |
SINRIdx |
Index of symbol to calculate the SINR. |
pDstSINR |
Pointer to an array of SINR for layer 1,2. |
scaleFactor |
Scale factor, refer to Integer Scaling. |
Description
This functionality is available only within the Intel® System Studio suite.
Calculate B=HH*H, where H is the channel matrix, and superscript H denotes Hermitian operator (transpose and conjugate).
Calculate A=HH*H+σ2n*INt, where
Nt is the number of transmit antennas (1 or 2).
Calculate A-1=(HH*H+σ2n*INt)-1.
Calculate Z=HH*y.
Calculate x=A-1*Z=(HH*H+σ2n*INt)-1*HH*y.
Calculate D=W*H=A-1*HH*H.
Signal to interference plus noise ratio (SINR) is computed as follows:
where
M is the number of subcarriers.
The destination data is grouped by four symbols.
Return Values
ippStsNoErr |
Indicates no error. |
ippStsNullPtrErr |
Indicates an error if one of the specified pointers is NULL. |
ippStsSizeErr |
Indicates an error if numSymb or numSC is less than, or equal to zero. |