RadarCaponROIPreprocess#

Functions#

NVCVStatus pvaCaponROIPreprocessCreate(NVCVOperatorHandle *handle, NVCVTensorRequirements const *const wazReqs, NVCVTensorRequirements const *const welReqs, NVCVTensorRequirements const *const detectionCountReqs, NVCVTensorRequirements const *const detDopplersReqs, NVCVTensorRequirements const *const elementSnapshotsReqs, NVCVTensorRequirements const *const rangeDopplerPlanesReqs, NVCVTensorRequirements const *const steeringLutReqs, NVCVTensorRequirements const *const snapshotsROIReqs, NVCVTensorRequirements const *const steeringVectorsROIReqs, PVACaponROIPreprocessParams const *const params)

Constructs the Capon ROI preprocessing (projection-core) operator.

NVCVStatus pvaCaponROIPreprocessSubmit(NVCVOperatorHandle handle, cudaStream_t stream, NVCVTensorHandle const waz, NVCVTensorHandle const wel, NVCVTensorHandle const detectionCount, NVCVTensorHandle const detDopplers, NVCVTensorHandle const elementSnapshots, NVCVTensorHandle const rangeDopplerPlanes, NVCVTensorHandle const steeringLut, NVCVTensorHandle snapshotsROI, NVCVTensorHandle steeringVectorsROI)

Submits the Capon ROI preprocessing (projection-core) operator to a CUDA stream.

NVCVStatus pvaCaponROIPreprocessSubmit(NVCVOperatorHandle handle, cupvaStream_t stream, NVCVTensorHandle const waz, NVCVTensorHandle const wel, NVCVTensorHandle const detectionCount, NVCVTensorHandle const detDopplers, NVCVTensorHandle const elementSnapshots, NVCVTensorHandle const rangeDopplerPlanes, NVCVTensorHandle const steeringLut, NVCVTensorHandle snapshotsROI, NVCVTensorHandle steeringVectorsROI)

Submits the Capon ROI preprocessing (projection-core) operator to a cuPVA stream.

Functions#

NVCVStatus pvaCaponROIPreprocessCreate(
NVCVOperatorHandle *handle,
NVCVTensorRequirements const *const wazReqs,
NVCVTensorRequirements const *const welReqs,
NVCVTensorRequirements const *const detectionCountReqs,
NVCVTensorRequirements const *const detDopplersReqs,
NVCVTensorRequirements const *const elementSnapshotsReqs,
NVCVTensorRequirements const *const rangeDopplerPlanesReqs,
NVCVTensorRequirements const *const steeringLutReqs,
NVCVTensorRequirements const *const snapshotsROIReqs,
NVCVTensorRequirements const *const steeringVectorsROIReqs,
PVACaponROIPreprocessParams const *const params,
)#

Constructs the Capon ROI preprocessing (projection-core) operator.

Limitations:

  • numTx and numRx in [1, 8], and numTx*numRx <= numAzVirt*numElVirt.

  • ddmaWeightRe and ddmaWeightIm finite for every TX in [0, numTx).

  • numAzVirt in [1, 69], numElVirt in [1, 35], and numAzVirt*numElVirt <= 69.

  • beamAz in [1, 128], beamEl in [1, 64]; beamAz*numAzVirt <= 4096 unless a coarse-DoA grid is used.

  • coarseElVirt (when nonzero) in [numElVirt, 35]; a grid (coarseElVirt != numElVirt) requires numElVirt == 1 and the direct-DFT coarse DoA.

  • roiAz in [1, beamAz], roiEl in [1, beamEl], and svDim == roiAz*roiEl <= 16.

  • roiAz*numElVirt*numDopplerBinsCovMat <= 1024.

  • numAzBins in [1, 16], numElBins in [1, 8].

  • numDopplerBinsCovMat (H) a multiple of 8 in [8, 64]; dopplerNeighborhood >= 2, and 2*floor(dopplerNeighborhood/2), doubled when enableFBA, must not exceed H (the remaining snapshot rows are zero).

  • rangeDopplerPlanes: every value a detection reads (its row among the first detectionCount, at the chirps of its Doppler window shifted by each TX’s relDopplerOffset) must decode to real and imaginary parts that are finite and below 2^24 in magnitude. Not checked: other values give that detection unspecified (saturated) snapshots.

  • elementSnapshots (first detectionCount rows): real and imaginary parts below 4 in magnitude, since coarse DoA converts them to 16-bit Q13. For numElVirt == 1 the azimuth beamspace power is also kept in 32 bits, which needs the sum of |elementSnapshots| over a row below 180 (about 181 less a rounding margin, for twiddles of complex magnitude at most 1; for example below 2.8 per element at numAzVirt == 64). Not checked; larger values saturate or wrap and can select the wrong ROI. Scaling a row by a common factor does not change its ROI.

  • Custom waz/wel (useCustomTwiddles) entries finite with complex magnitude at most 1; Submit rejects others.

  • For numElVirt > 1 the coarse DoA ranks the 2D beamspace on fixed-point power, with no floating-point refine (numElVirt == 1 re-scores its strongest candidates in FP32): when two peaks are within about 1e-3 in relative power, it can select another ROI than a double-precision reference.

  • inputQBits and steeringQBits in [0, 30]. snapshotsROI values are not saturated: a projected snapshot of magnitude 2^(31-inputQBits) or more wraps.

  • numChirps in [1, 512].

  • maxDet in [1, 1024]; above 256 it is tiled tileCols detections per DMA line, where tileCols is its smallest divisor >= ceil(maxDet/256) (there must be one <= 256), and tileCols times the larger of 2*numAzVirt*coarseElVirt and 2*numChirps*numRx must not exceed 65535.

  • coarseDoAUseFft requires useCustomTwiddles == false, numElVirt == 1, beamAz == 64, numAzVirt == 64, no coarse-DoA grid and the full coarse-DoA field of view.

  • The coarse-DoA field of view must be an ordered bin range inside [0, beamAz) x [0, beamEl).

  • Every tensor densely packed (no padding between rows).

Parameters:
  • handle – [out] Where the operator instance handle is written. Must not be NULL.

  • wazReqs – [in] Az twiddle table, HW [1, beamAz*numAzVirt] complex, NVCV_DATA_TYPE_2F32. Used only when params->useCustomTwiddles is set; may be NULL otherwise.

  • welReqs – [in] El twiddle table, HW [1, beamEl*coarseElVirt] complex, NVCV_DATA_TYPE_2F32 (coarseElVirt = numElVirt when no coarse-DoA grid is used). Used only when params->useCustomTwiddles is set; may be NULL otherwise.

  • detectionCountReqs – [in] Number of detections to process, W [1], NVCV_DATA_TYPE_S32; the value must be in [0, maxDet]. The values of the per-detection input rows at or beyond it are not used (the rows must still be valid memory), and their snapshotsROI / steeringVectorsROI rows are unspecified.

  • detDopplersReqs – [in] Per-detection detected Doppler bin, HW [1, maxDet], NVCV_DATA_TYPE_S32.

  • elementSnapshotsReqs – [in] Per-detection element snapshot for coarse DoA, HW [maxDet, numAzVirt*coarseElVirt] complex NVCV_DATA_TYPE_2F32; vc-index rx*numTx+tx, or position y*numAzVirt+x on a coarse-DoA grid.

  • rangeDopplerPlanesReqs – [in] Per-detection range-Doppler plane, HW [maxDet, numChirps*numRx] complex NVCV_DATA_TYPE_2F32 (host-gathered cube[rangeIdx]).

  • steeringLutReqs – [in] Precomputed per-ROI-position steering LUT, HW [numRoiPos, svDim*numElBins*numAzBins] complex NVCV_DATA_TYPE_2S32, where numRoiPos = (beamAz-roiAz+1)*(beamEl-roiEl+1) and the block for ROI start (roiAzStart, roiElStart) lives at row roiElStart*(beamAz-roiAz+1)+roiAzStart, quantized to Q(steeringQBits). Constant across frames; the device gathers one block per detection.

  • snapshotsROIReqs – [in] Output snapshots, NHW [maxDet, numDopplerBinsCovMat(H), svDim], NVCV_DATA_TYPE_2S32.

  • steeringVectorsROIReqs – [in] Output steering vectors, NCHW [maxDet, svDim, numElBins, numAzBins], NVCV_DATA_TYPE_2S32 (the LUT block selected for each detection’s ROI).

  • params – [in] Capon ROI preprocessing parameters. PVACaponROIPreprocessParams. Must not be NULL.

Return values:
  • NVCV_ERROR_INVALID_ARGUMENT – Handle, params or a required tensor requirement is NULL, or a parameter or tensor requirement is out of range.

  • NVCV_ERROR_OUT_OF_MEMORY – Not enough memory to create the operator.

  • NVCV_ERROR_INTERNAL – Internal error in the operator, e.g. building its program or data flows.

  • NVCV_SUCCESS – Operator created successfully.

NVCVStatus pvaCaponROIPreprocessSubmit(
NVCVOperatorHandle handle,
cudaStream_t stream,
NVCVTensorHandle const waz,
NVCVTensorHandle const wel,
NVCVTensorHandle const detectionCount,
NVCVTensorHandle const detDopplers,
NVCVTensorHandle const elementSnapshots,
NVCVTensorHandle const rangeDopplerPlanes,
NVCVTensorHandle const steeringLut,
NVCVTensorHandle snapshotsROI,
NVCVTensorHandle steeringVectorsROI,
)#

Submits the Capon ROI preprocessing (projection-core) operator to a CUDA stream.

Note

CUDA stream support requirements:

  • PVA SDK 2.7.0 or later

  • Jetpack 7 or later

  • DriveOS 7 or later

  • x86 Emulator is not supported

Note

detectionCount and custom waz/wel are read on the CPU during this call; see the file-level note.

Parameters:
  • handle – [in] Operator handle. Must not be NULL.

  • stream – [in] Valid CUDA stream.

  • waz – [in] Az twiddle table (see Create); may be NULL unless useCustomTwiddles is set.

  • wel – [in] El twiddle table (see Create); may be NULL unless useCustomTwiddles is set.

  • detectionCount – [in] Number of detections to process (see Create).

  • detDopplers – [in] Per-detection detected Doppler bin (see Create).

  • elementSnapshots – [in] Per-detection element snapshots for coarse DoA (see Create).

  • rangeDopplerPlanes – [in] Per-detection range-Doppler planes (see Create).

  • steeringLut – [in] Precomputed per-ROI-position steering LUT (see Create).

  • snapshotsROI – [out] Output ROI beamspace snapshots (see Create).

  • steeringVectorsROI – [out] Output per-detection steering vectors (see Create).

Return values:
  • NVCV_ERROR_INVALID_ARGUMENT – Handle or a required tensor is NULL, a tensor does not match the requirements passed to Create, a tensor read on the CPU is not host-accessible, detectionCount is outside [0, maxDet], or a custom waz/wel entry is not finite or has a complex magnitude above 1.

  • NVCV_ERROR_INTERNAL – Internal error in the operator, e.g. updating its data flows.

  • NVCV_ERROR_DEVICE – Submitting the PVA command failed.

  • NVCV_SUCCESS – Operation submitted successfully.

NVCVStatus pvaCaponROIPreprocessSubmit(
NVCVOperatorHandle handle,
cupvaStream_t stream,
NVCVTensorHandle const waz,
NVCVTensorHandle const wel,
NVCVTensorHandle const detectionCount,
NVCVTensorHandle const detDopplers,
NVCVTensorHandle const elementSnapshots,
NVCVTensorHandle const rangeDopplerPlanes,
NVCVTensorHandle const steeringLut,
NVCVTensorHandle snapshotsROI,
NVCVTensorHandle steeringVectorsROI,
)#

Submits the Capon ROI preprocessing (projection-core) operator to a cuPVA stream.

Note

detectionCount and custom waz/wel are read on the CPU during this call; see the file-level note.

Parameters:
  • handle – [in] Operator handle. Must not be NULL.

  • stream – [in] Valid cuPVA stream.

  • waz – [in] Az twiddle table (see Create); may be NULL unless useCustomTwiddles is set.

  • wel – [in] El twiddle table (see Create); may be NULL unless useCustomTwiddles is set.

  • detectionCount – [in] Number of detections to process (see Create).

  • detDopplers – [in] Per-detection detected Doppler bin (see Create).

  • elementSnapshots – [in] Per-detection element snapshots for coarse DoA (see Create).

  • rangeDopplerPlanes – [in] Per-detection range-Doppler planes (see Create).

  • steeringLut – [in] Precomputed per-ROI-position steering LUT (see Create).

  • snapshotsROI – [out] Output ROI beamspace snapshots (see Create).

  • steeringVectorsROI – [out] Output per-detection steering vectors (see Create).

Return values:
  • NVCV_ERROR_INVALID_ARGUMENT – Handle or a required tensor is NULL, a tensor does not match the requirements passed to Create, a tensor read on the CPU is not host-accessible, detectionCount is outside [0, maxDet], or a custom waz/wel entry is not finite or has a complex magnitude above 1.

  • NVCV_ERROR_INTERNAL – Internal error in the operator, e.g. updating its data flows.

  • NVCV_ERROR_DEVICE – Submitting the PVA command failed.

  • NVCV_SUCCESS – Operation submitted successfully.