RadarTranspose#

Functions#

NVCVStatus pvaRadarTransposeCreate(NVCVOperatorHandle *handle, NVCVTensorRequirements const *const inTensorRequirements, NVCVTensorRequirements const *const outTensorRequirements, PVARadarTransposeParams const *const params)

Constructs an instance of the RadarTranspose operator.

NVCVStatus pvaRadarTransposeSubmit(NVCVOperatorHandle handle, cudaStream_t stream, NVCVTensorHandle const in, NVCVTensorHandle out)

Submits the RadarTranspose operator to a CUDA stream.

NVCVStatus pvaRadarTransposeSubmit(NVCVOperatorHandle handle, cupvaStream_t stream, NVCVTensorHandle const in, NVCVTensorHandle out)

Submits the RadarTranspose operator to a cuPVA stream.

Structs#

PVARadarTransposeParams

Parameters for the RadarTranspose operator.

Functions#

NVCVStatus pvaRadarTransposeCreate(
NVCVOperatorHandle *handle,
NVCVTensorRequirements const *const inTensorRequirements,
NVCVTensorRequirements const *const outTensorRequirements,
PVARadarTransposeParams const *const params,
)#

Constructs an instance of the RadarTranspose operator.

Transposes data from PVA_FFT_LAYOUT_RANGE_RX_DOPPLER to the requested output layout.

Limitations:

  • nRange (range bins) and nDoppler (Doppler bins) must be in [1, 1024]. nRx (RX channels) must be in [1, 256].

  • Input and output tensors must be densely packed, with no padding between dimensions.

  • outputLayout must be PVA_FFT_LAYOUT_RANGE_DOPPLER_RX or PVA_FFT_LAYOUT_DOPPLER_RX_RANGE.

  • For PVA_FFT_LAYOUT_DOPPLER_RX_RANGE, nRange must be a multiple of 4, and a multiple of 8 above 512.

  • Some other combinations of nRange, nRx and nDoppler are not supported, for example all three at their maximum. Create rejects them with NVCV_ERROR_INVALID_ARGUMENT.

Input: Data Layout: [HCW], [NHCW] only when N = 1 H: Number of range bins. C: Number of RX channels. W: Number of Doppler bins.

Batches: [1]

Rank: 3 for [HCW] or 4 for [NHCW] only when N = 1.

Data Type

Allowed

8bit Unsigned

No

8bit Signed

No

16bit Unsigned

No

16bit Signed

No

32bit Unsigned

No

32bit Signed

No

32bit Float

No

64bit Float

No

2x16bit Unsigned

No

2x16bit Signed

No

2x32bit Unsigned

No

2x32bit Signed

Yes

Output: Data Layout: [HCW], [NHCW] only when N = 1 If outputLayout is PVA_FFT_LAYOUT_RANGE_DOPPLER_RX: H: Number of range bins. C: Number of Doppler bins. W: Number of RX channels. If outputLayout is PVA_FFT_LAYOUT_DOPPLER_RX_RANGE: H: Number of Doppler bins. C: Number of RX channels. W: Number of range bins. The H, C and W dimensions must exactly match the permutation selected by outputLayout.

Batches: [1]

Rank: 3 for [HCW] or 4 for [NHCW] only when N = 1.

Data Type

Allowed

8bit Unsigned

No

8bit Signed

No

16bit Unsigned

No

16bit Signed

No

32bit Unsigned

No

32bit Signed

No

32bit Float

No

64bit Float

No

2x16bit Unsigned

No

2x16bit Signed

No

2x32bit Unsigned

No

2x32bit Signed

Yes

Input/Output Dependency:

Property

Input == Output

Data Layout

Yes

Data Type

Yes

Number

Yes

Channels

No, permuted per outputLayout (see Output)

Width

No, permuted per outputLayout (see Output)

Height

No, permuted per outputLayout (see Output)

Parameters:
  • handle – [out] Where the operator instance handle will be written to.

    • Must not be NULL.

  • inTensorRequirements – [in] Pointer to the NVCVTensorRequirements structure which contains input Tensor layout rank, shape and data type information.

    • Must not be NULL.

  • outTensorRequirements – [in] Pointer to the NVCVTensorRequirements structure which contains output Tensor layout rank, shape and data type information.

    • Must not be NULL.

  • params – [in] Pointer to PVARadarTransposeParams.

    • Must not be NULL.

Return values:
  • NVCV_ERROR_INVALID_ARGUMENT – A required pointer is null, a tensor requirement does not use a supported layout, rank, shape or data type, or a parameter is invalid.

  • NVCV_ERROR_OUT_OF_MEMORY – Not enough memory.

  • NVCV_SUCCESS – Operation executed successfully.

NVCVStatus pvaRadarTransposeSubmit(
NVCVOperatorHandle handle,
cudaStream_t stream,
NVCVTensorHandle const in,
NVCVTensorHandle out,
)#

Submits the RadarTranspose 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

Parameters:
  • handle – [in] Handle to the operator.

  • stream – [in] Handle to a valid CUDA stream.

  • in – [in] Input tensor handle.

  • out – [out] Output tensor handle.

Return values:
  • NVCV_ERROR_INVALID_ARGUMENT – Handle is null or not a RadarTranspose operator, a tensor base pointer is null, or a tensor does not match the requirements used to create the operator.

  • NVCV_ERROR_INTERNAL – Updating the data flows or submitting to the stream failed.

  • NVCV_SUCCESS – Operation executed successfully.

NVCVStatus pvaRadarTransposeSubmit(
NVCVOperatorHandle handle,
cupvaStream_t stream,
NVCVTensorHandle const in,
NVCVTensorHandle out,
)#

Submits the RadarTranspose operator to a cuPVA stream.

Parameters:
  • handle – [in] Handle to the operator.

    • Must not be NULL.

  • stream – [in] Handle to a valid cuPVA stream.

  • in – [in] Input tensor handle.

  • out – [out] Output tensor handle.

Return values:
  • NVCV_ERROR_INVALID_ARGUMENT – Handle is null or not a RadarTranspose operator, a tensor base pointer is null, or a tensor does not match the requirements used to create the operator.

  • NVCV_ERROR_INTERNAL – Updating the data flows or submitting to the stream failed.

  • NVCV_SUCCESS – Operation executed successfully.