Collaboration diagram for DCT Type III Functions:

Functions
tpt_status	tpt_dct3_forward_f16 (f16_t aOutData, f16_t aInData, f16_t *aScratch, size_t aLogN)
	Processing function for the floating-point DCT3.
tpt_status	tpt_dct3_inverse_f16 (f16_t aOutData, f16_t aInData, f16_t *aScratch, size_t aLogN)
static tpt_status	tpt_dct3_f16 (f16_t aOutData, f16_t aInData, f16_t *aScratch, size_t aLogN, bool aIdctFlag)
tpt_status	tpt_dct3_forward_f32 (f32_t aOutData, f32_t aInData, f32_t *aScratch, size_t aLogN)
	Processing function for the floating-point DCT3.
tpt_status	tpt_dct3_inverse_f32 (f32_t aOutData, f32_t aInData, f32_t *aScratch, size_t aLogN)
static tpt_status	tpt_dct3_f32 (f32_t aOutData, f32_t aInData, f32_t *aScratch, size_t aLogN, bool aIdctFlag)
tpt_status	tpt_dct3_forward_f64 (f64_t aOutData, f64_t aInData, f64_t *aScratch, size_t aLogN)
	Processing function for the floating-point DCT3.
tpt_status	tpt_dct3_inverse_f64 (f64_t aOutData, f64_t aInData, f64_t *aScratch, size_t aLogN)
static tpt_status	tpt_dct3_f64 (f64_t aOutData, f64_t aInData, f64_t *aScratch, size_t aLogN, bool aIdctFlag)
tpt_status	tpt_dct3_forward_q15 (q15_t aOutData, q15_t aInData, q15_t *aScratch, size_t aLogN)
	Processing function for the Q15 DCT3.
tpt_status	tpt_dct3_inverse_q15 (q15_t aOutData, q15_t aInData, q15_t *aScratch, size_t aLogN)
static tpt_status	tpt_dct3_q15 (q15_t aOutData, q15_t aInData, q15_t *aScratch, size_t aLogN, bool aIdctFlag)
tpt_status	tpt_dct3_forward_q31 (q31_t aOutData, q31_t aInData, q31_t *aScratch, size_t aLogN)
	Processing function for the Q31 DCT3.
tpt_status	tpt_dct3_inverse_q31 (q31_t aOutData, q31_t aInData, q31_t *aScratch, size_t aLogN)
static tpt_status	tpt_dct3_q31 (q31_t aOutData, q31_t aInData, q31_t *aScratch, size_t aLogN, bool aIdctFlag)

Detailed Description

A discrete cosine transform (DCT) expresses a finite sequence of data points in terms of a sum of cosine functions oscillating at different frequencies. In particular, a DCT is a Fourier-related transform similar to the discrete Fourier transform (DFT), but using only real numbers.

Algorithm

This function implements DCT Type III (IDCT type II) Function:

           N-1
    x[k] = sum { c[n] * y[n] * cos(PI * (2 * k + 1) * n / (2 * N)) }
           n=0

where c[0] = 1 / 2.0 and c[n] = 1 for n != 0. The N-point DCT3 internally uses N-point CFFT implementation. It supports lengths of [16, 32, 64, ..., 1024] and the corresponding aLogN value is [4, 5, 6, ..., 10].

Normalizing factor: The normalizing factor is 1.0. If necessary, the caller should calculate the normalizing factor separately.

Function Documentation

◆ tpt_dct3_f16()

tpt_status tpt_dct3_f16	(	f16_t *	aOutData,
		f16_t *	aInData,
		f16_t *	aScratch,
		size_t	aLogN,
		bool	aIdctFlag )

inlinestatic

◆ tpt_dct3_f32()

tpt_status tpt_dct3_f32	(	f32_t *	aOutData,
		f32_t *	aInData,
		f32_t *	aScratch,
		size_t	aLogN,
		bool	aIdctFlag )

inlinestatic

◆ tpt_dct3_f64()

tpt_status tpt_dct3_f64	(	f64_t *	aOutData,
		f64_t *	aInData,
		f64_t *	aScratch,
		size_t	aLogN,
		bool	aIdctFlag )

inlinestatic

◆ tpt_dct3_forward_f16()

tpt_status tpt_dct3_forward_f16	(	f16_t *	aOutData,
		f16_t *	aInData,
		f16_t *	aScratch,
		size_t	aLogN )

Processing function for the floating-point DCT3.

Parameters

[out]	aOutData	points to the output buffer.
[in]	aInData	points to the input buffer.
[in]	aScratch	points to a temporary buffer of size 2^(aLogN+1).
[in]	aLogN	The length is 2^aLogN.

Returns

execution status

TPT_MATH_SUCCESS : Operation is successful.
TPT_MATH_ARGUMENT_ERROR : aLogN is not supported.

◆ tpt_dct3_forward_f32()

tpt_status tpt_dct3_forward_f32	(	f32_t *	aOutData,
		f32_t *	aInData,
		f32_t *	aScratch,
		size_t	aLogN )

Processing function for the floating-point DCT3.

Parameters

[out]	aOutData	points to the output buffer.
[in]	aInData	points to the input buffer.
[in]	aScratch	points to a temporary buffer of size 2^(aLogN+1).
[in]	aLogN	The length is 2^aLogN.

Returns

execution status

TPT_MATH_SUCCESS : Operation is successful.
TPT_MATH_ARGUMENT_ERROR : aLogN is not supported.

◆ tpt_dct3_forward_f64()

tpt_status tpt_dct3_forward_f64	(	f64_t *	aOutData,
		f64_t *	aInData,
		f64_t *	aScratch,
		size_t	aLogN )

Processing function for the floating-point DCT3.

Parameters

[in,out]	aInPlace	points to the in-place input and output buffer.
[in]	aScratch	points to a temporary buffer of size 2^(aLogN+1).
[in]	aLogN	The length is 2^aLogN.

Returns

execution status

TPT_MATH_SUCCESS : Operation is successful.
TPT_MATH_ARGUMENT_ERROR : aLogN is not supported.

◆ tpt_dct3_forward_q15()

tpt_status tpt_dct3_forward_q15	(	q15_t *	aOutData,
		q15_t *	aInData,
		q15_t *	aScratch,
		size_t	aLogN )

Processing function for the Q15 DCT3.

Parameters

[out]	aOutData	points to the output buffer.
[in]	aInData	points to the input buffer.
[in]	aScratch	points to a temporary buffer of size 2^(aLogN+1).
[in]	aLogN	The length is 2^aLogN.

Returns

execution status

TPT_MATH_SUCCESS : Operation is successful.
TPT_MATH_ARGUMENT_ERROR : aLogN is not supported.

Input an output formats: Internally inputs are downscaled in the CFFT process function to avoid overflows. Number of bits downscaled, depends on the size of the transform. The input and output formats for different DCT sizes and number of bits to upscale are mentioned in the table below:

Input and Output Formats for Q15 DCT3
aLogN	DCT3 Size	Input format	Output format	Number of bits to upscale
4	16	1.15	5.11	4
5	32	1.15	6.10	5
6	64	1.15	7.9	6
7	128	1.15	8.8	7
8	256	1.15	9.7	8
9	512	1.15	10.6	9
10	1024	1.15	11.5	10
......

◆ tpt_dct3_forward_q31()

tpt_status tpt_dct3_forward_q31	(	q31_t *	aOutData,
		q31_t *	aInData,
		q31_t *	aScratch,
		size_t	aLogN )

Processing function for the Q31 DCT3.

Parameters

[out]	aOutData	points to the output buffer.
[in]	aInData	points to the input buffer.
[in]	aScratch	points to a temporary buffer of size 2^(aLogN+1).
[in]	aLogN	The length is 2^aLogN.

Returns

execution status

TPT_MATH_SUCCESS : Operation is successful.
TPT_MATH_ARGUMENT_ERROR : aLogN is not supported.

Input an output formats: Internally inputs are downscaled in the CFFT process function to avoid overflows. Number of bits downscaled, depends on the size of the transform. The input and output formats for different DCT sizes and number of bits to upscale are mentioned in the table below:

Input and Output Formats for Q31 DCT3
aLogN	DCT3 Size	Input format	Output format	Number of bits to upscale
4	16	1.31	5.27	0
5	32	1.31	6.26	5
6	64	1.31	7.25	6
7	128	1.31	8.24	7
8	256	1.31	9.23	8
9	512	1.31	10.22	9
10	1024	1.31	11.21	10
......

◆ tpt_dct3_inverse_f16()

tpt_status tpt_dct3_inverse_f16	(	f16_t *	aOutData,
		f16_t *	aInData,
		f16_t *	aScratch,
		size_t	aLogN )

◆ tpt_dct3_inverse_f32()

tpt_status tpt_dct3_inverse_f32	(	f32_t *	aOutData,
		f32_t *	aInData,
		f32_t *	aScratch,
		size_t	aLogN )

◆ tpt_dct3_inverse_f64()

tpt_status tpt_dct3_inverse_f64	(	f64_t *	aOutData,
		f64_t *	aInData,
		f64_t *	aScratch,
		size_t	aLogN )

◆ tpt_dct3_inverse_q15()

tpt_status tpt_dct3_inverse_q15	(	q15_t *	aOutData,
		q15_t *	aInData,
		q15_t *	aScratch,
		size_t	aLogN )

◆ tpt_dct3_inverse_q31()

tpt_status tpt_dct3_inverse_q31	(	q31_t *	aOutData,
		q31_t *	aInData,
		q31_t *	aScratch,
		size_t	aLogN )

◆ tpt_dct3_q15()

tpt_status tpt_dct3_q15	(	q15_t *	aOutData,
		q15_t *	aInData,
		q15_t *	aScratch,
		size_t	aLogN,
		bool	aIdctFlag )

inlinestatic

◆ tpt_dct3_q31()

tpt_status tpt_dct3_q31	(	q31_t *	aOutData,
		q31_t *	aInData,
		q31_t *	aScratch,
		size_t	aLogN,
		bool	aIdctFlag )

inlinestatic