Collaboration diagram for Modified DCT Functions:

Functions
void	tpt_imdct12_f32 (f32_t aOutData, const f32_t aInData)
	Processing function for the floating-point 12-point IMDCT. More...

void	tpt_imdct12_q15 (q3_12_t aOutData, const q15_t aInData)
	Processing function for the Q15 12-point IMDCT. More...

void	tpt_imdct12_q31 (q3_28_t aOutData, const q31_t aInData)
	Processing function for the Q31 12-point IMDCT. More...

void	tpt_imdct36_f32 (f32_t aOutData, const f32_t aInData)
	Processing function for the floating-point 36-point IMDCT. More...

void	tpt_imdct36_q15 (q5_10_t aOutData, const q15_t aInData)
	Processing function for the Q31 36-point IMDCT. More...

void	tpt_imdct36_q31 (q5_26_t aOutData, const q31_t aInData)
	Processing function for the Q31 36-point IMDCT. More...

void	tpt_mdct12_f32 (f32_t aOutData, const f32_t aInData)
	Processing function for the floating-point 12-point MDCT. More...

void	tpt_mdct12_q15 (q4_11_t aOutData, const q15_t aInData)
	Processing function for the Q15 12-point MDCT. More...

void	tpt_mdct12_q31 (q4_27_t aOutData, const q31_t aInData)
	Processing function for the Q31 12-point MDCT. More...

void	tpt_mdct36_f32 (f32_t aOutData, const f32_t aInData)
	Processing function for the floating-point 36-point MDCT. More...

void	tpt_mdct36_q15 (q6_9_t aOutData, const q15_t aInData)
	Processing function for the Q15 36-point MDCT. More...

void	tpt_mdct36_q31 (q6_25_t aOutData, const q31_t aInData)
	Processing function for the Q31 36-point MDCT. More...

tpt_status	tpt_mdct_f32 (f32_t aOutData, f32_t aInData, size_t aLogN, bool aImdctFlag)
	Processing function for the floating-point MDCT. Source buffer is modified by this function. More...

tpt_status	tpt_mdct_q15 (q15_t aOutData, q15_t aInData, size_t aLogN, bool aImdctFlag)
	Processing function for the Q15 MDCT. Source buffer is modified by this function. More...

tpt_status	tpt_mdct_q31 (q31_t aOutData, q31_t aInData, size_t aLogN, bool aImdctFlag)
	Processing function for the Q31 MDCT. Source buffer is modified by this function. More...

Detailed Description

The Modified Discrete Cosine Transform (MDCT) is widely used in speech and audio coding as analysis/synthesis filter bank with time domain alias cancellation property. Since it is a lapped transform it is particularly useful in mitigating blocking artifacts that arise in audio coding because of quantization of the spectral coefficients.

Algorithm

1] forward transform

           N-1
    X[k] = sum { x[n] * cos(PI * (4n + 2 + N) * (2k + 1) / (4N) }
           n=0

           for k = 0, 1, ..., N/2 - 1

2] inverse transform

          N/2-1
    x[k] = sum { X[n] * cos(PI * (4k + 2 + N) * (2n + 1) / (4N) }
           n=0

           for k = 0, 1, ..., N - 1

The N-point MDCT internally uses N/2-point DCT4 implementation. It supports lengths of [64, 128, ..., 4096] and the corresponding aLogN value is [6, 7, ..., 12]. There are separate functions for 12-point and 36-point.

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

Function Documentation

◆ tpt_imdct12_f32()

void tpt_imdct12_f32	(	f32_t *	aOutData,
		const f32_t *	aInData
	)

Processing function for the floating-point 12-point IMDCT.

Parameters

[out]	aOutData	points to output buffer.
[in]	aInData	points to input buffer. Note: the input length is 6 and the output length is 12.

Returns: none.

◆ tpt_imdct12_q15()

void tpt_imdct12_q15	(	q3_12_t *	aOutData,
		const q15_t *	aInData
	)

Processing function for the Q15 12-point IMDCT.

Parameters

[out]	aOutData	points to output buffer.
[in]	aInData	points to input buffer. Note: the input length is 6 and the output length is 12.

Returns: none.

Input and output formats:: Internally inputs are downscaled in the IMDCT process function to avoid overflows. The return is in 4.12 format.

◆ tpt_imdct12_q31()

void tpt_imdct12_q31	(	q3_28_t *	aOutData,
		const q31_t *	aInData
	)

Processing function for the Q31 12-point IMDCT.

Parameters

[out]	aOutData	points to output buffer.
[in]	aInData	points to input buffer. Note: the input length is 6 and the output length is 12.

Returns: none.

Input and output formats:: Internally inputs are downscaled in the IMDCT process function to avoid overflows. The return is in 4.28 format.

◆ tpt_imdct36_f32()

void tpt_imdct36_f32	(	f32_t *	aOutData,
		const f32_t *	aInData
	)

Processing function for the floating-point 36-point IMDCT.

Parameters

[out]	aOutData	points to output buffer.
[in]	aInData	points to input buffer. Note: the input length is 18 and the output length is 36.

Returns: none.

◆ tpt_imdct36_q15()

void tpt_imdct36_q15	(	q5_10_t *	aOutData,
		const q15_t *	aInData
	)

Processing function for the Q31 36-point IMDCT.

Parameters

[out]	aOutData	points to output buffer.
[in]	aInData	points to input buffer. Note: the input length is 18 and the output length is 36.

Returns: none.

Input and output formats:: Internally inputs are downscaled in the IMDCT process function to avoid overflows. The return is in 6.10 format.

◆ tpt_imdct36_q31()

void tpt_imdct36_q31	(	q5_26_t *	aOutData,
		const q31_t *	aInData
	)

Processing function for the Q31 36-point IMDCT.

Parameters

[out]	aOutData	points to output buffer.
[in]	aInData	points to input buffer. Note: the input length is 18 and the output length is 36.

Returns: none.

Input and output formats:: Internally inputs are downscaled in the IMDCT process function to avoid overflows. The return is in 6.26 format.

◆ tpt_mdct12_f32()

void tpt_mdct12_f32	(	f32_t *	aOutData,
		const f32_t *	aInData
	)

Processing function for the floating-point 12-point MDCT.

Parameters

[out]	aOutData	points to output buffer.
[in]	aInData	points to input buffer. Note: the input length is 12 and the output length is 6.

Returns: none.

◆ tpt_mdct12_q15()

void tpt_mdct12_q15	(	q4_11_t *	aOutData,
		const q15_t *	aInData
	)

Processing function for the Q15 12-point MDCT.

Parameters

[out]	aOutData	points to output buffer.
[in]	aInData	points to input buffer. Note: the input length is 12 and the output length is 6.

Returns: none.

Input and output formats:: Internally inputs are downscaled in the MDCT process function to avoid overflows. The return is in 5.11 format.

◆ tpt_mdct12_q31()

void tpt_mdct12_q31	(	q4_27_t *	aOutData,
		const q31_t *	aInData
	)

Processing function for the Q31 12-point MDCT.

Parameters

[out]	aOutData	points to output buffer.
[in]	aInData	points to input buffer. Note: the input length is 12 and the output length is 6.

Returns: none.

Input and output formats:: Internally inputs are downscaled in the MDCT process function to avoid overflows. The return is in 5.27 format.

◆ tpt_mdct36_f32()

void tpt_mdct36_f32	(	f32_t *	aOutData,
		const f32_t *	aInData
	)

Processing function for the floating-point 36-point MDCT.

Parameters

[out]	aOutData	points to output buffer.
[in]	aInData	points to input buffer. Note: the input length is 36 and the output length is 18.

Returns: none.

◆ tpt_mdct36_q15()

void tpt_mdct36_q15	(	q6_9_t *	aOutData,
		const q15_t *	aInData
	)

Processing function for the Q15 36-point MDCT.

Parameters

[out]	aOutData	points to output buffer.
[in]	aInData	points to input buffer. Note: the input length is 36 and the output length is 18.

Returns: none.

Input and output formats:: Internally inputs are downscaled in the MDCT process function to avoid overflows. The return is in 7.9 format.

◆ tpt_mdct36_q31()

void tpt_mdct36_q31	(	q6_25_t *	aOutData,
		const q31_t *	aInData
	)

Processing function for the Q31 36-point MDCT.

Parameters

[out]	aOutData	points to output buffer.
[in]	aInData	points to input buffer. Note: the input length is 36 and the output length is 18.

Returns: none.

Input and output formats:: Internally inputs are downscaled in the MDCT process function to avoid overflows. The return is in 7.25 format.

◆ tpt_mdct_f32()

tpt_status tpt_mdct_f32	(	f32_t *	aOutData,
		f32_t *	aInData,
		size_t	aLogN,
		bool	aImdctFlag
	)

Processing function for the floating-point MDCT. Source buffer is modified by this function.

Parameters

[out]	aOutData	points to output buffer
[in]	aInData	points to input buffer (Source buffer is modified by this function.)
[in]	aLogN	length is 2^aLogN
[in]	aImdctFlag	flag that selects transform direction. value = false: forward transform value = true : inverse transform Note: For forward transform: the input length is 2^aLogN and the output length is 2^(aLogN-1). For inverse transform: the input length is 2^(aLogN-1) and the output length is 2^aLogN.

Returns

execution status

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

◆ tpt_mdct_q15()

tpt_status tpt_mdct_q15	(	q15_t *	aOutData,
		q15_t *	aInData,
		size_t	aLogN,
		bool	aImdctFlag
	)

Processing function for the Q15 MDCT. Source buffer is modified by this function.

Parameters

[out]	aOutData	points to output buffer
[in]	aInData	points to input buffer (Source buffer is modified by this function.)
[in]	aLogN	length is 2^aLogN
[in]	aImdctFlag	flag that selects transform direction. value = false: forward transform value = true : inverse transform Note: For forward transform: the input length is 2^aLogN and the output length is 2^(aLogN-1). For inverse transform: the input length is 2^(aLogN-1) and the output 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 MDCT process function to avoid overflows. Number of bits downscaled, depends on the size of the transform. The input and output formats for different MDCT sizes and number of bits to upscale are mentioned in the table below:

Input and Output Formats for Q15 MDCT
aLogN	MDCT Size	Input format	Output format	Number of bits to upscale
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
......

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

◆ tpt_mdct_q31()

tpt_status tpt_mdct_q31	(	q31_t *	aOutData,
		q31_t *	aInData,
		size_t	aLogN,
		bool	aImdctFlag
	)

Processing function for the Q31 MDCT. Source buffer is modified by this function.

Parameters

[out]	aOutData	points to output buffer
[in]	aInData	points to input buffer (Source buffer is modified by this function.)
[in]	aLogN	length is 2^aLogN
[in]	aImdctFlag	flag that selects transform direction. value = false: forward transform value = true : inverse transform Note: For forward transform: the input length is 2^aLogN and the output length is 2^(aLogN-1). For inverse transform: the input length is 2^(aLogN-1) and the output 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 MDCT process function to avoid overflows. Number of bits downscaled, depends on the size of the transform. The input and output formats for different MDCT sizes and number of bits to upscale are mentioned in the table below:

Input and Output Formats for Q31 MDCT
aLogN	MDCT Size	Input format	Output format	Number of bits to upscale
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
......

Input and Output Formats for Q31 IMDCT
aLogN	MDCT Size	Input format	Output format	Number of bits to upscale
6	64	1.31	6.26	5
7	128	1.31	7.25	6
8	256	1.31	8.24	7
9	512	1.31	9.23	8
10	1024	1.31	10.22	9
......