CMSIS-DSP  Version 1.4.4
CMSIS DSP Software Library
 All Data Structures Files Functions Variables Typedefs Enumerations Enumerator Macros Groups Pages
Vector Inverse Clarke Transform

Functions

static __INLINE void arm_inv_clarke_f32 (float32_t Ialpha, float32_t Ibeta, float32_t *pIa, float32_t *pIb)
 Floating-point Inverse Clarke transform.
 
static __INLINE void arm_inv_clarke_q31 (q31_t Ialpha, q31_t Ibeta, q31_t *pIa, q31_t *pIb)
 Inverse Clarke transform for Q31 version.
 

Description

Inverse Clarke transform converts the two-coordinate time invariant vector into instantaneous stator phases.

The function operates on a single sample of data and each call to the function returns the processed output. The library provides separate functions for Q31 and floating-point data types.

Algorithm
clarkeInvFormula.gif
where pIa and pIb are the instantaneous stator phases and Ialpha and Ibeta are the two coordinates of time invariant vector.
Fixed-Point Behavior
Care must be taken when using the Q31 version of the Clarke transform. In particular, the overflow and saturation behavior of the accumulator used must be considered. Refer to the function specific documentation below for usage guidelines.

Function Documentation

static __INLINE void arm_inv_clarke_f32 ( float32_t  Ialpha,
float32_t  Ibeta,
float32_t pIa,
float32_t pIb 
)
static
Parameters
[in]Ialphainput two-phase orthogonal vector axis alpha
[in]Ibetainput two-phase orthogonal vector axis beta
[out]*pIapoints to output three-phase coordinate a
[out]*pIbpoints to output three-phase coordinate b
Returns
none.
static __INLINE void arm_inv_clarke_q31 ( q31_t  Ialpha,
q31_t  Ibeta,
q31_t pIa,
q31_t pIb 
)
static
Parameters
[in]Ialphainput two-phase orthogonal vector axis alpha
[in]Ibetainput two-phase orthogonal vector axis beta
[out]*pIapoints to output three-phase coordinate a
[out]*pIbpoints to output three-phase coordinate b
Returns
none.

Scaling and Overflow Behavior:

The function is implemented using an internal 32-bit accumulator. The accumulator maintains 1.31 format by truncating lower 31 bits of the intermediate multiplication in 2.62 format. There is saturation on the subtraction, hence there is no risk of overflow.