DSP-Library with SIMD functions
SIMD stands for 'Single Instruction and Multiple Data Stream'. It represents an organization that includes many processing units under the supervision of a common control unit.
| Functions | Description |
|---|---|
DSP_Filter_init(M_Filt_t *filt) |
General filter initialize function |
DSPkalman_initTrust(Kalman*kalman,Trust type) |
This function provide simple initialize for kalman filter |
DSPkalman_init(Kalman *kalman ,float32_t Q, float32_t R) |
Normal initialize for kalman filter this function provide Q,R constant values.This filter is 1-Dimensional so we don't use vectors,transpoze and inverse calculation. |
DSPexp_moving_avg(M_Filt_t *filt,float32_t input,float Alpha) |
Exponential moving average filter |
DSPmoving_med(M_Filt_t *filt ,float32_t input, unsigned int Numpoints) |
Moving median filter y = median value(X[N]), X[N]={ sorted N numbers} |
DSPmoving_av(M_Filt_t *filt,float32_t input, unsigned int Numpoints) |
Moving average filter y[n] = ( x[n] + x[n+1]...x[N-1] ) / N |
DSP_fir16(M_Filt_t *filt ,int16_t Src) |
FIR(Finit Impulse Response)filter |
DSP_fastfir16(M_Filt_t *filt ,int16_t Src) |
Fast FIR (Finit Impulse Response) filter with SIMD instructions. |
DSPstatistic_32(Statistic *st , int16_t sig, float32_t Sens_RMSnoise) |
32 - bit Statistical calculation |
DSPstatistic_32(Statistic *st , int16_t sig, float32_t Sens_RMSnoise) |
16 - bit Statistical calculation. |
FPU_enable() |
Device FPU(floating point unit ) activation |
DSPkalman(Kalman*kalman, float32_t data |
Kalman prediction |
insertionSort(float arr[], unsigned int n) |
instertion sort |
- FIR Filter float
- FIR Filter 16 - bit
- Fast FIR Filter
- Exponential Moving Average Filter
- Moving Average Filter
- Moving Median Filter
- Statistic calculation 16 and 32 - bit (Standard deviation,mean,SNR,variance)
- SIMD support
- CMSIS DSP Library
- Cortex M4
