You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

356 lines
16 KiB
C++

/**
*
* \section COPYRIGHT
*
* Copyright 2013-2021 Software Radio Systems Limited
*
* By using this file, you agree to the terms and conditions set
* forth in the LICENSE file which can be found at the top level of
* the distribution.
*
*/
/******************************************************************************
* File: vector.h
*
* Description: Vector functions using SIMD instructions where possible.
*
* Reference:
*****************************************************************************/
#ifndef SRSRAN_VECTOR_H
#define SRSRAN_VECTOR_H
#ifdef __cplusplus
extern "C" {
#endif
#include "srsran/config.h"
#include <math.h>
#include <stdbool.h>
#include <stdint.h>
#include <stdio.h>
#define SRSRAN_MEM_ALLOC(T, N) ((T*)srsran_vec_malloc((uint32_t)sizeof(T) * (N)))
#define SRSRAN_MEM_ZERO(Q, T, N) \
do { \
T* ptr_ = (Q); \
srsran_vec_zero((void*)ptr_, (uint32_t)sizeof(T) * (N)); \
} while (false)
#define SRSRAN_MAX(a, b) ((a) > (b) ? (a) : (b))
#define SRSRAN_MIN(a, b) ((a) < (b) ? (a) : (b))
#define SRSRAN_CEIL(NUM, DEN) (((NUM) + ((DEN)-1)) / (DEN))
#define SRSRAN_FLOOR(NUM, DEN) ((NUM) / (DEN))
#define SRSRAN_ROUND(NUM, DEN) ((uint32_t)round((double)(NUM) / (double)(DEN)))
// Complex squared absolute value
#define SRSRAN_CSQABS(X) (__real__(X) * __real__(X) + __imag__(X) * __imag__(X))
// Cumulative moving average
#define SRSRAN_VEC_CMA(data, average, n) ((average) + ((data) - (average)) / ((n) + 1))
// Cumulative moving average
#ifdef __cplusplus
#define SRSRAN_VEC_SAFE_CMA(data, average, n) (std::isnormal(average) ? SRSRAN_VEC_CMA(data, average, n) : (data))
#else
#define SRSRAN_VEC_SAFE_CMA(data, average, n) (isnormal(average) ? SRSRAN_VEC_CMA(data, average, n) : (data))
#endif
// Proportional moving average
#define SRSRAN_VEC_PMA(average1, n1, average2, n2) (((average1) * (n1) + (average2) * (n2)) / ((n1) + (n2)))
// Exponential moving average
#define SRSRAN_VEC_EMA(data, average, alpha) ((alpha) * (data) + (1 - alpha) * (average))
// Safe exponential moving average
#ifdef __cplusplus
#define SRSRAN_VEC_SAFE_EMA(data, average, alpha) \
(std::isnormal(average) ? SRSRAN_VEC_EMA(data, average, alpha) : (data))
#else
#define SRSRAN_VEC_SAFE_EMA(data, average, alpha) (isnormal(average) ? SRSRAN_VEC_EMA(data, average, alpha) : (data))
#endif
static inline float srsran_convert_amplitude_to_dB(float v)
{
return 20.0f * log10f(v);
}
static inline float srsran_convert_power_to_dB(float v)
{
return 10.0f * log10f(v);
}
static inline float srsran_convert_power_to_dBm(float v)
{
return srsran_convert_power_to_dB(v) + 30.0f;
}
static inline float srsran_convert_dB_to_amplitude(float v)
{
return powf(10.0f, v / 20.0f);
}
static inline float srsran_convert_dB_to_power(float v)
{
return powf(10.0f, v / 10.0f);
}
/*!
* Computes \f$ z = x \oplus y \f$ elementwise.
* \param[in] x A pointer to a vector of uint8_t with 0's and 1's.
* \param[in] y A pointer to a vector of uint8_t with 0's and 1's.
* \param[out] z A pointer to a vector of uint8_t with 0's and 1's.
* \param[in] len Length of vectors x, y and z.
*/
SRSRAN_API void srsran_vec_xor_bbb(const uint8_t* x, const uint8_t* y, uint8_t* z, const uint32_t len);
/** Return the sum of all the elements */
SRSRAN_API float srsran_vec_acc_ff(const float* x, const uint32_t len);
SRSRAN_API cf_t srsran_vec_acc_cc(const cf_t* x, const uint32_t len);
SRSRAN_API void* srsran_vec_malloc(uint32_t size);
SRSRAN_API cf_t* srsran_vec_cf_malloc(uint32_t size);
SRSRAN_API float* srsran_vec_f_malloc(uint32_t size);
SRSRAN_API int32_t* srsran_vec_i32_malloc(uint32_t nsamples);
SRSRAN_API uint32_t* srsran_vec_u32_malloc(uint32_t nsamples);
SRSRAN_API int16_t* srsran_vec_i16_malloc(uint32_t nsamples);
SRSRAN_API uint16_t* srsran_vec_u16_malloc(uint32_t nsamples);
SRSRAN_API int8_t* srsran_vec_i8_malloc(uint32_t nsamples);
SRSRAN_API uint8_t* srsran_vec_u8_malloc(uint32_t nsamples);
SRSRAN_API void* srsran_vec_realloc(void* ptr, uint32_t old_size, uint32_t new_size);
/* Zero memory */
SRSRAN_API void srsran_vec_zero(void* ptr, uint32_t nsamples);
SRSRAN_API void srsran_vec_cf_zero(cf_t* ptr, uint32_t nsamples);
SRSRAN_API void srsran_vec_f_zero(float* ptr, uint32_t nsamples);
SRSRAN_API void srsran_vec_i8_zero(int8_t* ptr, uint32_t nsamples);
SRSRAN_API void srsran_vec_u8_zero(uint8_t* ptr, uint32_t nsamples);
SRSRAN_API void srsran_vec_i16_zero(int16_t* ptr, uint32_t nsamples);
SRSRAN_API void srsran_vec_u32_zero(uint32_t* ptr, uint32_t nsamples);
/* Copy memory */
SRSRAN_API void srsran_vec_cf_copy(cf_t* dst, const cf_t* src, uint32_t len);
SRSRAN_API void srsran_vec_f_copy(float* dst, const float* src, uint32_t len);
SRSRAN_API void srsran_vec_u8_copy(uint8_t* dst, const uint8_t* src, uint32_t len);
SRSRAN_API void srsran_vec_i8_copy(int8_t* dst, const int8_t* src, uint32_t len);
SRSRAN_API void srsran_vec_u16_copy(uint16_t* dst, const uint16_t* src, uint32_t len);
SRSRAN_API void srsran_vec_i16_copy(int16_t* dst, const int16_t* src, uint32_t len);
/* print vectors */
SRSRAN_API void srsran_vec_fprint_c(FILE* stream, const cf_t* x, const uint32_t len);
SRSRAN_API void srsran_vec_fprint_f(FILE* stream, const float* x, const uint32_t len);
SRSRAN_API void srsran_vec_fprint_b(FILE* stream, const uint8_t* x, const uint32_t len);
SRSRAN_API void srsran_vec_fprint_bs(FILE* stream, const int8_t* x, const uint32_t len);
SRSRAN_API void srsran_vec_fprint_byte(FILE* stream, const uint8_t* x, const uint32_t len);
SRSRAN_API void srsran_vec_fprint_i(FILE* stream, const int* x, const uint32_t len);
SRSRAN_API void srsran_vec_fprint_s(FILE* stream, const int16_t* x, const uint32_t len);
SRSRAN_API void srsran_vec_fprint_hex(FILE* stream, uint8_t* x, const uint32_t len);
SRSRAN_API uint32_t srsran_vec_sprint_hex(char* str, const uint32_t max_str_len, uint8_t* x, const uint32_t len);
SRSRAN_API void srsran_vec_sprint_bin(char* str, const uint32_t max_str_len, const uint8_t* x, const uint32_t len);
/* Saves/loads a vector to a file */
SRSRAN_API void srsran_vec_save_file(char* filename, const void* buffer, const uint32_t len);
SRSRAN_API void srsran_vec_load_file(char* filename, void* buffer, const uint32_t len);
/* sum two vectors */
SRSRAN_API void srsran_vec_sum_fff(const float* x, const float* y, float* z, const uint32_t len);
SRSRAN_API void srsran_vec_sum_ccc(const cf_t* x, const cf_t* y, cf_t* z, const uint32_t len);
SRSRAN_API void srsran_vec_sum_sss(const int16_t* x, const int16_t* y, int16_t* z, const uint32_t len);
/* substract two vectors z=x-y */
SRSRAN_API void srsran_vec_sub_fff(const float* x, const float* y, float* z, const uint32_t len);
SRSRAN_API void srsran_vec_sub_ccc(const cf_t* x, const cf_t* y, cf_t* z, const uint32_t len);
SRSRAN_API void srsran_vec_sub_sss(const int16_t* x, const int16_t* y, int16_t* z, const uint32_t len);
SRSRAN_API void srsran_vec_sub_bbb(const int8_t* x, const int8_t* y, int8_t* z, const uint32_t len);
/* scalar product */
SRSRAN_API void srsran_vec_sc_prod_cfc(const cf_t* x, const float h, cf_t* z, const uint32_t len);
SRSRAN_API void srsran_vec_sc_prod_fcc(const float* x, const cf_t h, cf_t* z, const uint32_t len);
SRSRAN_API void srsran_vec_sc_prod_ccc(const cf_t* x, const cf_t h, cf_t* z, const uint32_t len);
SRSRAN_API void srsran_vec_sc_prod_fff(const float* x, const float h, float* z, const uint32_t len);
SRSRAN_API void srsran_vec_convert_fi(const float* x, const float scale, int16_t* z, const uint32_t len);
SRSRAN_API void srsran_vec_convert_conj_cs(const cf_t* x, const float scale, int16_t* z, const uint32_t len);
SRSRAN_API void srsran_vec_convert_if(const int16_t* x, const float scale, float* z, const uint32_t len);
SRSRAN_API void srsran_vec_convert_fb(const float* x, const float scale, int8_t* z, const uint32_t len);
SRSRAN_API void srsran_vec_lut_sss(const short* x, const unsigned short* lut, short* y, const uint32_t len);
SRSRAN_API void srsran_vec_lut_bbb(const int8_t* x, const unsigned short* lut, int8_t* y, const uint32_t len);
SRSRAN_API void srsran_vec_lut_sis(const short* x, const unsigned int* lut, short* y, const uint32_t len);
/* vector product (element-wise) */
SRSRAN_API void srsran_vec_prod_ccc(const cf_t* x, const cf_t* y, cf_t* z, const uint32_t len);
SRSRAN_API void srsran_vec_prod_ccc_split(const float* x_re,
const float* x_im,
const float* y_re,
const float* y_im,
float* z_re,
float* z_im,
const uint32_t len);
/* vector product (element-wise) */
SRSRAN_API void srsran_vec_prod_cfc(const cf_t* x, const float* y, cf_t* z, const uint32_t len);
/* conjugate vector product (element-wise) */
SRSRAN_API void srsran_vec_prod_conj_ccc(const cf_t* x, const cf_t* y, cf_t* z, const uint32_t len);
/* real vector product (element-wise) */
SRSRAN_API void srsran_vec_prod_fff(const float* x, const float* y, float* z, const uint32_t len);
SRSRAN_API void srsran_vec_prod_sss(const int16_t* x, const int16_t* y, int16_t* z, const uint32_t len);
// Negate sign (scrambling)
SRSRAN_API void srsran_vec_neg_sss(const int16_t* x, const int16_t* y, int16_t* z, const uint32_t len);
SRSRAN_API void srsran_vec_neg_bbb(const int8_t* x, const int8_t* y, int8_t* z, const uint32_t len);
SRSRAN_API void srsran_vec_neg_bb(const int8_t* x, int8_t* z, const uint32_t len);
/* Dot-product */
SRSRAN_API cf_t srsran_vec_dot_prod_cfc(const cf_t* x, const float* y, const uint32_t len);
SRSRAN_API cf_t srsran_vec_dot_prod_ccc(const cf_t* x, const cf_t* y, const uint32_t len);
SRSRAN_API cf_t srsran_vec_dot_prod_conj_ccc(const cf_t* x, const cf_t* y, const uint32_t len);
SRSRAN_API float srsran_vec_dot_prod_fff(const float* x, const float* y, const uint32_t len);
SRSRAN_API int32_t srsran_vec_dot_prod_sss(const int16_t* x, const int16_t* y, const uint32_t len);
/* z=x/y vector division (element-wise) */
SRSRAN_API void srsran_vec_div_ccc(const cf_t* x, const cf_t* y, cf_t* z, const uint32_t len);
SRSRAN_API void srsran_vec_div_cfc(const cf_t* x, const float* y, cf_t* z, const uint32_t len);
SRSRAN_API void srsran_vec_div_fff(const float* x, const float* y, float* z, const uint32_t len);
/* conjugate */
SRSRAN_API void srsran_vec_conj_cc(const cf_t* x, cf_t* y, const uint32_t len);
/* average vector power */
SRSRAN_API float srsran_vec_avg_power_cf(const cf_t* x, const uint32_t len);
SRSRAN_API float srsran_vec_avg_power_sf(const int16_t* x, const uint32_t len);
SRSRAN_API float srsran_vec_avg_power_bf(const int8_t* x, const uint32_t len);
/* Correlation between complex vectors x and y */
SRSRAN_API float srsran_vec_corr_ccc(const cf_t* x, cf_t* y, const uint32_t len);
/* return the index of the maximum value in the vector */
SRSRAN_API uint32_t srsran_vec_max_fi(const float* x, const uint32_t len);
SRSRAN_API uint32_t srsran_vec_max_abs_fi(const float* x, const uint32_t len);
SRSRAN_API uint32_t srsran_vec_max_abs_ci(const cf_t* x, const uint32_t len);
/*!
* Quantizes an array of floats into an array of 16-bit signed integers. It is
* ensured that *-inf* and *inf* map to -32767 and 32767, respectively (useful
* when quantizing on less than 16 bits).
* \param[in] in Real values to be quantized.
* \param[out] out Quantized values.
* \param[in] gain Quantization gain, controls the output range.
* \param[in] offset Quantization offset, for asymmetric quantization.
* \param[in] clip Saturation value.
* \param[in] len Number of values to be quantized.
*/
SRSRAN_API void srsran_vec_quant_fs(const float* in, int16_t* out, float gain, float offset, float clip, uint32_t len);
/*!
* Quantizes an array of floats into an array of 8-bit signed integers. It is
* ensured that *-inf* and *inf* map to -127 and 127, respectively (useful
* when quantizing on less than 8 bits).
* \param[in] in Real values to be quantized.
* \param[out] out Quantized values.
* \param[in] gain Quantization gain, controls the output range.
* \param[in] offset Quantization offset, for asymmetric quantization.
* \param[in] clip Saturation value.
* \param[in] len Number of values to be quantized.
*/
SRSRAN_API void srsran_vec_quant_fc(const float* in, int8_t* out, float gain, float offset, float clip, uint32_t len);
/* quantify vector of floats or int16 and convert to uint8_t */
SRSRAN_API void srsran_vec_quant_fuc(const float* in,
uint8_t* out,
const float gain,
const float offset,
const uint8_t clip,
const uint32_t len);
SRSRAN_API void srsran_vec_quant_fus(const float* in,
uint16_t* out,
const float gain,
const float offset,
const uint16_t clip,
const uint32_t len);
SRSRAN_API void srsran_vec_quant_suc(const int16_t* in,
uint8_t* out,
const float gain,
const float offset,
const uint8_t clip,
const uint32_t len);
SRSRAN_API void srsran_vec_quant_sus(const int16_t* in,
uint16_t* out,
const float gain,
const float offset,
const uint16_t clip,
const uint32_t len);
/* magnitude of each vector element */
SRSRAN_API void srsran_vec_abs_cf(const cf_t* x, float* abs, const uint32_t len);
SRSRAN_API void srsran_vec_abs_square_cf(const cf_t* x, float* abs_square, const uint32_t len);
/**
* @brief Extracts module in decibels of a complex vector
*
* This function extracts the module in decibels of a complex array input. Abnormal absolute value inputs (zero,
* infinity and not-a-number) are set to default_value outputs.
*
* Equivalent code:
* for (int i = 0; i < len; i++) {
* float mag = x[i];
*
* // Check boundaries
* if (isnormal(mag)) {
* // Avoid infinites and zeros
* abs[i] = 20.0f * log10f(mag);
* } else {
* // Set to default value instead
* abs[i] = default_value;
* }
* }
*
* @param x is the input complex vector
* @param default_value is the value to use in case of having an abnormal absolute value.
* @param abs is the destination vector
* @param len is the input and output number of samples
*
*/
SRSRAN_API void srsran_vec_abs_dB_cf(const cf_t* x, float default_value, float* abs, const uint32_t len);
/**
* @brief Extracts argument in degrees from a complex vector
*
* This function extracts the argument from a complex vector. Infinity and not-a-number results are set to
* default_value.
*
* Equivalent code:
* for(int i = 0; i < len; i++) {
* arg[i] = cargf(x[i]) * (180.0f / M_PI);
*
* if (arg[i]!=0.0f && !isnormal(arg[i])) {
* arg[i] = default_value;
* }
* }
*
* @param x is the input complex vector
* @param default_value is the value to use in case of having an abnormal result.
* @param arg is the destination vector
* @param len is the input and output number of samples
*
*/
SRSRAN_API void srsran_vec_arg_deg_cf(const cf_t* x, float default_value, float* arg, const uint32_t len);
SRSRAN_API float srsran_mean_arg_cf(const cf_t* x, uint32_t len);
SRSRAN_API void srsran_vec_interleave(const cf_t* x, const cf_t* y, cf_t* z, const int len);
SRSRAN_API void srsran_vec_interleave_add(const cf_t* x, const cf_t* y, cf_t* z, const int len);
SRSRAN_API void srsran_vec_gen_sine(cf_t amplitude, float freq, cf_t* z, int len);
SRSRAN_API void srsran_vec_apply_cfo(const cf_t* x, float cfo, cf_t* z, int len);
SRSRAN_API float srsran_vec_estimate_frequency(const cf_t* x, int len);
#ifdef __cplusplus
}
#endif
#endif // SRSRAN_VECTOR_H