vulcan::rng Namespace Reference

Classes
class	RNG
	Vulcan Random Number Generator. More...

Functions
template<typename RNGType>
allan::IMUNoiseInput< double >	generate_imu_noise (RNGType &rng)
	Generate IMU noise input for Allan variance model.
template<typename RNGType>
Eigen::Vector3d	generate_noise3 (RNGType &rng)
	Generate 3-axis white noise.
template<int N, typename RNGType>
Eigen::Vector< double, N >	generate_noiseN (RNGType &rng)
	Generate N-dimensional white noise.
template<int N, typename RNGType>
Eigen::Vector< double, N >	generate_correlated_noise (RNGType &rng, const Eigen::Matrix< double, N, N > &L)
	Generate correlated noise with given Cholesky factor.
template<typename RNGType>
Eigen::VectorXd	generate_correlated_noise (RNGType &rng, const Eigen::MatrixXd &L)
	Generate correlated noise for dynamic-size matrices.
template<typename RNGType>
void	generate_turbulence_noise (RNGType &rng, double &noise_u, double &noise_v, double &noise_w)
	Generate turbulence forcing noise (3 independent channels).
uint64_t	hardware_seed ()
	Generate seed from std::random_device.
std::seed_seq	create_seed_seq (std::initializer_list< uint32_t > values)
	Create seed sequence from multiple values.
uint64_t	seed_from_string (std::string_view name)
	Hash-based seed from string.
uint64_t	stream_seed (uint64_t base_seed, uint64_t stream_index)
	Combine base seed with stream index for parallel streams.
uint64_t	advance_seed (uint64_t seed, uint64_t n)
	Advance seed to skip N sequences.

Function Documentation

advance_seed()

uint64_t vulcan::rng::advance_seed ( uint64_t seed, uint64_t n )

inline

Advance seed to skip N sequences.

Useful for checkpointing: compute the seed that would result from drawing N samples.

Parameters

seed	Current seed
n	Number of positions to advance

Returns

New seed value

create_seed_seq()

std::seed_seq vulcan::rng::create_seed_seq ( std::initializer_list< uint32_t > values )

inline

Create seed sequence from multiple values.

Provides robust initialization for MT19937 which benefits from multiple seed values to initialize its large state.

Parameters

values

Initializer list of seed values

Returns

seed_seq for RNG initialization

generate_correlated_noise() [1/2]

template<int N, typename RNGType>

Eigen::Vector< double, N > vulcan::rng::generate_correlated_noise	(	RNGType &	rng,
		const Eigen::Matrix< double, N, N > &	L )

Generate correlated noise with given Cholesky factor.

For covariance matrix P, compute L = chol(P) and pass L. Returns L * w where w ~ N(0, I).

Template Parameters

N	Vector dimension
RNGType	Any type with gaussian() method

Parameters

rng	Random number generator
L	Lower-triangular Cholesky factor of covariance

Returns

Correlated noise vector ~ N(0, P)

generate_correlated_noise() [2/2]

template<typename RNGType>

Eigen::VectorXd vulcan::rng::generate_correlated_noise	(	RNGType &	rng,
		const Eigen::MatrixXd &	L )

Generate correlated noise for dynamic-size matrices.

Template Parameters

RNGType

Any type with gaussian() method

Parameters

rng	Random number generator
L	Lower-triangular Cholesky factor of covariance (n×n)

Returns

Correlated noise vector ~ N(0, P) (n×1)

generate_imu_noise()

template<typename RNGType>

allan::IMUNoiseInput< double > vulcan::rng::generate_imu_noise

(

RNGType &

rng

)

Generate IMU noise input for Allan variance model.

Fills all 18 noise channels (3 axes × 2 sensors × 3 noise types) with independent N(0,1) samples.

Example:

vulcan::rng::RNG rng(42);
auto state = vulcan::allan::init_state<double>();
auto coeffs = vulcan::allan::consumer_grade_coeffs(0.01);
 
auto input = vulcan::rng::generate_imu_noise(rng);
auto noise = vulcan::allan::step(state, coeffs, input);

Template Parameters

RNGType

Any type with gaussian() method returning double

Parameters

rng	Random number generator

Returns

IMUNoiseInput with all channels filled

generate_noise3()

template<typename RNGType>

Eigen::Vector3d vulcan::rng::generate_noise3

(

RNGType &

rng

)

Generate 3-axis white noise.

Template Parameters

RNGType

Any type with gaussian() method

Parameters

rng	Random number generator

Returns

3D vector of independent N(0,1) samples

generate_noiseN()

template<int N, typename RNGType>

Eigen::Vector< double, N > vulcan::rng::generate_noiseN

(

RNGType &

rng

)

Generate N-dimensional white noise.

Template Parameters

N	Vector dimension
RNGType	Any type with gaussian() method

Parameters

rng	Random number generator

Returns

N-dimensional vector of independent N(0,1) samples

generate_turbulence_noise()

template<typename RNGType>

void vulcan::rng::generate_turbulence_noise	(	RNGType &	rng,
		double &	noise_u,
		double &	noise_v,
		double &	noise_w )

Generate turbulence forcing noise (3 independent channels).

For Dryden/von Kármán turbulence filters.

Template Parameters

RNGType

Any type with gaussian() method

Parameters

rng	Random number generator
noise_u	Output: longitudinal noise
noise_v	Output: lateral noise
noise_w	Output: vertical noise

hardware_seed()

uint64_t vulcan::rng::hardware_seed ( )

inline

Generate seed from std::random_device.

Use only for initial seeding. The returned seed should be logged for reproducibility in future runs.

Warning

Not guaranteed to be truly random on all platforms. Some implementations may fall back to PRNG.

Returns

Hardware-generated seed value

seed_from_string()

uint64_t vulcan::rng::seed_from_string ( std::string_view name )

inline

Hash-based seed from string.

Useful for named simulation runs, e.g., seed_from_string("monte_carlo_run_42") Uses std::hash which provides reasonable distribution.

Parameters

name	String identifier for the simulation

Returns

Deterministic seed value

stream_seed()

uint64_t vulcan::rng::stream_seed ( uint64_t base_seed, uint64_t stream_index )

inline

Combine base seed with stream index for parallel streams.

Uses golden ratio hashing for good distribution of stream seeds. Each stream will produce an independent sequence.

Example:

uint64_t base = 42;
for (int i = 0; i < 100; ++i) {
    RNG stream_rng(stream_seed(base, i));
    // Independent simulation...
}

Parameters

base_seed	Base seed for all streams
stream_index	Index of this particular stream

Returns

Unique seed for this stream