deterministicGauss1​D() – Deterministic Gaussian Samples

Optimally placed samples of the standard normal density in the scalar case

samples = deterministicGauss1D( L )

Deterministic Sampling of Standard Normal Gaussian Distribution

Advantages of deterministic sampling vs random sampling like randn()
- Reproducible results
- Samples are optimally placed
- Less samples needed for same quality results
- Methods may fail occassionally due to poor choice of random samples

Input
- L : number of samples

Output
- samples : (L x 1) vector with 1D sample locations

Example : Get 7 deterministic samples with a standard deviation of 3 and a mean of 5.
>> samples = deterministicGauss1D(5)*3 + 5;

https://isas.iar.kit.edu/Publications.php : Theory of deterministic sampling in general
https://nonlinearestimation.bitbucket.io/ : Deterministic Gaussian sampling in higher dimensions
https://github.com/libDirectional/ : Deterministic sampling in non-Euclidean manifolds