Name
Abstract | ||
|---|---|---|
The need for generating samples that approximate statistical distributions within reasonable error limits and with less computational cost, necessitates the search for alternatives. In this work, we focus on the approximation of Gaussian distribution using the convolution of integer sequences. The results show that we can approximate Gaussian profile within 1% error. Though Bessel function based discrete kernels have been proposed earlier, they involve computations on real numbers and hence increasing the computational complexity. However, the integer sequence based Gaussian approximation, discussed in this paper, offer a low cost alternative to the one using Bessel functions. |
Year | DOI | Venue |
|---|---|---|
2014 | 10.1007/978-3-319-04960-1_19 | ADVANCES IN SIGNAL PROCESSING AND INTELLIGENT RECOGNITION SYSTEMS |
Field | DocType | Volume |
Discrete mathematics,Applied mathematics,Convolution,Gaussian,Probability distribution,Real number,Gaussian function,Mathematics,Bessel function,Integer sequence,Computational complexity theory | Conference | 264 |
ISSN | Citations | PageRank |
2194-5357 | 0 | 0.34 |
References | Authors | |
2 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Arulalan M. Rajan | 1 | 1 | 0.70 |
| Ashok Rao | 2 | 200 | 19.14 |
| R. Vittal Rao | 3 | 0 | 0.34 |
| H. S. Jamadagni | 4 | 160 | 30.14 |