Stochastic computational models for accurate reliability evaluation of logic circuits
Proceedings of the 20th symposium on Great lakes symposium on VLSI
Stochastic decoding of turbo codes
IEEE Transactions on Signal Processing
Comparing the performance of stochastic simulation on GPUs and OpenMP
International Journal of Computational Science and Engineering
Survey of Stochastic Computing
ACM Transactions on Embedded Computing Systems (TECS) - Special Section on Probabilistic Embedded Computing
Hi-index | 0.00 |
To date essentially only two fundamentally different representations of numbers have been used in electronic computers: One is the analog representation by a voltage or current inside a given range, the other one, the digital representation which maps a sequence of 0's and 1's onto a spatial or temporal sequence of voltage or current pulses. Of late, interest has arisen in the use of random pulse sequences as information carriers. (1), (2), (3), (4), (5), (6). It turns out that the use of random pulse sequences leads to the use of digital ANDs or ORs for the fundamental operations of multiplication and summation and therefore to a very considerable reduction in cost of the computational equipment. It is the purpose of this paper to present the theory of these random pulse sequences as well as their practical circuit implementations, and to give some systems design examples. In the final section some non-Von Neumann organizations will be discussed which are made possible by the use of stochastic computing elements.