We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
New Structure of CCR with an AOANN Threshold.
- Authors
Rabehi, Abdelhalim; Djebbari, Ali; Hafaifa, Ahmed; Souahlia, Abdelkerim; Taleb-Ahmed, Abdelmalik
- Abstract
In this paper, artificial neural network-based adaptive optimal threshold estimation for a two-dimensional optical code division multiple access conventional correlation receiver is proposed. A multilayer perceptron neural network with back-propagation learning algorithm is considered. This estimator uses the weight (w) and the length (F) of the code word, the number of active users (Ν) and the signal to noise ratio as inputs to estimate the required optimal threshold. We have evaluated the proposed approach on a data set of 46,200 samples. We have found that it gives accurate results: 0.029 for the root mean square error, 0.37% for the relative root mean square error and 99.984% for the correlation coefficient (R), which reflects the efficiency of the proposed optimal threshold estimator.
- Subjects
CODE division multiple access; STANDARD deviations; SIGNAL-to-noise ratio; TWO-dimensional bar codes; ODDS ratio; MACHINE learning; MULTILAYER perceptrons
- Publication
Journal of Optical Communications, 2021, Vol 42, Issue 1, p103
- ISSN
0173-4911
- Publication type
Article
- DOI
10.1515/joc-2018-0028