Due to the various features of fiber-optic cables, the fields of use vary from one type of cable to another. Therefore, identifying the cable is important before its use. Aging or other factors can make cables inaccessible, such as underground cables that can lose their markings or have torn or worn labels. By analyzing the behaviors of a particular fiber type, it is possible to classify using machine learning algorithms such as K-Nearest Neighbors (KNN). Meanwhile, the possibilities of quantum computing are still being explored in terms of advantages and limitations. The idea of making the best use of both machine learning and quantum computing is the foundation for this work. So, the problem statement of the current project was to first implement the classical KNN algorithm and then develop an efficient KNN algorithm with the support of quantum technology. Finally, the classical and quantum-based KNN algorithms are compared in terms of time complexity and prediction accuracy. During the project, it was found that the quantum KNN algorithm can achieve the same accuracy as conventional algorithms, along with reducing the complexity, i.e., classical version has log complexity whereas quantum version has quadratic complexity.
Application of Quantum-Based K-Nearest Neighbors Algorithm in Optical Fiber Classification
Lect. Notes Electrical Eng.
International Conference on Robotics, Control, Automation and Artificial Intelligence ; 2022 November 24, 2022 - November 26, 2022
2023-11-18
14 pages
Article/Chapter (Book)
Electronic Resource
English
Non-Data-Aided k-Nearest Neighbors Technique for Optical Fiber Nonlinearity Mitigation
British Library Online Contents | 2018
|A Fast Algorithm for Finding k-Nearest Neighbors with Non-Metric Dissimilarity
British Library Conference Proceedings | 2002
|