Additionally, the proposed KNN classifier has been compared with existing models to ascertain its superiority and proclaim its state-of-the-art performance. We found that the model which is trained using the augmented dataset and additionally includes the absolute DFT values pertaining to the secondary harmonic frequency values of the eight fundamental DTMF frequencies as the features, achieved the best performance with a macro classification F1 score of 0.980835, a 5-fold stratified cross-validation accuracy of 98.47% and test dataset detection accuracy of 98.1053%. These models are validated with an unseen test data set created to simulate real-time noise as observed in telecommunication channels. The proposed KNN classifier model is configured in four different manners which differ in being trained with or without augmented data, as well as, with or without inclusion of secondary harmonic frequency DFT coefficient values as features. The features required for training the proposed KNN classifier are extracted using Goertzel’s Algorithm employed to estimate the absolute Discrete Fourier Transform (DFT) coefficient values for the fundamental DTMF frequencies with or without their secondary harmonic frequencies. In this research article, a novel machine learning based approach to detect DTMF tones perturbed by noise, frequency and time variations by employing the K-Nearest Neighbour (KNN) algorithm is proposed. Owing to the continuous and rapid evolution of telecommunication equipment, the demand for a more efficient and noise-robust detection of Dual-tone multi-frequency (DTMF) signals is conspicuous.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |