Deep learning for communication signal classification - part II

Over the past several years, Deep Learning (DL) has been widely regarded as a fundamental technology of the Fourth Industrial Revolution (4IR or Industry 4.0), which encompasses artificial intelligence (AI) and machine learning (ML). In this project, the performance of Convolutional neural networ...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Lew, Mei Tong
مؤلفون آخرون: Alex Chichung Kot
التنسيق: Final Year Project
اللغة:English
منشور في: Nanyang Technological University 2023
الموضوعات:
الوصول للمادة أونلاين:https://hdl.handle.net/10356/167412
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
المؤسسة: Nanyang Technological University
اللغة: English
الوصف
الملخص:Over the past several years, Deep Learning (DL) has been widely regarded as a fundamental technology of the Fourth Industrial Revolution (4IR or Industry 4.0), which encompasses artificial intelligence (AI) and machine learning (ML). In this project, the performance of Convolutional neural network (CNN), Long short-term memory (LSTM), and Convolutional, long short-term memory deep neural network (CLDNN) models towards those standard modulation signals had been analyzed. Furthermore, the relationship between the variety of modulation signals and the validation accuracy of all these three neural network models was obtained at different signal-to-noise ratio (SNR) levels. By providing additional training data to the neural network models, their capacity to accurately classify modulation signals had been enhanced. As a result, both CNN and CLDNN have demonstrated superior performance at high SNR levels, even when faced with a diverse range of modulation signals in the dataset. However, the classification ability of LSTM deteriorates with an increase in the number of modulation signal types present in the dataset. Furthermore, signals with a similar modulation scheme have a higher likelihood of causing misclassification. However, increasing the SNR level of the modulation signals can improve this situation. Therefore, LSTM is the optimal model for dealing with signals with similar modulation schemes.