• Navya Holla K.
• Deon Gabriel Dsouza
• MD Sohail Shaikh
• Sai Nath Das
• Raghav Gupta

Deep learning techniques are applied to detect MIMO-DCSK signals in highly dynamic environments. The complexity of signal detection increases with the number of antennas, necessitating efficient techniques. By training deep neural networks (DNNs) on labeled MIMO-DCSK signals, complex mappings between received and transmitted symbols are learned. Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) extract spatial and temporal dependencies, enabling robust detection in fading and interference. Techniques like data augmentation, regularization, and transfer learning improve generalization. Deep learning outperforms traditional methods, exhibiting better BER and symbol detection accuracy, as well as increased robustness against fading and interference. Integration of deep learning enhances MIMO-DCSK-based communication systems, improving reliability and capacity for future wireless networks.

Massive MIMO, Non-coherent Detection, Differential Detection, Deep-learning, Neural Networks.

"Comparison Between ML, MSME And Deep Learning Techniques For MIMO-DCSK Signals", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 5, page no.o228-o234, May-2023, Available :http://www.jetir.org/papers/JETIR2305F27.pdf

"Comparison Between ML, MSME And Deep Learning Techniques For MIMO-DCSK Signals", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 5, page no. ppo228-o234, May-2023, Available at : http://www.jetir.org/papers/JETIR2305F27.pdf