This Paper's Primary Objective Is to Investigate and Describe the Several Deep Learning
Approaches Currently Used For Picture Steganography. Traditional Approaches, the Most Common Deep
Learning Approaches For Photo Steganography Are Convolutional Neural Network (Cnn) Based Methods
And Generative Adversarial Network (Gan) Based Methods. the Authors of This Work Set Out to Aid Their
Fellow Researchers By Gathering Pertinent Data on the Most Recent Developments, Difficulties, And
Potential Future Directions In This Area. the Pictures to Be Concealed Are Embedded By First Transforming
The Cover Image to Luminance and Chrominance Components. an Excellent Example of the Class Of
Permutation-Based Algorithms That May Better Survive Channel Degradations Is the Chaotic Baker Map,
Which Is Used to Encrypt the Secret Pictures. In This Study, an Ofdm System With Channel Equalization
Was Used For Wireless Communication.