GAN-IoTVS: A Novel Internet of Multimedia Things-Enabled Video Streaming Compression Model Using GAN and Fuzzy Logic

This study proposes a novel method for streaming data compression and encoding protocols based on generative adversarial networks (GANs) and fuzzy logic. The concept of GAN and fuzzy logic integration creates a different prospect as compared to the existing multimedia streaming data compression and security protocols. This article proposed a collaborative model for multimedia data compression by integrating GAN with fuzzy logic. It guarantees data compression using a secure channel for the verification platform of multimedia-enabled videos. Thus, it defines a fuzzy-based weight function for controlling the frequency of the current objects by the contents that effectively handle a string of code generation, which is designed and developed. It also eliminates adversarial loss removes streaming data redundancy and saves resource constraints using static–dynamic memory, transmission bandwidth, and computational power. In addition, this proposed model reduces the number of lossless video files before encryption by increasing the entropy of the sequence of coded images. Fuzzy logic code provides a more secure multimedia information mapping on the code contents. Substantially, the private key is generated when multimedia-enabled video files are embedded and sent to the receiver, and decoding information is possible using the same key. To illustrate the simulation of the theoretical results and model experiments are performed under the Python 3.9 tool. The experimental results show that a better compression ratio is achieved compared with the other state-of-the-art models. There is no difference between the original video and the compression after decoding, and the rate is increased to over 30.13%.