A GAN-based generalized signal encoding processor for IoT applications in 22nm ULL technology

In this project, the chip will implement the newly proposed GAN based generalized signal encoding algorithm for audio/image/video compression. In the algorithm, the compressed signal is represented by a latent vector fed into a GAN which is trained to produce high-quality signals that minimize a target objective function. Therefore, the chip will implement both convolutional neural network (CNN) and generative adversarial network (GAN) for both forward- and backward- propagation.