CTLE Adaptation Using Deep Learning in High- speed SerDes Link

To speed up a serial link simulation, it is critical to model the Serializer/Deserializer (SerDes) circuit behavior accurately. In this research, we focus on building a model for high-speed SerDes receiver CTLE adaptation behavior, which has a fast simulation speed and high-precision prediction. The proposed modeling method doesn't need any substantial domain knowledge. Deep neural networks model will be used to mimic the behavior of the CTLE adaptation process in the receiver. The proposed modeling method shows high correlations with the CTLE adaptation codes.