Receiver Behavior Modeling Based on System Identification
Bowen Li, P. M. Franzen, Yongjin Choi, Christopher Cheng · 2018
In moderm high-speed chip to chip SerDes (Serializer-Deserializer) links, the measured eye diagram at the receiver input is often closed. A receiver behavioral model is needed to predict the inner signal at the output of the receiver. In this paper, receiver modeling based on system identification approach is obtained. The advantage of this approach is that receiver system identification model can be performed only using the input and output time domain signals. In this research, model performances of linear and nonlinear system identification models are presented. System identification models are compared with Recurrent neural networks(RNN) and show better results.