Optimization of Eye Diagram Characteristics of MLGNR Interconnect Networks Using Fast ML Assisted Evolutionary Algorithm

In this paper, a knowledge based artificial neural network (KBANN) assisted evolutionary algorithm (EA) is presented for optimization of eye diagram characteristics of on-chip multi-layered graphene nanoribbon (MLGNR) interconnect network driven with nanosheet FET (NSFET) inverters. First, a KBANN model is trained to mimic the eye diagram characteristics of the MLGNR interconnect network. The next step is to use particle swarm optimization (PSO) and EA (such as strength pareto evolutionary algorithm (SPEA2)) for optimizing the eye diagram characteristics obtained from the outputs of the KBANN.