Self-Heating of FinFET Circuitry Simulated by Multi-Correlated Recurrent Neural Networks

A series of multi-correlated recurrent neural networks is used to predict the relative temperature of inverter chains folded in 3 rows. The circuit hotspot temperature is predicted by a fully connected neural network. The correlated recurrent neural networks trained by the SPICE data within 17 stages can predict T up to 37 stages (2.2x SPICE complexity) with the error as low as 0.9 degrees C, outperforming the previous fully connected neural network (1.9x SPICE, 3 degrees C error) and non-correlated recurrent neural network (2.2x SPICE, 3 degrees C error) by considering the thermal coupling between rows. The precise prediction of temperature profiles and hotspot positions indicate that the thermal physics is learned by correlated recurrent neural networks. Therefore, an 82-stage folded inverter chain can be predicted and optimized confidently by neural networks, while SPICE can only simulate a 37-stage chain due to the high computational cost. A 100-stage chain is also predicted.

Automated summary

This article uses a series of multi-correlated recurrent neural networks to predict the relative temperature of inverter chains folded in 3 rows, while a fully connected neural network is used to predict the circuit's hotspot temperature. The correlated recurrent neural networks, trained using SPICE data, exhibit better accuracy in temperature prediction by considering the thermal coupling between rows, outperforming previous neural network models.