Multi-bias Small-signal Modelling of Microwave Transistors Using PSO-BP Neural Network

In this paper, a multi-bias small-signal S-parameters modeling method for InP heterojunction bipolar transistor (HBT) devices is explored using a particle swarm algorithm (PSO) to optimize back-propagation neural network (BPNN). To improve the modeling accuracy, eight separate NNs were used to model the real and imaginary parts of the S-parameters respectively. Good agreement was achieved between measured and modeled S-parameters over in a frequency range of 1 to 50 GHz over a wide range of bias points.