A Genetic algorithm for the optimal design of a multistage amplifier

Abstrak

The optimal sizing of analog circuits is one of the most complicated processes, because of the number of variables taken into, to the number of required objectives to be optimized and to the constraint functions restrictions. The aim is to automate this activity in order to accelerate the circuits design and sizing. In this paper, we deal with the optimization of the three stage bipolar transistor amplifier performances namely the voltage gain (AV), the input impedance (ZIN), the output impedance (ZOUT), the power consumption (P) and the low and the high cutoff frequency (FL,FH), through the Genetic Algorithm (GA). The presented optimization problem is of multi-dimensional parameters, and the trade-off of all parameters. In fact, the passive components (Resistors and Capacitors) are selected from manufactured constant values (E12, E24, E48, E96, E192) for the purpose of reduce the cost of design; also, the intrinsic parameters of transistors (hybrid parameters and the junction capacitances) are considered variables in order not to be limited in design. SPICE simulation is used to validate the obtained result/performances.