Optimization of low noise amplifier designs by genetic algorithms

The genetic algorithms (GAs) are employed as an optimization tool for low noise amplifier (LNA) designs. In the optimization, the input and output matching circuits for an LNA are encoded by a chromosome representation. A fitness function is then defined to quantitatively measure the circuit performances of the LNA. Following the evolving processes of the GAs, the matching circuits can be optimized to obtain a high-performance LNA. To demonstrate the optimization algorithms, 2.4 and 5.2 GHz LNAs are designed and implemented. For both the examples, the GAs take about 70 iterations to acquire the optimal results. In addition, the simulated and measured results show that the obtained LNA designs well satisfy the desired design targets, which validate the capability of GAs in the LNA designs.