Implementation of nonlinear blind source separation for CHEMFET sensor arrays

In this study, a method to improve selectivity of chemically field-effect transistor (CHEMFET) sensor towards the main ion concentration in mixed solution is discussed. The approach is based on artificial neural network (ANN) as a post processing stage that performs the estimation of ion concentration in a mixed solution. CHEMFET sensor is viewed as non-linear model producing signal fed to blind-source separation algorithm. To describe how the ions interfere with main ion, the source signal of CHEMFET sensor is generated based on CHEMFET model. The sensor response is converted to frequency by using voltage to frequency converter (VFC). Simulation results confirm that the algorithm is able to separate the mixing signal.