Model-Based Processing of Microcantilever Sensor Arrays

In this paper, we have developed a model-based processor (MBP) for a microcantilever-array sensor to detect target species in solution. We perform a proof-of-concept experiment, fit model parameters to the measured data and use them to develop a Gauss-Markov simulation. We then investigate two cases of interest, averaged deflection data and multichannel data. For this evaluation we extract model parameters via a model-based estimation, perform a Gauss-Markov simulation, design the optimal MBP and apply it to measured experimental data. The performance of the MBP in the multichannel case is evaluated by comparison to a "smoother" (averager) typically used for microcantilever signal analysis. It is shown that the MBP not only provides a significant gain (~80 dB) in signal-to-noise ratio (SNR), but also consistently outperforms the smoother by 40-60 dB. Finally, we apply the processor to the smoothed experimental data and demonstrate its capability for chemical detection. The MBP performs quite well, apart from a correctable systematic bias error