Triple-Sensing with An Ion-Sensitive-Field-Effect-Transistor via Machine-Learning Algorithm

Internet-of-thing (IoT) has driven a paradigm shift of the next-generation sensor architecture. With edge-computing capabilities, it can reduce high-cost sensing devices with low-cost computations. To match up this paradigm shift, single sensing device with multiple sensing capabilities would be the future trend of sensor technologies. Based on machine learning algorithms, in this work, we demonstrate an active-controlled ion-sensitive-field-effect-transistor (ISFET) can sense pH (5 similar to 9), illumination (0 mu W/cm(2) similar to 750 mu W/cm(2)), and temperature (22 degrees C similar to 52 degrees C) simultaneously. Utilizing the understanding of the sensing device and experimental data analysis, we interpret the sensing system as a transformation between physical variable space (PVS) and virtual sensor space (VSS). With the active-controlled ISFET, the transformation can be learned and utilized as a multi-target sensing device. Due to advantages of low power and small size, the developed multi-sensing device has a good potential for IoT applications.