The recent surge in the internet-of-things envisions the integration of trillions of sensors in various applications such as environmental monitoring. A myriad of parameters such as temperature, pressure, humidity, and proximity are required to be collected simultaneously by resistive (R) or capacitive (C) transducers [1-4]. The traditional sinusoidal-stimulated sensor system exhibits high precision [5] and scales for multiple channels through multiplexing and demultiplexing as shown in Fig. 1, top. However, these systems are not suitable for emerging energy-limited systems and cost-sensitive applications in massive-channel sensing due to: 1) The signal-to-noise ratio (SNR) of the signal chain is directly traded with power consumption and core area, leading to high power and large die area. 2) The multiplexing methods in conventional sensor interface take 2 additional I/O pads to support 1 more sensor, and the large pad frame significantly increases the total chip area.