A basic premise behind modern secure computation is the demand for lightweight cryptographic primitives, like identifier or key generator. From a circuit perspective, the development of cryptographic modules has also been driven by the aggressive scalability of complementary metal-oxide-semiconductor (CMOS) technology. While advancing into nano-meter regime, one significant characteristic of today's CMOS design is the random nature of process variability, which limits the nominal circuit design. With the continuous scaling of CMOS technology, instead of mitigating the physical variability, leveraging such properties becomes a promising way. One of the famous products adhering to this double-edged sword philosophy is the Physically Unclonable Functions (PUFs), which extract secret keys from uncontrollable manufacturing variability on integrated circuits (ICs). However, since PUFs take advantage of microscopic process variations, thus many specialized issues including variability, modeling attacks and noise sensitivity need to be considered and addressed. In this dissertation, we present our recent work on PUF based secure computation from three aspects: variability, modeling and noise sensitivity, which are deemed the foundations of our study. Moreover, we found that the three factors coordinate with each other in our study, for example, the modeling technique can be utilized to improve the unsatisfied reliability caused by noise sensitivity, quantifying the variability can effectively eliminate the impact from noise, and modeling can help with characterizing the physical variability precisely.