Stochastic optimization algorithm and random signal nonlinear optimal filtering algorithm, as the two computing cores of high-performance intelligent calculation in intelligent control, are typical representatives of intelligent optimization algorithms in the rapidly developing scientific research and engineering application technology fields. In order to effectively improve the efficiency of information filtering and optimal quantization modeling of a class of partially observable random signals with massive characteristics in intelligent control, this paper uses stochastic optimal control theory and methods to discuss in depth the mathematical construction of this type of random signal transmission system. Two optimal control mathematical models are obtained via the optimal calculation and random optimization analysis of the allowable coding function and decoding function under this type of signal. It provides an idea mathematical processing method for improving the intelligent control and information management efficiency of the signal transmission system.