This talk investigates the mean-square stability and stabilizability of a networked SISO feedback system over a communication channel with random transmission delay and packet dropout. The transmission delay for each data is an integer step within a given bounded range. A sequence of transmission delays for all transmitted data is an independent and identically distributed (i.i.d.) stochastic process and the probability mass function (PMF) for the transmission delays and packet dropout is known. A nominal closed-loop system is obtained of the networked feedback system, which consists of the plant and controller in the system as well as the channel in mean sense. The other is a stochastic channel uncertainty which is with a random finite impulse response. Each element in the impulse response is with zero mean. It is shown that this stochastic channel uncertainty is a colored multiplicative noise and the networked system is a feedback loop comprised by the nominal closed-loop system and this noise. The ratio between the spectral density factorization of the multiplicative noise and the transfer function of the channel in mean sense is called a colored signal-to-noise ratio. A necessary and sufficient condition is found for the mean-square stability of the network feedback system with a given controller, which is determined by the H2-norm of the product between the nominal system’s transfer function and the colored signal-to-noise ratio. Moreover, the mean-square stabilizable condition of the networked system is presented and the interaction between the colored signal-to-noise ratio and the plant’s unstable poles is studied in the mean-square stabilization problem. According to the mean-square stabilizable condition, an optimal output feedback controller is designed in the mean-square stabilization for the networked system.