Statistical simulation of signal-to-noise ratio estimation

Abstract

The research addresses an urgent issue of operational control of the signal-to-noise ratio in radio-electronic systems. The proposed signal-to-noise ratio estimation technique is based on the maximum likehood estimation of the signal amplitude when observed against the white Gaussian noise. The paper presents block diagrams of sumilation in Simulink, which implement the proposed technique, in particular, a module for calculating the maximum likehood estimation of the signal amplitude, which involves quadrature heterodyning of the input signal, accumulation of the readings of the obtained components in the observation interval, and calculation of the square root of the sum of squares of in-phase and quadrature correlations. The statistical simulation (1,000 tests) was performed to verify the accuracy of the proposed technique. It was established that the mean-squared deviation of the signal-tonoise ratio estimation error is approximately 0.1 dB in the case of unlimited (by width) input and reference signals. With requirements to signal width reduced, or with sign approximation of input and reference signals, the mean-squared deviation of the signal-to-noise ratio estimation error does not exceed 0.2 dB. The results obtained confirm high accuracy and feasibility of the proposed estimator. The materials of this paper can be of use in the operative quality control of space communication channels.

Keywords: signal-to-noise ratio, signal amplitude estimation, white Gaussian noise, simulation model, statistical simulation