Resumen
The subject matter of article is the process of measuring the parameters of microwave signals and tracts. The goal of work is implementation on FPGA of a multi-robe method for measuring parameters of signals and tracts with an increasing of accuracy due to optimal processing of information received from sensors. The following tasks have been solved in the article: creation of a model of a multiprobe microwave multimeter with an redundant number of sensors using Kalman filtering in the method of multiprobe microwave measurements and its implementation on field-programmable gate array (FPGA). The following methods were used: linear algebra when defining intermediate variables from the inverse matrix of the system of equations describing a multiprobe system when creating a sensor signal processing algorithm for indirect measuring the power and reflection coefficient from sensor signals, estimation theory when filtering intermediate variables, where the sum main diagonal elements of the variance and covariance matrix is used as an objective function, the smaller this sum, the smaller the error, by analogy with the least squares method, where the D-optimal experiment design minimizes the product of the elements of the main diagonal of the variance and covariance matrix, because the product of the elements of the main diagonal introduces the main contribution to the calculation of the determinant for negligible off-diagonal elements of the variance and covariance matrix, in turn, the determinant of the variance and covariance matrix is visualized by the scattering ellipsoid, the smaller which is, the more accurate the measurement. The following results were obtained a mathematical model of a multiprobe method for measuring parameters of signals and microwave tracts, based on the conversion of signals from sensors located along the direction of power transmission in the tract into incident, reflected and passing power and a complex reflection coefficient of the termination, which differs by filtering intermediate variables, which made it possible to improve accuracy; a finite state machine (FSM) was proposed with states such as forecasting and updating the Kalman filter algorithm and its modeling using FPGA. Conclusions: improvement of signal processing of multiprobe microwave multimeter sensors has improved the measurement accuracy.