Research on a method of high precision frequency measurement based on coordinate rotation digital computer algorithm and Kalman filter

Abstract

Frequency measurement is one of the key techniques in high-precision data acquisition technology of broadband signals. Generally, frequency measurement not only needs to deal with a large amount of data processing but also requires a high precision, but these two aspects are sometimes difficult to reconcile. Some algorithms are overly dependent on the accuracy of the to-be-measured data, which might not be the desired option for real projects since it is almost impossible to get ideal error-free data. This article adopts a frequency measurement method based on the coordinate rotation digital computer algorithm, differential algorithm, and Kalman filter. The use of these algorithms for the frequency measurement process would not only simplify the calculation but also reduce the effect of the measurement error. This method can measure all signals that satisfy the sampling theorem and can also measure multi-channel parallel signals. The experimental results of data simulation and actual measurement on the hardware platform show that the accurate frequency measurement algorithm has a strong data processing ability, stable measurement, and steady improvement in the accuracy of measurement results, which can meet the needs of most instruments for accurate frequency measurement. The measurement error could be reduced to the percentile by the Kalman filter and could be reduced to below the thousandth by the combining the algorithms.