Instantaneous frequency measurement method based on Cordic algorithm

Abstract

The invention discloses an instantaneous frequency measurement method based on a Cordic algorithm, and the method comprises the steps: constructing a differential equation through the relation between each signal and an adjacent signal, and the relation of a corresponding differential signal, rapidly solving the result of the differential algorithm by means of a Cordic algorithm kernel in an FPGA, carrying out the calculation and compensation of the signal through the autocorrelation of the signal, and finally carrying out the calculation and compensation of the signal through the sampling characteristics. And solving the signal frequency. According to the method, mathematical changes and the characteristics of signals are ingeniously utilized, and the dependency on a large number of data sources is greatly reduced through the mathematical changes and built-in kernel calculation. Meanwhile, the characteristic derivation of adjacent signals is utilized, the dependence on periodicity or standard is reduced, and the instantaneous frequency measurement of a single-path structure and a multi-path structure can be supported.

Description

technical field [0001] The invention belongs to the technical field of electronic measurement, and more specifically relates to an instantaneous frequency measurement method based on a Cordic algorithm, which is used for instantaneous frequency measurement of an I / Q signal based on a zero-IF architecture. Background technique [0002] With the rapid development and continuous application of the communication field, instantaneous frequency measurement has become a particularly important technical index in the communication system. Whether it is in radar positioning, signal capture, remote sensing measurement, precision instruments, etc., instantaneous frequency measurement plays a key role. At present, the most widely used instantaneous frequency measurement methods mostly adopt the principle of fast Fourier transform (FFT) with mature theory, and calculate through the phase difference characteristics of the signal itself. However, most of these algorithms depend on the accu...

AI Technical Summary

Problems solved by technology

[0005] 1. The demand for data is too large, and a large amount of complete data is required for iteration

If the amount of data provided is insufficient, the results will be greatly affected

[0006] 2. The formula or calculation is too complicated. FPGA requires high timing, and FPGA is composed of a large number of gate calculations. If there are too many too complex formulas or calculations, it is easy to cause confusion in the calculation logic of FPGA.

[0007] 3. Rely on the data of the whole period, or need to clarify the period point. Some algorithms use the periodical transformation characteristics of the signal to calculate, and need to rely on periodical information, but it is difficult to implement in a real-time sampling system

[0008] 4. Too much reliance on the standardization of I / Q two-way signals

In this process, errors will inevitably occur, and the generated two-way signals must be accompanied by amplitude and phase errors. Some algorithms are based on the characteristics that the I / Q two-way signals are completely orthogonal, or the amplitudes are completely equal. These algorithms are in In the actual hardware system, signal deviation is easy to occur

[0009] 5. Multi-channel signal constraints. The current mainstream algorithm is to analyze one signal. However, in the new zero-IF architecture, due to changes in cost and structure, a new split system (4-channel, 8-channel, etc.) is designed. ), there are more and more algorithms for measuring single-channel signals, especially for frequency measurement and calculation, the use in multi-channel systems is very limited

[0010] It can be seen that most of the current traditional instantaneous frequency measurement methods have high requirements for data volume, data integrity, and specificity, especially in the applicability of new structures and the efficiency of algorithms. low efficiency

Images