Symbol rate estimation method for discontinuous phase 2fsk signal based on fpga

Abstract

The invention discloses a method for estimating the symbol rate of a discontinuous phase 2FSK signal based on FPGA, comprising: acquiring discontinuous phase 2FSK baseband signal data, selecting a carrier frequency and extracting carrier frequency parameters; utilizing DDS, adder and multiplication of FPGA The device IP core performs secondary down-conversion on the I and Q channels of the discontinuous phase 2FSK baseband signal data through the carrier frequency parameters to obtain the corresponding output results; use the FPGA filter IP to check the I channel output results and the Q channel output results Carry out low-pass filtering respectively; use the FFT IP core of FPGA to calculate the envelope power spectrum of the discontinuous phase 2FSK baseband signal according to the output results of the I channel and the Q channel after low-pass filtering; for the discontinuous phase 2FSK baseband signal The envelope power spectrum is subjected to spectral line highlighting processing, the symbol rate spectral line is extracted from the processing result according to the threshold, and the symbol rate is calculated from the symbol rate spectral line according to the current sampling rate and the current FFT point number. The invention greatly reduces the complexity of symbol rate estimation and can save FPGA hardware resources.

Description

Technical field [0001] The present invention relates to the field of electronic information technology, in particular to a symbolic rate estimation method for discontinuous phase 2FSK signals based on an FPGA. Background [0002] Most of the existing 2FSK modulation signal symbol rate estimation algorithms are divided into two categories: the first type, which uses the wavelet transform to extract the local frequency change moment in the signal through the wavelet transform, so as to obtain the symbol rate and the jump time. Some literature proposes that the symbolic jump time is extracted by using the haar wavelet transform twice, and then the symbol rate is obtained by using the Fourier transform, but the error is large under the low signal-to-noise ratio, and the wavelet transform scale has a greater influence on the result and is affected by frequency bias and symbol rate, which makes the wavelet transform difficult to select. On the basis of the above processing process, the...

AI Technical Summary

Problems solved by technology

Some literature proposes to use two haar wavelet transforms to extract the symbol jump time, and then use Fourier transform to obtain the symbol rate, but the error is large under low signal-to-noise ratio, and the wavelet transform scale has a great influence on the result and is affected by frequency offset , symbol rate and other influences make it difficult to select the wavelet transform

On the basis of the above processing process, morlet wavelet is further used to extract the ridge line, and the short-time variance is calculated to extract the discrete spectral line, which improves the anti-noise performance of the algorithm to a certain extent, but there are still difficulties in parameter selection and anti-noise performance. Poor problem, and the implementation of the above algorithms on FPGA is relatively complicated, and consumes more hardware resources

The second category uses the cyclostationary statistical properties of digitally modulated signals and adopts cyclic autocorrelation processing to obtain the peak value containing symbol rate information, and obtain the symbol rate by detecting the peak value. The performance is not stable enough, and the algorithm complexity is very high, which is not suitable for FPGA implementation

Images