38results about How to "Fast Fourier Transform" patented technology

The invention discloses a digital processing method of an altimeter of a linear frequency modulation continuous wave radar, which belongs to the field of the radio altimeters. The digital processing method comprises the following steps: step 1, obtaining beating signals from an orthogonal mixer arranged at the front end of the altimeter of the radar, and carrying out A / D (analog-to-digital) sampling to obtain a digital signal sequence; step 2, performing Fourier transform on the digital signal sequence, and calculating the amplitude of a spectral line; step 3, judging according to the amplitude of the spectral line of an amplitude spectrum, and measuring the beating frequency of the digital signal sequence; step 4, refining the frequency spectrum, calculating the slope correction of saw-tooth wave, and correcting the slope of the voltage signals of the saw-tooth wave; and step 5, calculating the high value of the current altimeter according to the slope of the voltage signals of the saw-tooth wave. The FFT-CZT method adopted by the invention is insensitive to noises and can perfectly work at lower signal-to-noise ratio, but the measurement precision is reduced. Because of having no limit on the properties of the targets, the digital processing method can be applied to the measurement and the analysis of point targets or surface targets.

A polynomial modular multiplication coprocessor based on a lattice-based cryptosystem comprises a first register file, a second register file, a third register file and a fourth register file, and further comprises a multiplexer, an input register file, a fast Fourier transform module and an output register file, wherein the input end of the multiplexer is connected with the third register file, the fourth register file and the output of a modular multiplication inverse element operation unit; the input register file is connected with the multiplexer and an address mapping table; the fast Fourier transform module is connected with the input register file and the output end of the first register file under modular computation; the input end of the output register file is connected with the output end of the fast Fourier transform module under modular computation, and the output end of the output register file is connected with the multiplexer; the output end of a demultiplexer is connected with an output end and a modular multiplication unit, and the output end of the modular multiplication unit is connected with the modular multiplication inverse element operation unit; the input end of the address mapping table is connected with the second register file. According to the coprocessor, an FPGA platform is used for hardware design, and the polynomial modular multiplication coprocessor based on the lattice-based cryptosystem is obtained.

The invention discloses a two-phase flow pressure drop and noise spectrum analysis-based flow pattern discrimination method. The method is used for judging the flow pattern of a two-phase flow in a pipeline. The method comprises the steps of respectively acquiring continuous pressure-drop time-domain signals at the two ends of a measurement section of a pipeline and the two ends of a liquid pump; subjecting the continuous pressure-drop time-domain signals to the pressure-drop high-frequency sampling treatment to obtain pressure-drop time-domain figures; judging the relevance between the pressure-drop time-domain figure at the two ends of the measurement section of a pipeline and the pressure-drop time-domain figure at the two ends of the liquid pump to obtain the flow pattern of the two-phase flow. According to the technical scheme of the invention, the two-phase flow pressure drop and noise spectrum analysis-based flow pattern discrimination method has the advantage of accurate recognition effect.

The invention discloses a microwave frequency modulation thermal wave imaging system and method. The system uses microwaves modulated by the amplitude of the FM signal to heat the object to be inspected, and uses a thermal imager to record the surface temperature of the object to be inspected. Set the frequency modulation signal or the temperature signal of the non-defective area as the reference signal. Perform fast Fourier transform on the temperature signal and the reference signal, extract the amplitude, phase and peak frequency of a specific frequency as eigenvalues; perform cross-correlation on the temperature signal and the reference signal, and extract the correlation coefficient of the specific delay time and the peak delay time As an eigenvalue; perform frequency-domain cross-correlation on the temperature signal and a reference signal, and extract the cross-correlation amplitude and phase of a specific delay time and peak delay time as eigenvalues. Imaging is performed using the eigenvalues ​​of all pixels in the inspected area. The method and system have the advantages of strong anti-interference, rich depth information, short detection time, etc., and can be applied to the fields of non-destructive testing, medical imaging, target detection and the like.

The invention discloses a rapid channel estimation and signal synchronization method suitable for an underwater acoustic modem, and is suitable for the field of real-time online demodulation of the underwater acoustic modem. The underwater acoustic communication modem physical layer data link block pilot signal is designed, the rapid Fourier transform and the subspace projection are utilized to realize the rapid estimation and online decoding process of an underwater acoustic channel, and meanwhile, the periodic autocorrelation characteristic of a pseudorandom sequence is utilized to realize the rapid decorrelation process and realize the physical layer signal synchronization. Compared with a general time domain underwater acoustic channel estimation and signal synchronization method, thealgorithm has the advantages that the algorithm can be rapidly realized by using systems such as a digital signal processor (DSP), a field programmable gate array (FPGA) and the like, and the operation complexity is reduced while the channel estimation is ensured. The method is especially suitable for sparse underwater acoustic channel estimation.