CN101251390A - A Weak Signal Detection Device Based on Time-Frequency Transformation - Google Patents

Abstract

The invention relates to a weak signal detection device based on time-frequency conversion, which includes a signal channel unit, an analog-to-digital conversion unit, and a processor; a modulated signal to be detected is input to the signal channel unit, and is amplified and filtered by the signal channel unit; The analog-to-digital conversion unit samples the analog signal output by the signal channel unit and then outputs a quantized signal; the processor receives the quantized signal output by the analog-to-digital conversion unit and converts it into a frequency domain signal for power spectrum analysis to obtain a measurement result. Because the present invention uses a processor to convert time-domain signals into frequency-domain signals, detects weak signals in the frequency domain and obtains measurement results, and does not need a device for generating specific reference signals, the hardware structure is simple and the cost is low.

Description

A Weak Signal Detection Device Based on Time-Frequency Transformation

Technical field:

The invention relates to a weak signal detection device, in particular to a weak signal detection device based on time-frequency transformation.

Background technique:

Weak signal detection is a cutting-edge field in measurement technology. Generally, sensors are used for power conversion to convert the detection object into power. When the useful signal is covered by a large amount of noise, the measurement is greatly limited. The main means of weak signal detection is to improve the signal-to-noise ratio of the signal. At present, the method of weak signal detection technology is through the lock-in amplification method, and the lock-in amplification method is divided into a digital lock-in detection method and an analog phase-lock detection method. The device for detecting weak signals by using the analog phase-locking detection method is mainly composed of a signal channel unit, a reference unit, a phase-sensitive detection unit, an analog-to-digital conversion unit, and a processor. The working process of the device is as follows: the signal a to be detected modulated to a fixed frequency is amplified and filtered by the signal channel unit to output signal b; the reference unit outputs signal d with the same frequency and phase as signal b; signal b and signal d are simultaneously sent to the phase sensor The detection unit, the phase-sensitive detection unit filters out the double frequency signal of the modulation frequency, obtains the DC component of the difference frequency and phase to obtain the output signal e, and then obtains the quantized output signal f by sampling the analog-to-digital conversion unit, and the processor reads the analog-to-digital conversion The quantized output signal f of the unit is used to obtain the final measurement result.

Digital phase-locked amplification detection method The device for detecting weak signals is mainly composed of a signal channel unit, a reference unit, an analog-to-digital conversion unit and a processor, and the phase-sensitive detection unit is omitted. The working process of the device is as follows: the signal g to be detected is amplified and filtered by the signal channel unit, and the output signal h is output; the reference signal is phase-shifted by the reference unit to output the signal j, and the signal h and signal j are sent to the analog-to-digital conversion unit, and the analog-to-digital conversion unit After the signal h and signal j are digitized, the output signal k is sent to the processor, and the processor calculates the final measurement result according to the signal h and signal j. In fact, the processor completes the work of phase-sensitive detection.

Although the device for detecting weak signals using digital phase-locking detection method and analog phase-locking detection method can realize the detection of weak signals, the system hardware is relatively complicated. No matter the analog phase-locking method or the digital phase-locking method, not only the processing of the signal channel is required , it is also necessary to generate a specific reference signal through some methods. In addition, the analog phase-locking method also requires a phase-sensitive detector (analog multiplier); the digital phase-locking method requires two analog-to-digital conversions to sample the reference signal and the signal to be detected respectively.

Invention content:

The technical problem to be solved by the present invention is to provide a weak signal detection device based on time-frequency transformation with simple hardware structure and low cost.

The weak signal detection device based on time-frequency transformation of the present invention includes a signal channel unit, an analog-to-digital conversion unit, and a processor; the modulated signal to be detected is input to the signal channel unit, and is amplified and filtered by the signal channel unit; the analog-to-digital conversion unit After sampling the analog signal output by the signal channel unit, the quantized signal is output; the processor receives the quantized signal output by the analog-to-digital conversion unit and converts it into a frequency domain signal for power spectrum analysis to obtain the measurement result.

Based on Parseval's theorem, the total energy of a signal in the time domain is equal to its total energy in the frequency domain, so signal detection in the time domain can be performed in the frequency domain. The signal to be detected is modulated to a fixed frequency, and only the energy of this frequency needs to be obtained in the frequency domain to obtain the desired measurement result. Because the present invention uses a processor to convert time-domain signals into frequency-domain signals, detects weak signals in the frequency domain and obtains measurement results, and does not need a device for generating specific reference signals, the hardware structure is simple and the cost is low.

The processor application software includes a Fourier spectrum analysis module for performing power spectrum analysis on the output signal of the analog-to-digital conversion unit to obtain the relative strength of the signal energy to be detected.

The signal channel unit includes an amplifier circuit, a gain control circuit, and an anti-aliasing filter; the modulated signal to be measured is amplified by the amplifier circuit and sent to the gain control circuit, and the signal is amplified by the gain control circuit; the gain control The signal output by the circuit is sent to the anti-aliasing filter, and the high-frequency noise is filtered out by the anti-aliasing filter and the impedance matching of the AD analog-to-digital converter is completed, and then sent to the analog-to-digital conversion unit.

After the signal to be tested is amplified and output by the amplifying circuit, the gain control circuit is used to control the gain of the amplified signal, and the amplified signal is adjusted to the best sampling level of the analog-to-digital conversion unit, which improves the signal-to-noise ratio and the dynamic range of weak signal measurement; An anti-aliasing filter is used to filter out high-frequency noise, which avoids aliasing and distortion of subsequent sampling signals, and suppresses high-frequency noise at the same time, further improving the signal-to-noise ratio.

The amplifying circuit is composed of a preamplifier and a secondary amplifier; the signal to be measured is amplified by the preamplifier and then input to the secondary amplifier, and then further amplified by the secondary amplifier and then input to the gain control circuit.

The preamplifier is selected as a low-noise amplifier; the low-noise amplifier with high input impedance and high common-mode rejection ratio is used to reduce the noise introduced and complete the impedance matching with the signal source. Because the signal is very weak, pre-amplification alone is not enough, so a secondary amplifier is used for further amplification.

The gain control circuit is a programmable gain controller. The programmable gain controller can adjust the gain multiple through software programming, so the gain multiple of the gain control circuit can be adjusted according to the strength of the signal to be measured, so that the signal output by the amplifying circuit can be effectively adjusted to the best sampled by the analog-to-digital conversion unit level, improving the signal-to-noise ratio and dynamic range for weak signal measurements.

The anti-aliasing filter is an active low-pass filter.

The analog-to-digital conversion unit adopts a bipolar AD converter.

The anti-aliasing filter may also include a DC bias circuit; the analog-to-digital conversion unit adopts a unipolar Σ-Δ analog-to-digital converter.

The function of the DC bias circuit is to add a DC component to the output signal of the gain control circuit. The input analog signal is effectively sampled by the analog-to-digital conversion unit. In each sampling period, the unipolar Σ-Δ analog-to-digital converter monitors the input signal and outputs the average value of the signal in the sampling period, thereby reducing the impact of noise on the system. Due to the particularity of sinusoidal signals, the sampling frequency of the unipolar Σ-Δ ADC must be set to f _s =(m+1)f _c , where m is a positive integer and f _c is the signal carrier frequency.

The application software in the processor can also include:

A DC component processing module for performing DC component processing on the data sampled and output by the Σ-Δ analog-to-digital converter; the Fourier spectrum analysis module performs power spectrum analysis on the signal after the DC processing, and obtains the relative energy of the signal to be detected. strength.

The key principle of the present invention is Parseval's theorem, that is, the total energy of a signal in the time domain is equal to the total energy in the frequency domain. Expressed as:

$\underset{\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}}\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

According to Parseval's theorem, signal detection in the time domain can be performed on frequency; the signal to be detected is modulated to a fixed frequency, and only the energy of this frequency needs to be obtained in the frequency domain to obtain the desired measurement result.

Using discrete-time signal Fourier transform (DFT) and Parseval theorem, the energy of each frequency point can be obtained, and then the energy of the signal modulation frequency can be obtained to realize the relative measurement of weak signals.

The processor may also include a calibration correction module for performing calibration correction on the energy relative strength of the signal to be detected obtained by power spectrum analysis.

In order to accurately measure the absolute size of the signal energy to be detected, it is necessary to perform calibration before the test work. First, the signal to be detected is set as a known stronger precision signal V _H , and the power spectrum is calculated as _PH , then the signal to be detected is set as a known weaker precision signal V _L , and the power spectrum is calculated as _PL , from the stronger The system correction coefficient s is obtained by calculating the signal power spectrum and the weaker signal power spectrum:

$\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; L</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; L</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

According to the system correction coefficient s, the absolute size v of the signal energy to be detected can be obtained:

$\mathrm{<span\; class="notranslate">\; v</span>}<span\; class="notranslate">\; =</span>\sqrt{\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; u</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

Where u is the relative intensity of the energy of the signal to be detected obtained after power spectrum analysis by the Fourier spectrum analysis module.

The invention reduces the phase-sensitive detector and reference signal generation channel required by the general weak signal detection system, the hardware circuit structure is simple and reliable, and the production cost is reduced; the function of the digital signal processing method in the weak signal detection is strengthened, and the Transformation, the signal to be detected is measured from the frequency domain, the measurement sensitivity is high, and the dynamic range of weak signal measurement is wide.

The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

Description of drawings:

Fig. 1 is a structural block diagram of the present invention. In the figure, 1 is a signal channel unit, 2 is an analog-to-digital conversion unit, and 3 is a processor.

FIG. 2 is a structural block diagram of the signal channel unit 1 of the present invention. 11 amplifying circuit, 12 gain control circuit, 13 anti-aliasing filter, 111 preamplifier, 112 secondary amplifier.

FIG. 3 is a schematic diagram of application software function modules of the processor 3 of the present invention. 31 DC component processing module, 32 Fourier spectrum analysis module, 33 calibration correction module.

Detailed ways:

As shown in FIG. 1 , the present invention includes a signal channel unit 1 , an analog-to-digital conversion unit 2 and a processor 3 . The signal channel unit 1 includes an amplifying circuit 11 , a gain control circuit 12 and an anti-aliasing filter 13 ; wherein the amplifying circuit 11 includes a preamplifier 111 and a secondary amplifier 112 . The pre-amplifier 111 is a weak-signal pre-stage operational amplifier, using a high-precision, low-noise instrumentation amplifier INA118; the secondary amplifier 112 is an OP07 amplifier. The gain control circuit 12 adopts a programmable gain controller, and its model is AD8321. Anti-aliasing filter 13 uses an operational amplifier (OP07) to implement active low-pass filtering. The analog-to-digital conversion unit 2 adopts a bipolar AD converter, the model of which is AD676. The analog-to-digital conversion unit 2 can also use a unipolar Σ-Δ type analog-to-digital converter, the model of which is ADS1255. When the analog-to-digital conversion unit 2 adopts a unipolar Σ-Δ type analog-to-digital converter, a DC bias circuit is set in the anti-aliasing filter 13 in a manner known in the art, so that a DC bias circuit is added to the output signal of the gain control circuit portion. The processor 3 is a digital signal processor 3 whose model is TMS320C5509.

The working process of the present invention is as follows: the input signal is the signal to be detected x ₀ modulated to a frequency of 125 Hz. The signal x ₀ to be detected is amplified by the preamplifier 111 to output a signal x ₁ , and the signal x ₁ is sent to the secondary amplifier 112 , and further amplified by the secondary amplifier 112 to form a signal x ₂ . In order to prevent the amplitude of the signal _x2 from exceeding the input range measured by the analog-to-digital conversion unit 2 and improve the measurement accuracy and dynamic range, the output signal _x2 of the secondary amplifier 112 is sent to the gain control circuit 12, and the signal _x2 is controlled by the programmed gain The signal x ₃ is formed after gain control of the filter, and the signal x ₃ is sent to the anti-aliasing filter 13 . In anti-aliasing filter 13, in order to eliminate the influence of aliasing distortion as much as possible, an operational amplifier (OP07) is used to realize a 4th-order active low-pass filter, and its 3dB bandwidth is set to 150Hz; in order to meet the requirement of unipolar Σ-Δ According to the requirements of the analog-to-digital converter on the sampling level, a DC bias circuit is set in the active low-pass filter in a manner known in the art, so that a DC component is added to the signal x ₃ . The signal x ₃ is filtered by the anti-aliasing filter 13 to output the signal x ₄ , and the signal x ₄ is sent to the analog-to-digital conversion unit 2 . Due to the characteristics of sinusoidal sampling, the sampling frequency of the analog-to-digital conversion unit 2 is set to 1000 Hz, that is, the signal is oversampled by 8 times. In order to effectively sample data of the input signal, the analog-to-digital conversion unit 2 adopts a Σ-Δ type analog-to-digital converter. The Σ-Δ analog-to-digital converter monitors the input signal x ₄ in each sampling period, and outputs the signal average value x ₅ in the sampling period, thereby reducing the impact of noise on the system. The signal average value x ₅ in each sampling period obtained by sampling is sent to the processor 3 . In the processor 3 , firstly, the DC component processing module 31 performs DC component removal processing on the signal average x ₅ to form a signal x ₆ , and the signal x ₆ is sent to the Fourier spectrum analysis module 32 .

In order to accurately obtain the spectrum of the signal, the longer the sampling time, the better, but the longer the sampling time, the greater the amount of calculation and storage, and the signal detection needs to measure the signal to be tested in the shortest possible time, so There is a tradeoff between sampling accuracy and sampling time. According to actual needs, select the frequency resolution Δf = 0.1Hz to determine the number of points N required for Fourier transform:

N=f _s /Δf=10000 (4)

Where f _s is the sampling frequency.

。Then perform Fourier transform directly on the 10,000-point data x(n) to obtain the spectral density

.

Then take the square of its amplitude and divide it by N to get the true power spectrum of the sampled signal:

$\mathrm{<span\; class="notranslate">\; P</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}}{<span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; N</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

Since the sampling frequency is f _s and the number of sampling points is N, the frequency resolution is f _s /N, and the signal power is Es. At this time, the noise bandwidth is B _N , and the noise average power spectral density is G _N . The system before Fourier transform The signal-to-noise ratio of is:

${\mathrm{<span\; class="notranslate">\; SNR</span>}}_{\mathrm{<span\; class="notranslate">\; in</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 20</span>}\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; (</span>\frac{{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; in</span>}}}{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; in</span>}}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 20</span>}\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; Es</span>}}{{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; N</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; N</span>}}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 5</span>}<span\; class="notranslate">\; )</span>$

${\mathrm{<span\; class="notranslate">\; SNR</span>}}_{\mathrm{<span\; class="notranslate">\; out</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 20</span>}\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; (</span>\frac{{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; out</span>}}}{{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; out</span>}}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 20</span>}\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; Es</span>}}{{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; N</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; N</span>}}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; N</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 20</span>}\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; Es</span>}<span\; class="notranslate">\; \&Center\; Dot;</span>\mathrm{<span\; class="notranslate">\; N</span>}}{{\mathrm{<span\; class="notranslate">\; G</span>}}_{\mathrm{<span\; class="notranslate">\; N</span>}}<span\; class="notranslate">\; \&Center\; Dot;</span>{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; N</span>}}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 6</span>}<span\; class="notranslate">\; )</span>$

S _in is the input signal power, S _out is the output signal power, N _in is the input noise power, N _out is the output noise power. From (5) and (6), the output and input signal-to-noise ratio can be obtained as:

$\frac{{\mathrm{<span\; class="notranslate">\; SNR</span>}}_{\mathrm{<span\; class="notranslate">\; out</span>}}}{{\mathrm{<span\; class="notranslate">\; SNR</span>}}_{\mathrm{<span\; class="notranslate">\; in</span>}}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 20</span>}\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 7</span>}<span\; class="notranslate">\; )</span>$

Therefore, when the number of sampling points N is 10000, the signal-to-noise ratio of the system is improved to 80dB, which greatly improves the ability of weak signal detection.

The output data x ₇ transformed by the Fourier spectrum analysis module 32 is sent to the calibration correction module 33 .

In order to accurately measure the absolute size of the signal energy to be detected, it is necessary to perform calibration before the test work. First, the signal to be detected is set as a known stronger precision signal V _H , and the power spectrum is calculated as _PH , then the signal to be detected is set as a known weaker precision signal V _L , and the power spectrum is calculated as _PL , from the stronger The system correction coefficient s is obtained by calculating the signal power spectrum and the weaker signal power spectrum:

$\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; L</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; L</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

According to the system correction coefficient s, the absolute size v of the signal energy to be detected can be obtained:

$\mathrm{<span\; class="notranslate">\; v</span>}<span\; class="notranslate">\; =</span>\sqrt{\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; u</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

Where u is the relative intensity of the energy of the signal to be detected obtained after power spectrum analysis by the Fourier spectrum analysis module 32 .

The present invention is not limited to the above embodiments, and the processor 3 can also use other devices with signal analysis and processing capabilities such as programmable logic device (FPGA), ARM, single-chip microcomputer or PC. As long as a device with signal analysis and processing capabilities is used to convert the weak time-domain signal to be measured into a frequency-domain signal for power spectrum analysis and obtain measurement results, it does not deviate from the idea of the present invention and is within the scope of protection intended by the present invention. .

Claims (10)

1, a kind of apparatus for detecting weak signal based on time-frequency conversion comprises signalling channel unit (1), AD conversion unit (2), processor (3); Signal to be detected through ovennodulation is input to signalling channel unit (1), is amplified and filtering by signalling channel unit (1); It is characterized in that exporting quantized signal behind the analog signal sampling of AD conversion unit (2) to signalling channel unit (1) output; Processor (3) receives the quantized signal of AD conversion unit (2) output and is converted into frequency-region signal and carries out power spectrumanalysis, obtains measurement result.

2, the apparatus for detecting weak signal based on time-frequency conversion according to claim 1, it is characterized in that described processor (3) application software comprises is used for AD conversion unit (2) output signal is carried out power spectrumanalysis, obtains the fourier spectra analysis module (32) of signal energy relative intensity to be detected.

3, the apparatus for detecting weak signal based on time-frequency conversion according to claim 1 is characterized in that described signalling channel unit (1) comprises amplifying circuit (11), gain control circuit (12), frequency overlapped-resistable filter (13); After the measured signal of ovennodulation amplifies through amplifying circuit (11), send into gain control circuit (12), by gain control circuit (12) to the signal amplification that gains; The signal of gain control circuit (12) output is sent into frequency overlapped-resistable filter (13), sends into AD conversion unit (2) by frequency overlapped-resistable filter (13) filter away high frequency noise and after finishing the impedance matching of AD analog to digital converter.

4, the apparatus for detecting weak signal based on time-frequency conversion according to claim 3 is characterized in that described amplifying circuit (11) is made up of prime amplifier (111) and two-stage amplifier (112); Measured signal amplifies back input two-stage amplifier (112) through prime amplifier (111), further amplifies back input-gain control circuit (12) through two-stage amplifier (112).

5, the apparatus for detecting weak signal based on time-frequency conversion according to claim 4 is characterized in that described prime amplifier (111) selects low noise amplifier for use.

6, the apparatus for detecting weak signal based on time-frequency conversion according to claim 3 is characterized in that described gain control circuit (12) selects the programmable-gain controller for use.

7, the apparatus for detecting weak signal based on time-frequency conversion according to claim 3 is characterized in that described frequency overlapped-resistable filter (13) selects active low-pass filter for use.

8, the apparatus for detecting weak signal based on time-frequency conversion according to claim 7 is characterized in that comprising dc bias circuit in the described frequency overlapped-resistable filter (13); Described AD conversion unit (2) adopts unipolarity ∑-Δ pattern number converter.

9, the apparatus for detecting weak signal based on time-frequency conversion according to claim 8 is characterized in that described processor (3) application software also comprises and is used for DC component treatment module (31) that the data of unipolarity AD converter sampling output are carried out DC processing; Described fourier spectra analysis module (32) carries out power spectrumanalysis with the signal after the DC processing, obtains signal energy relative intensity to be detected.

10,, it is characterized in that described processor (3) also comprises the calibration correction module (33) that is used for the signal energy relative intensity to be detected that power spectrumanalysis obtains is carried out scaled correction according to claim 2 or the described apparatus for detecting weak signal of 9 each claims based on time-frequency conversion.

Images