An asynchronous frequency shifting method based on periodic rectangular signals

Abstract

This invention provides an asynchronous frequency shifting method based on a periodic rectangular signal. The steps are as follows: Step a, acquiring an input periodic rectangular wave signal; Step b, segmenting the periodic rectangular wave signal into an effective pulse width signal and an ineffective pulse width signal; Step c, applying different frequency shift amounts to the effective pulse width signal and the ineffective pulse width signal respectively to generate an asynchronously frequency-shifted signal; Step d, obtaining the frequency shift difference value through interaction between the asynchronously frequency-shifted signal and the target echo signal, and separating the signal from noise based on the frequency shift difference value to obtain the processed signal. This invention is applied in the field of radar signal processing technology.

Description

Technical Field

[0001] This invention relates to the field of radar signal processing technology, and in particular to an asynchronous frequency shifting method based on periodic rectangular signals. Background Technology

[0002] In lidar systems, the transmitted signal typically uses a periodic rectangular wave (such as a pulsed laser). Within one cycle of this signal, the duration of the high-level signal (amplitude A) carries the majority of the detection energy and is called the "effective pulse width" (τ); while the duration of the low-level signal (amplitude 0) is mainly used to distinguish pulse intervals and does not directly carry detection information; this is called the "ineffective pulse width." Figure 1 As shown, the time-domain structure of a periodic rectangular wave signal is intuitively illustrated, clearly distinguishing between the effective pulse width τ and the ineffective pulse width. The effective pulse width τ is an inherent, pre-set parameter of the transmitted signal, determined by the radar system design, and used to control the duration of the laser pulse. The pulse width characteristics of the target echo signal mainly affect the energy distribution and specific shape of the echo signal within the effective pulse width period, but the signal division is based on τ and T defined by the transmitted signal itself.

[0003] When the target echo pulse width exceeds the preset effective pulse width τ, the echo signal energy may diffuse into the invalid pulse width period, leading to separation errors. To solve this problem, a reasonable τ value needs to be preset according to the target detection scenario (e.g., τ is usually set to 100ns~1μs for aerosol detection). If the echo pulse width remains abnormal, dynamic matching can be achieved by adaptively adjusting τ or increasing the pulse interval T. The division is always based on the transmitted signal τ; the target echo only affects the energy distribution, not the defined boundary.

[0004] In radar systems, frequency shifting is a step in the signal modulation process that transmits information by changing the frequency of the carrier signal. After a lidar emits its signal, a Doppler frequency shift occurs because the aerosol (or other moving target object) is moving at a certain speed. v d In signal processing, to extract Doppler frequency shift information, the signal first needs to be frequency-shifted. In lidar, the laser is divided into local oscillator light and emitted laser light, where the frequency of the local oscillator light is assumed to be... v 0, modulation frequency shift is v M After frequency shifting of the emitted laser, we have: the emitted laser frequency is v 0+ v M The frequency of the echo signal detected by the lidar is v 0+ v M + v dThe echo signal and the local oscillator signal are converted into intermediate frequency electrical signals. Then, digital signal processing technology is used to detect relevant parameters based on the Doppler frequency shift component in the spectrum. Since the reflected laser signal is very weak and easily interfered with by noise, signal processing is necessary to obtain a high signal-to-noise ratio (SNR) for accurate extraction of the Doppler frequency shift. If the SNR is too low, the accuracy of the measurement results will be affected. Reducing noise impact is crucial for improving SNR during signal processing. Existing frequency shifting techniques perform a global frequency shift, failing to achieve different frequency shifts in different time slots. This is detrimental to improving the signal-to-noise ratio. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide an asynchronous frequency shifting method based on periodic rectangular signals. In this method, by generating different frequency shifts between the effective pulse width time slot and the ineffective pulse width time slot, signal-to-noise separation is achieved through the frequency difference between the signal and the noise, thereby improving the signal-to-noise ratio.

[0006] The technical solution adopted in this invention is an asynchronous frequency shifting method based on a periodic rectangular signal, which includes the following steps:

[0007] Step a: Obtain the input periodic rectangular wave signal, and set the period of the rectangular wave transmission signal to T and the amplitude to A;

[0008] Step b: The periodic rectangular wave signal is segmented into an effective pulse width signal τ and an invalid pulse width signal. The signal within the time period with an amplitude of A in each cycle of the rectangular wave signal is defined as the effective pulse width signal τ, and the signal within the time period with an amplitude of 0 in each cycle of the rectangular wave signal is defined as the invalid pulse width signal. This division is independent of the specific pulse width characteristics of the target echo.

[0009] Step c: Apply different frequency shifts V to the effective pulse width signal and the ineffective pulse width signal respectively. M1 and V M2 This generates an asynchronously frequency-shifted signal; where V M1 ≠V M2 , and |V M1 -V M2 |> Δv_min, where Δv_min is the minimum detectable frequency shift resolution of the radar receiving system;

[0010] Step d: By interacting with the asynchronously frequency-shifted signal and the target echo signal, the frequency shift difference is obtained. Based on the frequency shift difference, the signal and noise are separated to obtain the processed signal.

[0011] Further, the specific steps of step a are as follows: Transmit a periodic rectangular wave signal. Denote the transmitted signal as f(t). Then the mathematical expression of the transmitted signal in the time domain is:

[0012]

[0013] where A is the amplitude of the transmitted signal, T is the period of the transmitted signal, the effective pulse width is set as τ, n is the number of periods, taking values 0, 1, …, N, and the frequency of the transmitted signal is V = 1 / T.

[0014] Further, step b is specifically as follows:

[0015] Divide each period of the transmitted signal into two parts: an effective pulse width signal and an ineffective pulse width signal based on its inherent time structure. This division is completely based on the mathematical definition of the transmitted signal f(t) and the preset parameters T and τ, which is known and controllable and does not depend on the characteristics of the target echo signal. Among them,

[0016] Effective pulse width signal: corresponding to the time period of the time domain expression f(t) = A, that is, nT - τ / 2 < t < nT + τ / 2. This time period is the main part carrying energy in the transmitted signal;

[0017] Ineffective pulse width signal: corresponding to the time period of the time domain expression f(t) = 0, that is, nT + τ / 2 < t < (n + 1)T - τ / 2. This time period is used to separate pulses;

[0018] where n is the number of periods, taking values 0, 1, …, N.

[0019] Further, the specific steps of step c are as follows: Let the effective pulse width signal generate a frequency shift of V M1 , and let the ineffective pulse width signal generate a frequency shift of V M2 . Then the frequency shift amount of the transmitted signal is expressed as a piecewise function:

[0020] ;

[0021] The transmitted signal interacts with aerosol particles in the air, generating a Doppler frequency shift. Let the generated Doppler frequency shift signal be V d . Then the frequency shift amount of the echo signal is:

[0022] ;

[0023] where the specific numerical selection of the frequency shift amounts V M1 and V M2 still needs to meet the following conditions:

[0024] Frequency shift difference |V M1 - V M2|> Δv_min, where Δv_min is the minimum detectable frequency shift resolution of the radar receiving system.

[0025] Further, the specific steps of step d are as follows: Solving for the difference in frequency shift between the effective pulse width signal and the ineffective pulse width signal, as follows:

[0026] (V) M1 +V d )-(V M2 +V d )=V M1 -V M2 ,

[0027] By using asynchronous frequency shifting, invalid pulse widths are separated from valid pulse widths, thus achieving the separation of valid and invalid pulse width signals.

[0028] The beneficial effects of this invention are: This invention performs time-domain analysis on the input signal, divides the effective / ineffective pulse width based on a preset parameter τ, and applies conditions satisfying |V M1 -V M2 The differential frequency shift processing under the Δv_min condition achieves the separation of the effective pulse width signal and the ineffective pulse width signal (and accompanying noise) in the frequency domain. The difference in frequency shift (V) between the effective and ineffective pulse width periods is calculated. M1 -V M2 This difference serves as a key basis for identifying and separating these two types of signal components in the frequency domain. Finally, the separated effective signal (located in V) M1 +V d Spectral analysis was performed on the frequency band to extract the Doppler frequency shift component V. d The motion parameters of the target object are obtained and further optimized. This invention effectively solves the problem of separating effective information and noise in periodic rectangular wave signals, improves the accuracy and reliability of signal detection, and has important application value in radar, communication and other fields. Attached Figure Description

[0029] Figure 1 This is a time-domain diagram of the radar system's signal;

[0030] Figure 2 This is the original frequency amplitude diagram;

[0031] Figure 3 This is the frequency amplitude diagram (V) after processing by existing frequency shifting methods. M The region contains both valid signals and noise.

[0032] Figure 4 This is the frequency amplitude diagram obtained by the processing of this invention, where V M1 For an effective pulse width signal, V M2 This is an invalid pulse width signal. Detailed Implementation

[0033] like Figure 1 and Figure 4 As shown, an asynchronous frequency shifting method based on a periodic rectangular signal includes the following steps:

[0034] Step a: Obtain the input periodic rectangular wave signal, and set the period of the rectangular wave transmission signal to T and the amplitude to A;

[0035] Step b: The periodic rectangular wave signal is segmented into an effective pulse width signal τ and an invalid pulse width signal. The signal within the time period with an amplitude of A in each cycle of the rectangular wave signal is defined as the effective pulse width signal τ, and the signal within the time period with an amplitude of 0 in each cycle of the rectangular wave signal is defined as the invalid pulse width signal.

[0036] Step c: Apply different frequency shifts V to the effective pulse width signal and the ineffective pulse width signal respectively. M1 and V M2 This generates an asynchronously frequency-shifted signal; where V M1 ≠V M2 , and |V M1 -V M2 |> Δv_min, where Δv_min is the minimum detectable frequency shift resolution of the radar receiving system;

[0037] Step d: By interacting with the asynchronously frequency-shifted signal and the target echo signal, the frequency shift difference is obtained. Based on the frequency shift difference, the signal and noise are separated to obtain the processed signal.

[0038] Specifically, step a consists of the following steps:

[0039] If a periodic rectangular wave signal is transmitted, and the transmitted signal is denoted as f(t), then the time-domain mathematical expression of the transmitted signal is:

[0040]

[0041] Where A is the amplitude of the transmitted signal, T is the period of the transmitted signal, the effective pulse width is τ, n is the number of periods, taking values ​​of 0, 1, ..., N, and the frequency of the transmitted signal is V = 1 / T.

[0042] The specific steps of step b are as follows:

[0043] Each cycle of the transmitted signal is divided into two parts based on its inherent time structure: an effective pulse width signal and an ineffective pulse width signal. This division is entirely based on the mathematical definition of the transmitted signal f(t) and preset parameters T and τ, and is known and controllable, independent of the characteristics of the target echo signal. The target echo only affects the energy distribution.

[0044] Effective pulse width signal: corresponding to the time period of the time-domain expression f(t)=A, that is, nT - τ / 2 < t < nT + τ / 2, and this time period is the main part carrying energy in the transmitted signal;

[0045] Ineffective pulse width signal: corresponding to the time period of the time-domain expression f(t)=0, that is, nT + τ / 2 < t < (n + 1)T - τ / 2, and this time period is used to separate pulses;

[0046] where n is the number of cycles, and the values are 0, 1, …, N.

[0047] The specific steps of step c are as follows:

[0048] Let the effective pulse width signal generate a frequency shift V M1 , and let the ineffective pulse width signal generate a frequency shift V M2 , then the frequency shift amount of the transmitted signal is expressed as a piecewise function:

[0049] ;

[0050] The transmitted signal interacts with aerosol particles in the air to generate a Doppler frequency shift. Let the generated Doppler frequency shift signal be V d , then the frequency shift amount of the echo signal is:

[0051] ;

[0052] where the specific numerical values of the frequency shift amounts V M1 and V M2 also need to meet the following conditions:

[0053] The frequency shift difference |V M1 - V M2 | > Δv_min, where Δv_min is the minimum detectable frequency shift resolution of the radar receiving system (or the resolution bandwidth of spectrum analysis).

[0054] Δv_min is an inherent performance parameter of the radar system, which is determined by receiver design (such as local oscillator phase noise, intermediate frequency filter bandwidth) and signal processing capabilities (such as FFT points, sampling rate). Those skilled in the art can select appropriate values of V M1 and V M2 to ensure that their difference (V M1 - V M2 ) can be clearly distinguished and detected in the spectrum. For example, if the minimum detectable frequency shift of the system is Δv_min, then it is necessary to select |V M1 - V M2 | > Δv_min.

[0055] The specific steps of step d are as follows: Solve for the difference between the frequency shift of the effective pulse width signal and the ineffective pulse width signal, which is:

[0056] (V) M1 +V d )-(V M2 +V d )=V M1 -V M2 ,

[0057] By using asynchronous frequency shifting, invalid pulse widths are separated from valid pulse widths, thus achieving the separation of valid and invalid pulse width signals.

[0058] Specifically, in the echo spectrum, the energy of the effective pulse width signal will be mainly concentrated in V. M1 +V d Nearby, the energy of invalid pulse width signals (and noise at similar frequencies) will be mainly concentrated in V. M2 +V d Nearby. Due to |V M1 -V M2 | > Δv_min (the minimum detectable frequency shift of the system), these two spectral peaks can be clearly distinguished. Therefore, by designing a suitable bandpass filter, only V can be extracted. M1 +V d The nearby spectral components (i.e., the echo information corresponding to the effective pulse width signal), while V M2 +V d Nearby spectral components (mainly containing invalid pulse width signals and some noise) are filtered out, thereby separating the signal (effective information) from the noise and significantly improving the signal-to-noise ratio (SNR).

[0059] Furthermore, because the target echo carries a Doppler frequency shift V during the effective pulse width period... d The invalid pulse width period mainly includes environmental noise. By using V M1 With V M2 The difference is set to be greater than Δv_min to ensure that separable double peaks are formed in the spectrum. Figure 4 Even if the target echo pulse width exceeds τ, the noise energy is always distributed in V. M2 +V d Nearby, a fixed bandpass filter (center frequency V) can still be used. M1 +V d Extract valid signals.

[0060] Figure 4 This demonstrates the effective pulse width signal (V) after asynchronous frequency shifting. M1 (region) and invalid pulse width signal (V) M2 The region is separated in the spectrum. V M1and V M2 The spacing needs to be greater than the system resolution Δv_min to achieve clear separation in the diagram. Where V... M1 The region mainly contains the energy of the effective pulse width signal (concentrated in V). M1 +V d (nearby), V M2 The region mainly contains the energy of invalid pulse width signals and noise (concentrated in V). M2 +V d nearby). Figure 4 In the middle, V M1 The peak amplitude of the region is significantly higher than that of V. M2 The region (typical ratio > 3:1) demonstrates that the effective signal energy is concentrated in the preset time period τ. The distance between the two peaks |V M1 -V M2 |=100 MHz (Example 2) is clearly visible, satisfying the separation condition.

[0061] Compared with existing technologies, this invention achieves signal-to-noise separation by applying differential frequency shifts to the effective and ineffective pulse widths and utilizing the frequency shift difference. Experiments show that this method can improve the signal-to-noise ratio by more than 30%, significantly enhancing the detection capability of weak signals, and is suitable for high-precision detection systems of lidar.

[0062] The present invention will now be further described with reference to specific embodiments.

[0063] Example 1: Aerosol Wind Speed ​​Detection (Low-Frequency Scenarios)

[0064] The period of the input periodic rectangular wave signal is set to T=10 μs, the pulse repetition frequency is 100 kHz, and the effective pulse width is τ=1 μs (duty cycle 10%).

[0065] Different frequency shifts are applied to the effective pulse width signal and the ineffective pulse width signal, respectively, where V M1 =50MHz, V M2 =30MHz, Doppler frequency shift is V d =0.5MHz, wind speed 10 m / s;

[0066] Generate a periodic rectangular wave signal according to the above settings, such as... Figure 1 As shown; the effective pulse width period is shifted to 50MHz, and the ineffective period is shifted to 30MHz; in the echo spectrum, the effective signal peak is located at 50.5MHz, and the noise peak is located at 30.5MHz (reference). Figure 4 The 50.5MHz signal was extracted using a bandpass filter with a bandwidth of 2MHz.

[0067] Finally, the processed signal achieved an SNR of 18 dB, while the signal processed by the conventional method achieved an SNR of 12 dB. This means that the signal-to-noise ratio (SNR) obtained by the method of this invention is 50% higher than that obtained by the conventional method.

[0068] Example 2: Drone Tracking (High-Frequency Scenarios)

[0069] The period of the input periodic rectangular wave signal is set to T=2 μs, the pulse repetition frequency is 500 kHz, and the effective pulse width is τ=200 ns (duty cycle 10%).

[0070] Different frequency shifts are applied to the effective pulse width signal and the ineffective pulse width signal, respectively, where V M1 =200MHz, V M2 =100MHz, Doppler frequency shift is V d =10MHz, drone speed 50 m / s;

[0071] Generate a periodic rectangular wave signal according to the above settings, and divide the effective / ineffective pulse width into τ=200 ns, as follows. Figure 1 As shown; the effective pulse width period is shifted to 200MHz, and the ineffective period is shifted to 100MHz; in the echo spectrum, the effective signal peak is located at 210MHz, and the noise peak is located at 110MHz (reference). Figure 4 ); Filter out noise in the frequency band below 110MHz.

[0072] Finally, the processed signal achieved an SNR of 15 dB, while the signal processed by the traditional method achieved an SNR of 8 dB. This means that the signal-to-noise ratio (SNR) obtained by the method of this invention is 87.5% higher than that obtained by the traditional method.

[0073] Finally, it should be emphasized that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various changes and modifications. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An asynchronous frequency shifting method based on a periodic rectangular signal, characterized in that, The method includes the following steps: Step a: Obtain the input periodic rectangular wave signal, set the period of the rectangular wave transmission signal as T, and the amplitude as A; Step b: Perform segmented processing on the periodic rectangular wave signal, and divide the signal into an effective pulse width signal τ and an ineffective pulse width signal. Among them, the signal in the period with an amplitude of A in each period of the rectangular wave signal is defined as the effective pulse width signal τ, and the signal in the period with an amplitude of 0 in each period of the rectangular wave signal is defined as the ineffective pulse width signal; Step c: Apply different frequency shifts V to the effective pulse width signal and the ineffective pulse width signal respectively. M1 and V M2 This generates an asynchronously frequency-shifted signal; where V M1 ≠V M2 , and |V M1 -V M2 |> Δv_min, where Δv_min is the minimum detectable frequency shift resolution of the radar receiving system; Step d: Through the interaction between the signal after asynchronous frequency shift and the target echo signal, obtain the frequency shift difference, and separate the signal from the noise according to the frequency shift difference to obtain the processed signal.

2. The asynchronous frequency shifting method based on a periodic rectangular signal according to claim 1, characterized in that, The specific steps of step a are as follows: Transmit a periodic rectangular wave signal, and record the transmitted signal as f(t). Then the time-domain mathematical expression of the transmitted signal is: ， where A is the amplitude of the transmitted signal, T is the period of the transmitted signal, assume the effective pulse width is τ, n is the number of periods, taking values 0, 1,..., N, and the frequency of the transmitted signal is V = 1 / T.

3. The asynchronous frequency shifting method based on a periodic rectangular signal according to claim 2, characterized in that, The specific steps of step b are as follows: Divide each period of the transmitted signal into two parts, an effective pulse width signal and an ineffective pulse width signal, based on its inherent time structure. This division is completely based on the mathematical definition of the transmitted signal f(t) and the preset parameters T and τ, and is known and controllable, and does not depend on the characteristics of the target echo signal. Among them, Effective pulse width signal: Corresponding to the period of the time-domain expression f(t)=A, that is, nT - τ / 2 < t < nT + τ / 2. This period is the main part of the transmitted signal carrying energy; Ineffective pulse width signal: Corresponding to the period of the time-domain expression f(t)=0, that is, nT + τ / 2 < t < (n + 1)T - τ / 2. This period is used to separate pulses; where n is the number of periods, taking values 0, 1,..., N.

4. The asynchronous frequency shifting method based on a periodic rectangular signal according to claim 3, characterized in that, The specific steps of step c are as follows: The effective pulse width signal is frequency shifted by V. M1 This causes the invalid pulse width signal to generate a frequency shift V. M2 Then the frequency shift of the transmitted signal can be expressed as a piecewise function as: ， The emitted signal interacts with aerosol particles in the air, producing a Doppler frequency shift. Let the generated Doppler frequency shift signal be V. d Then the frequency shift of the echo signal is: ， Among them, frequency shift V M1 and V M2 The specific numerical selection must also meet the following conditions: Frequency shift difference |V M1 -V M2 |> Δv_min, where Δv_min is the minimum detectable frequency shift resolution of the radar receiving system.

5. The asynchronous frequency shifting method based on a periodic rectangular signal according to claim 4, characterized in that, The specific steps of step d are as follows: Solve the difference between the frequency shift amounts of the effective pulse width signal and the ineffective pulse width signal, which is: (V M1 +V d )-(V M2 +V d )=V M1 -V M2 ， Through the method of asynchronous frequency shift, strip the ineffective pulse width from the effective pulse width to achieve the separation of the effective pulse width signal and the ineffective pulse width signal.

Images