Dynamic shift quantization modulation strong and weak signal separation method and device for order disturbance noise suppression

Abstract

The invention discloses a dynamic shift quantization modulation strong and weak signal separation method and device for order disturbance noise suppression, and relates to the signal processing technology, and the method comprises the steps: carrying out the symbol extension of an input signal; counting the maximum value Amax and the minimum value Amin of the signal by adopting a specified sliding window for the symbol expansion signal, and dynamically calculating required parameters; first disturbance d1 is injected into the symbol expansion signal, the symbol expansion signal injected with the disturbance d1 is shifted leftwards by m bits and amplified according to the calculated parameters, second disturbance d2 is injected into the amplified signal, the amplified signal is shifted rightwards by n bits and shifted leftwards by n bits in sequence, strong signal components are extracted, and weak signals are separated through subtraction; the strong signal component and the weak signal component are shifted rightwards by m bits to recover the magnitude; and the strong and weak signals are cut off and output. According to the method, the problem of high-precision separation of extreme signal-to-noise ratio mixed signals confronted by the prominent challenge of'covering weak echoes by strong direct waves' in non-cooperative signal processing is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of signal processing, and relates to non-cooperative signal processing under an external radiation source detection system, and in particular to a dynamic shift quantization modulation strength signal separation method and device for hierarchical disturbance noise suppression. BACKGROUND

[0002] In an electromagnetic sensing scenario, the received signal usually contains a strong direct wave (directly from the transmitting source) and a weak echo (reflected by the target or scattered by the environment). The dynamic range difference between the two is more than 60dB (for example, the strong signal amplitude is ±32767, and the weak signal is only ±328). The strong direct wave accounts for more than 90% of the signal energy, masking the weak echo and causing the target information to be lost. The weak echo contains target characteristics (such as distance, speed, and angle), but the amplitude is extremely low and is easily overwhelmed by noise and quantization errors. For example, in external radiation source radar: using a broadcast / communication signal as an illuminating source, the dynamic range difference between the strong direct wave and the weak target echo in the received signal is more than 80dB; in electronic reconnaissance: strong jamming signals from the enemy and weak communication signals from the own side are mixed, and traditional methods are difficult to separate.

[0003] Implementing efficient and faithful separation of strong direct wave and weak echo signals is the basis for subsequent effective demodulation, positioning, identification, and tracking. Three types of methods are usually used: first, the reference antenna method, which uses an independent reference antenna channel to obtain a "clean" strong direct wave signal, and eliminates the strong signal component through adaptive filtering or cancellation technology, thereby extracting the weak echo. Second, the transform domain separation method, which converts the signal into the frequency domain, time-frequency domain (such as short-time Fourier transform), or sparse representation domain, and uses the distribution difference of strong / weak signals in different domains for separation. Third, the blind source separation method, which is based on independent component analysis (ICA) or principal component analysis (PCA), assumes that the strong / weak signals are statistically independent, and realizes separation through matrix decomposition.

[0004] Comparing the three methods: the reference antenna method has relatively high cancellation accuracy (SNR improvement > 20dB), but it relies on additional hardware, increases system complexity, has high cost and is sensitive to the environment, has poor adaptability, and multipath interference causes phase inconsistency, resulting in a decrease in cancellation performance; the transform domain separation method suppresses the strong signal band in the frequency domain or time-frequency domain (such as FFT zeroing), and frequency domain processing is intuitive and suitable for narrowband signals, but it is computationally complex, has poor real-time performance, and it is difficult to guarantee phase consistency; the blind source separation method does not require prior information, has strong theoretical universality, but the assumption conditions are harsh (the strong / weak signals may be related due to multipath effects), the actual scene performance is unstable, and the iterative convergence is slow and delayed (> 10ms).

[0005] There are also time-domain adaptive filtering separation methods, including LMS / NLMS algorithm and RLS algorithm, the principle of which is to use an adaptive filter to suppress strong signals and extract weak signals by adjusting filter coefficients; relatively speaking, the RLS algorithm has a faster convergence speed and is suitable for real-time processing. For stable mixed signals, the weak signal separation effect is acceptable, but for changing signals, the effect is poor, and the calculation resource consumption is large.

[0006] Disadvantages and deficiencies of the prior art Hardware dependence and poor adaptability to complex environments: The prior art relies on independent reference antennas, floating-point operations or complex frequency domain transformations, resulting in high hardware cost, poor real-time performance, and performance degradation in scenarios such as multipath interference and signal mutation. For example, the reference antenna method requires additional hardware (cost increase of more than 30%), and multipath interference causes phase mismatch; the reference antenna method requires the deployment of high-isolation antennas, and in complex terrain (such as urban multipath environment), the antenna spacing error exceeds λ / 10 (wavelength tenth), and the cancellation performance decreases by more than 10 dB; the frequency domain separation method (FFT / STFT) is computationally complex (delay > 1 ms), and cannot meet the real-time requirements of radar pulse level.

[0007] Quantization noise has a drowning effect on weak signals: The amplitude of weak signals is close to the quantization noise level (such as 16-bit ADC quantization step of 1), and traditional methods amplify weak signals while simultaneously amplifying noise, resulting in insufficient improvement in signal-to-noise ratio (SNR) after separation. For example, RLS filtering can only suppress noise by about 15 dB, and weak signals are still drowned.

[0008] Dynamic range and real-time performance conflict: The signal dynamic range fluctuates dramatically (such as radar echo instantaneous change of 40-70 dB), and traditional adaptive algorithms (such as LMS / RLS) require iterative convergence (delay > 200 μs), which cannot balance real-time performance and accuracy. For example, fixed parameter design cannot adapt to dynamic range mutations (such as sudden appearance of target causing signal amplitude jump); real-time parameter adjustment algorithms (such as sliding window statistics) have high computational complexity and are difficult to deploy in low-power hardware.

[0009] Waste of resources for strong signals: Existing methods focus on "suppressing strong signals" as the core goal, resulting in loss of phase and amplitude information of strong signals after separation, which cannot be used for subsequent tasks such as synchronous demodulation and reference signal generation. For example, strong signals cannot be reused after being canceled or zeroed in the frequency domain; phase truncation error (such as right shift operation low position zero) causes distortion of strong signals. Summary of the invention

[0010] The embodiment of the application provides a dynamic shift quantization modulation strong-weak signal separation method and device for providing a low-complexity, high-fidelity, strong-real-time strong-weak mixed signal separation technology, solving the high-precision separation problem of an extreme signal-to-noise ratio mixed signal faced by the prominent challenge of a strong direct wave covering a weak echo in non-cooperative signal processing, and providing an efficient and universal mixed signal separation method for radar detection, communication anti-jamming, electronic reconnaissance and the like.

[0011] The embodiment of the application provides a dynamic shift quantization modulation non-cooperative strong-weak signal separation method, comprising: performing symbol extension on an input signal; statistically obtaining a maximum value Amax and a minimum value Amin of the signal by using a specified sliding window, and dynamically calculating required parameters for the symbol extended signal; injecting a first disturbance d1 into the symbol extended signal, and left shifting the symbol extended signal amplified by m bits according to the calculated parameters, and injecting a second disturbance d2 into the amplified signal; right shifting and left shifting the signal injected with the second disturbance d2 by n bits in sequence to extract a strong signal component, and separating a weak signal through subtraction; right shifting the strong signal component and the weak signal component by m bits to restore the order of magnitude; adopting symmetric saturation truncation for the strong signal component, and adopting asymmetric truncation for the weak signal output.

[0012] The embodiment of the application provides a dynamic shift quantization modulation non-cooperative strong-weak signal separation device, comprising: a symbol extension module configured to perform symbol extension on an input signal; a dynamic parameter calculation module configured to statistically obtain a maximum value Amax and a minimum value Amin of the signal by using a specified sliding window, and dynamically calculating required parameters for the symbol extended signal; a shift amplification module configured to left shift the symbol extended signal by m bits according to the calculated parameters; a phased disturbance injection module configured to inject a first disturbance d1 and a second disturbance d2 into the signal before and after the left shifting of the symbol extended signal by m bits; a strong signal extraction module configured to extract a strong signal component by right shifting and left shifting by n bits; a weak signal separation module configured to separate a weak signal through subtraction; an order of magnitude recovery module configured to right shift the strong signal component and the weak signal component by m bits to restore the order of magnitude; a truncation output module configured to adopt symmetric saturation truncation for the strong signal component, and adopt asymmetric truncation for the weak signal output.

[0013] The embodiment of the present application provides a low-complexity, high-fidelity and strong real-time strong-weak signal separation technology, solves the high-precision separation problem of an extreme signal-to-noise ratio mixed signal faced by the prominent challenge of "strong direct wave covering weak echo" in non-cooperative signal processing, and provides an efficient and universal mixed signal separation method for radar detection, communication anti-jamming, electronic reconnaissance and the like.

[0014] The above description is only a summary of the technical scheme of the present application, in order to enable the technical means of the present application to be more clearly understood, and to be implemented according to the content of the description, and in order to enable the above and other purposes, characteristics and advantages of the present application to be more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0015] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of the preferred embodiments and are not meant to limit the present application. Furthermore, the same reference numerals are used throughout the several drawings to refer to same or like parts. In the drawings: Figure 1 Another flowchart of the non-cooperative strong-weak signal separation method of the embodiment of the present application is shown in FIG. 4; Figure 2 A DSQM flowchart of the non-cooperative strong-weak signal separation device of the embodiment of the present application is shown in FIG. 5; Figure 3 A disturbance dynamic start-stop control state machine transition diagram of the non-cooperative strong-weak signal separation device of the embodiment of the present application is shown in FIG. 6. DETAILED DESCRIPTION

[0016] Exemplary embodiments of the present disclosure will be described more fully hereinafter with reference to the accompanying drawings, in which exemplary embodiments of the present disclosure are shown. It is to be understood that the present disclosure can be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the present disclosure to those skilled in the art.

[0017] The embodiment of the present application proposes a dynamic shift quantization modulation strong-weak signal separation method with step-by-step disturbance noise suppression, as shown in FIG. 1, including the following steps: Figure 1 In step S101, the input signal is symbol-extended. In some embodiments, symbol-extending the input signal includes expanding the bit width from B in to B ext =2×B in , and filling the high bit with a symbol bit.

[0018] ​In step S102, for the symbol expansion signal, the maximum value Amax and the minimum value Amin of the signal are counted by using a specified sliding window, and the required parameters are dynamically calculated; In step S103, after the symbol expansion, a first disturbance d1 is injected, and the amplitude |d1|=clamp(2^(m-n), 1, Amin / 8, the bit width is low log2(Amin)+3.

[0019] In step S104, according to the calculated parameters, the expansion signal injected with the first disturbance d1 is enlarged by being left shifted by m bits; In step S105, after the enlargement, a second disturbance d2 is injected, and the amplitude |d2|=clamp(2^(m-n), 1, Amin·2^m / 8, the bit width is low m+log2(Amin)+3.

[0020] In step S106, the strong signal component is extracted by being right shifted by n bits and left shifted by n bits in turn.

[0021] In step S107, the weak signal is separated by subtraction, and the strong signal component and the weak signal component are right shifted by m bits to recover the order of magnitude.

[0022] In step S108, the strong signal component is output by using symmetric saturation truncation, and the weak signal is output by using asymmetric truncation.

[0023] The application solves the core contradictions such as “strong signal suppression leading to weak signal phase distortion”, “quantization noise and dynamic range contradiction”, “hardware cost and real-time performance cannot be achieved simultaneously” and the like in the traditional method by using a dynamic shift quantization modulation architecture, and provides a new technical path for the external radiation source detection system.

[0024] The application further realizes quantization noise depth suppression (SNR improvement ≥40 dB), phase error optimization (≤0.8°) and dynamic range expansion (adaptation to 86 dB instantaneous fluctuation) on the basis of maintaining low delay and low resource characteristics by using a phased disturbance injection and parameter linkage control mechanism. The application is suitable for high-precision signal processing scenes such as external radiation source radar, 5G communication anti-jamming, medical ultrasonic imaging and the like.

[0025] Based on the foregoing embodiment, in the further embodiment of the application, the start and stop of the first disturbance and the second disturbance are further controlled in the following manner: The d1 disturbance is enabled when Amin<1000 or k<5; The d2 disturbance is enabled when m-n>8; When k≥12, and, for a preset number of cycles, all disturbances are closed.

[0026] wherein the generation of the first disturbance and the second disturbance is implemented by a single 32-bit LFSR configured with a specified polynomial to generate the disturbance sequences of the first disturbance and the second disturbance in time, for example, in some embodiments, the polynomial is to generate the disturbance sequences of the first disturbance d1 and the second disturbance d2 in time.

[0027] In some embodiments, the start-stop control of the first disturbance and the second disturbance is implemented by a finite state machine including three states: IDLE, d1_ONLY, and d1+d2, with a hysteresis of 2 dB (for the first disturbance d1) and 3 bits (for the second disturbance d2).

[0028] In some embodiments, the dynamically calculated parameters include: Right shift number: n = floor(log2(Amax / max(Amin, 1))); Signal-to-noise ratio: SNRdB = 20log10(Amax / Amin); Adaptive shift number: when SNRdB≥10 dB, k = ceil(SNRdB / 3); when SNRdB<10 dB, k = 2; Left shift number: m = clamp(n+k, 3, B), where the overflow protection bit number B = Bext-Bin.

[0029] In some embodiments, the window size of the sliding window is 2N, and a double-buffered parallel comparator tree is used to obtain Amax and Amin within a single cycle. In some embodiments, the overflow protection bit number B is adapted to the corresponding bit number of the input signal and is configurable, for example, B is adapted to 8 / 16 / 24 / 32-bit input signals.

[0030] In some embodiments, the dynamic calculation is implemented based on a 32-bit barrel shifter, and the right shift number n and the adaptive shift number k are calculated by looking up a table or shifting approximation, with a total delay of ≤7 clock cycles.

[0031] In some examples, the dynamic parameter calculation includes a parallel comparator tree and a 32-bit barrel shifter, with a delay of ≤7 cycles at 100 MHz and resource occupation of <500 LUTs.

[0032] In some examples, the phase error root mean square value after subtraction separation is <1°, and the weak signal signal-to-noise ratio is improved by ≥30 dB.

[0033] The application method is applicable to the following scenarios: radar: separation of strong direct wave (broadcast signal) and weak target echo (dynamic range >80 dB) in external radar.

[0034] Communication: suppress strong interference signals (such as adjacent base station interference), and extract weak user signals (signal-to-noise ratio <20 dB).

[0035] Medical imaging: separation of strong tissue reflection signals and weak blood flow signals in ultrasound imaging (SNR < 60 dB).

[0036] Electronic reconnaissance: separation of strong enemy interference and weak communication signals of one's own in a non-cooperative environment.

[0037] The embodiment of the present application also proposes a dynamic shift quantization modulation strong and weak signal separation device for stage-by-stage disturbance noise suppression, comprising: a symbol extension module configured to perform symbol extension on an input signal; a dynamic parameter calculation module, which calculates the required parameters by using a specified sliding window to count the maximum value Amax and the minimum value Amin of the symbol extension signal; a shift amplification module, which shifts the symbol extension signal left by m bits for amplification according to the calculated parameters; a stage-by-stage disturbance injection module, which injects d1 / d2 disturbances after symbol extension and left shift amplification; a strong signal extraction module, which extracts a strong signal component by right shifting n bits and left shifting n bits; a weak signal separation module, which separates a weak signal by subtraction; a dynamic start-stop control module, which controls the disturbance injection state based on the Amin, k, and m-n parameters.

[0038] a magnitude recovery module, which right shifts the strong signal component and the weak signal component by m bits to recover the magnitude; a truncation output module, which adopts symmetric saturation truncation for the strong signal component and asymmetric truncation for the weak signal for output.

[0039] The embodiment of the present application aims to integrate disturbance injection depth into the DSQM architecture, to realize the collaborative optimization of noise suppression and hardware efficiency through the linkage of stage-by-stage disturbance and dynamic parameters, and to achieve: Quantization noise deep suppression: the SNR of weak signals is increased from 25 dB to more than 40 dB through stage-by-stage disturbance injection.

[0040] Phase fidelity enhancement: the phase distortion caused by truncation error is reduced from 1.5° to less than 0.8°.

[0041] Dynamic disturbance control: a start-stop strategy linked with parameters is constructed to adapt to the instantaneous fluctuation of signal amplitude of 40-86 dB.

[0042] Hardware efficiency maintenance: on the basis of the basic patent 500 LUTs, the resource increment is less than 20% (total occupation < 600 LUTs).

[0043] The embodiment of the present application proposes a stage-by-stage disturbance enhanced DSQM architecture, as shown in Figure 2 . Signal flow: input signal → sign extension → [d1 injection] → left shift amplification → [d2 injection] → split dual-channel processing; Strong signal path: strong signal extraction → strong signal post-processing (strong signal recovery magnitude → strong signal symmetric truncation, recovered bit width output); Weak signal path: weak signal separation → weak signal post-processing (weak signal recovery magnitude → weak signal asymmetric truncation, recovered bit width output); Control flow: sign extension signal → dynamic parameter calculation (optimized) → disturbance start-stop control → LFSR disturbance generation; Key innovation modules: Two-stage disturbance injection module: d1 stage (after sign extension): break the periodicity of original quantization noise, reduce noise floor by 3-6 dB; d2 stage (after left shift amplification): suppress truncation error, reduce weak signal phase error by 40%; Dynamic amplitude formula: |d1|=clamp(2^(m-n), 1, Amin / 8) |d2|=clamp(2^(m-n), 1, (Amin<<m) / 8) Bit width mask design: d1 mask: low log2(Amin)+3 bits (cover 3σ fluctuation); d2 mask: low m+log2(Amin)+3 bits (adapt to bit width expansion after amplification).

[0044] Joint disturbance injection mechanism in stages, disturbance injection timing and physical significance Stage Injection position Core role Quantization noise suppression mechanism d1 After symbol extension, before left-shift amplification Break the original quantization noise periodicity, improve the effective resolution of weak signal (ENOB+0.5bit) Convert quantization noise energy from discrete spectrum to white noise by pseudo-random disturbance, noise floor reduced by 3-6dB. d2 After left-shift amplification, before strong signal extraction Suppress quantization noise and truncation error after amplification Disturbance offsets low n-bit truncation error caused by right-shift-left-shift operation, improves weak signal linearity (phase error reduced by 40%). Precise design of disturbance amplitude and bit width d1 amplitude formula: |d1|=clamp(2^(m-n), 1, Amin / 8) Parameter linkage: 2^(m-n): reflects the dynamic range difference between strong and weak signals (m is the left shift number, n is the strong signal truncation bit number); ensures that the disturbance amplitude does not exceed the quantization noise suppression capability.

[0045] Amin / 8 constraint: prevent disturbance from drowning weak signals (Amin is the minimum value of signals in the sliding window), take 8 for easy shift calculation, ensure that the disturbance does not exceed 12.5% of the weak signal amplitude.

[0046] Lower limit covers the original quantization step (Δ=1), ensures effective noise whitening.

[0047] Physical significance: dynamic amplitude ensures the optimal disturbance ratio of weak signal amplitude and noise floor.

[0048] d1 bit width design: log2(Amin) + 3 log2(Amin): The original effective bit width of the weak signal (e.g., Amin = 328 → 9 bits).

[0049] +3 bits: Based on the 3σ principle, it covers statistical fluctuations and instantaneous errors.

[0050] The formula for the magnitude of d2 is: |d2| = clamp(2^(mn), 1, Amin2^m / 8) Parameter linkage: Amin·2^m / 8: Matches the amplitude of the weak signal after amplification to prevent overflow (e.g., Amin=328, m=16 → upper limit of disturbance = 328×65536 / 8=2,687,872), where 2^m is the left shift amplification factor.

[0051] d2 bit width design: m + log2(Amin) + 3 m: Bit width expansion corresponding to the left shift amplification factor.

[0052] log2(Amin): Preserves the original details of weak signals.

[0053] +3 bits: Covers the noise floor fluctuation after amplification.

[0054] Dynamic disturbance start-stop control strategy like Figure 3 As shown, the control logic is linked to the parameters (threshold is configurable). Disturbance Enable condition Disable condition Hysteresis margin Physical meaning d1 Amin<1000 or k<5 k ≥ 12 2dB Enable when weak signal is close to noise floor or signal-to-noise ratio is insufficient, disable when signal-to-noise ratio recovers. d2 m-n>5 m-n≤ 2 3bit Enable when truncation error is significant, disable when small Parameter definition: k=ceil(SNRdB / 3): Signal-to-noise ratio adaptive coefficient (SNR=20dB→k=7).

[0055] mn: Cut-off strength index (e.g., m=16, n=5→mn=11).

[0056] Hysteresis tolerance: 2dB (d1), 3 bits (d2) to prevent state oscillation.

[0057] State machine implementation like Figure 3 As shown, the state is defined as follows: S0 (IDLE): Disturbance off, monitoring parameters.

[0058] S1(d1_ONLY): Enable only the d1 perturbation.

[0059] S2(d1+d2): Jointly activate d1 and d2 disturbances.

[0060] State transition logic: S0 → S1 : when Amin < 1000 or k < 5; S1 → S2: when m-n > 8 and lasts for 3 periods; S2 → S1 : when m-n ≤ 5; S1 / S2 → S0: when k ≥ 12 and lasts for 5 periods Hardware optimization of the embodiments of the present application: Dynamic parameter calculation module optimization: Fast calculation of log2(Amin): Implementation: priority encoder + lookup table (LUT), input Amin, output its most significant bit position (e.g. 328 → 9).

[0061] Resource occupation: 16-bit input requires 4-level MUX, occupying ≈25 LUTs.

[0062] Efficient generation of 2^(m-n): Implementation: barrel shifter left shift (m-n) bits, supporting 32-bit shift (5-level MUX, delay 5ns).

[0063] Scramble generation module (multiplexing optimization) LFSR sharing: single 32-bit LFSR generates pseudo-random sequence, generating d1 and d2 disturbances in time-sharing mode.

[0064] Polynomial: , with a period of -1.

[0065] Multiplexing logic: high and low bits are scaled and applied respectively Resource multiplexing technique Shifter multiplexing: barrel shifter of dynamic parameter module is shared to the scramble generation module, reducing redundant logic.

[0066] Mask dynamic generation: using existing log2(Amin) and m value to calculate bit width mask in real time, avoiding pre-storage consumption of RAM.

[0067] The embodiments of the present application have the following advantages: Quantization noise depth suppression: through phased disturbance injection, the SNR of weak signals is increased from 25dB in the basic patent to more than 40dB.

[0068] Phase fidelity enhancement: the phase error caused by truncation nonlinearity is reduced from 1.5° to less than 0.9°, improving Doppler velocity measurement and imaging accuracy.

[0069] Dynamic adaptability enhancement: supporting 86dB instantaneous fluctuation of signal amplitude.

[0070] Hardware efficiency preservation: resource footprint only increased by 16% (500→580 LUTs), real-time latency preserved at 70ns @ 100MHz.

[0071] It should be noted that, in the embodiments of the present application, the terms "comprising", "containing" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0072] The above-mentioned sequence numbers of the embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0073] From the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be realized by means of software and necessary general hardware platforms, of course, they can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes a plurality of instructions for making a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) execute the methods described in the embodiments of the present application.

[0074] The embodiments of the present application are described above in combination with the drawings, but the present application is not limited to the above-mentioned specific embodiments, the above-mentioned specific embodiments are only illustrative, not restrictive, and those skilled in the art can make many forms under the inspiration of the present application without departing from the scope of the present application and the protection scope of the claims.

Claims

1. A method for separating strong and weak signals in dynamic shift quantization modulation with graded perturbation noise suppression, characterized in that, include: Sign extension of the input signal; For the symbol-extended signal, the maximum value Amax and minimum value Amin of the signal are statistically analyzed using a specified sliding window, and the required parameters are dynamically calculated. For the symbol-spread signal, a first perturbation d1 is injected, and according to the calculated parameters, the symbol-spread signal injected with the first perturbation d1 is shifted left by m bits to amplify it, and a second perturbation d2 is injected into the amplified signal. For the signal injected with the second disturbance d2, the strong signal component is extracted by shifting it right by n bits and left by n bits in sequence, and the weak signal is separated by subtraction. The magnitude is recovered by shifting strong and weak signal components to the right by m bits. Strong signal components are symmetrically saturated and truncated, while weak signals are asymmetrically truncated and output.

2. The non-cooperative strong / weak signal separation method of dynamic shift quantization modulation as described in claim 1, characterized in that, The first perturbation d1 has an amplitude of |d1|=clamp(2^(m−n),1,Amin / 8) and a bit width of low log2(Amin)+3 bits. The second perturbation d2 has an amplitude of |d2|=clamp(2^(m−n),1,Amin·2^m / 8) and a bit width of low m+log2(Amin)+3 bits.

3. The non-cooperative strong / weak signal separation method of dynamic shift quantization modulation as described in claim 2, characterized in that, It also includes controlling the start and stop of the first and second disturbances in the following ways: d1 perturbation is enabled when Amin < 1000 or k < 5; When mn>8, the d2 perturbation is enabled; When k≥12, and for a preset number of periods, all disturbances are turned off.

4. Among them, The generation of the first and second perturbations is implemented using a single-channel 32-bit LFSR, which is configured with a specified polynomial to generate the perturbation sequences of the first and second perturbations in a time-division manner.

5. The non-cooperative strong / weak signal separation method of dynamic shift quantization modulation as described in claim 3, characterized in that, The start-stop control of the first and second disturbances is implemented using a finite state machine, which includes three states: IDLE, d1_ONLY, and d1+d2, with a hysteresis tolerance of 2dB and 3 bits.

6. The non-cooperative strong / weak signal separation method of dynamic shift quantization modulation as described in claim 1, characterized in that, Sign extension of the input signal includes increasing the bit width from B... in Extend to B ext =2×B in The high-order bit is filled with the sign bit.

7. The non-cooperative strong / weak signal separation method of dynamic shift quantization modulation as described in claim 5, characterized in that, The parameters calculated dynamically include: Right shift by bits: n = floor(log2(Amax / max(Amin,1))); Signal-to-noise ratio: SNRdB = 20log10(Amax / Amin); Adaptive shifter: when SNRdB ≥ 10dB, k = ceil(SNRdB / 3); when SNRdB < 10dB, k = 2; Left shift number: m = clamp(n + k, 3, B), where B = Bext - Bin.

8. The non-cooperative strong / weak signal separation method of dynamic shift quantization modulation as described in claim 6, characterized in that, The sliding window has a window size of 2N, and a double-buffered parallel comparator tree is used to obtain Amax and Amin in a single cycle.

9. The non-cooperative strong / weak signal separation method of dynamic shift quantization modulation as described in claim 6, characterized in that, The dynamic calculation is based on a 32-bit barrel shifter. The right shift number n and the adaptive shift number k are calculated by lookup table or shift approximation, with a total delay of ≤7 clock cycles.

10. A dynamic shift quantization modulation strong / weak signal separation device for graded perturbation noise suppression, characterized in that, include: The sign extension module is configured to extend the sign of the input signal; The dynamic parameter calculation module uses a specified sliding window to statistically analyze the maximum value Amax and minimum value Amin of the symbol-extended signal and dynamically calculates the required parameters. The shift amplification module amplifies the sign-extended signal by shifting it left by m bits based on the calculated parameters. The staged perturbation injection module is used to inject a first perturbation d1 and a second perturbation d2 into the signal before and after the sign-extended signal is amplified by shifting it left by m bits. The strong signal extraction module extracts strong signal components by shifting the signal right by n bits and left by n bits. The weak signal separation module separates weak signals through subtraction. The magnitude recovery module recovers the magnitude by shifting the strong and weak signal components to the right by m bits. The output module uses symmetrical saturation truncation for strong signal components and asymmetrical truncation for weak signals.

Images