A red light parameter autonomous regulation system

Abstract

This invention relates to the field of optical technology, and more particularly to an autonomous red light parameter control system, comprising a dual-mode coded light emission module, a multi-angle scattered light detection module, a phase-locked demodulation module, a correlation prediction module, an all-optical differential controller, a liquid crystal microstructure array, a penetration energy detection module, and a depth-locking controller. The system constructs parallel independent dual control loops. The fast-response loop consists of phase-locked demodulation, correlation prediction, the all-optical differential controller, and the liquid crystal microstructure array; the slow-response loop consists of penetration energy detection, the depth-locking controller, and the liquid crystal microstructure array. The two loops work synergistically on the same liquid crystal microstructure array. In this invention, the fast loop adjusts the wavefront energy distribution in real time based on the spatial scattering field signal to avoid local high-scattering regions, while the slow loop adjusts the overall beam curvature based on penetration energy feedback to lock the target energy. This achieves simultaneous optimization of red light transmission efficiency and effectiveness in dynamically turbid media, overcoming the shortcomings of existing single-loop control methods.

Description

Technical Field

[0001] This invention relates to the field of optical technology, and in particular to a system for autonomous adjustment of red light parameters. Background Technology

[0002] Red light, due to its excellent penetration into biological tissues, low damage, and stable transmission characteristics, is widely used in deep biological tissue therapy, underwater communication, and turbid environment detection. In these applications, red light needs to be transmitted through dynamic turbid media, such as living biological tissues and water containing suspended particles. The autonomous red light parameter control system is the core equipment to ensure the transmission efficiency of red light and guarantee its final effectiveness, such as therapeutic effect, communication quality, and detection accuracy. By adjusting parameters such as the wavefront morphology and energy distribution of red light, it reduces the adverse effects of dynamic turbid media on red light transmission, achieving precise action of red light on the target area. However, existing red light parameter control systems generally employ a single control loop design, making it difficult to simultaneously achieve adaptive optimization of red light spatial transmission efficiency and final effect performance in dynamic turbid media. Specifically, some existing systems only design fast-response control mechanisms for the local transmission state of red light. While this can avoid local high scattering regions to some extent, it cannot stably lock the final effect energy of the target region, resulting in large fluctuations in effect performance. Other systems only design slow-response control mechanisms around the final effect energy. Although this can maintain the target energy roughly stable, it cannot respond in real time to the rapid changes in local scattering characteristics in dynamic turbid media, leading to low red light spatial transmission efficiency and a large amount of energy being lost through scattering. These shortcomings of existing technologies have consistently limited the application effect of red light in dynamic turbid media. Summary of the Invention

[0003] To overcome the above deficiencies, this invention provides an autonomous control system for red light parameters, which aims to improve the shortcomings of existing single control loops that cannot simultaneously optimize the spatial transmission efficiency and final effectiveness of red light in dynamic turbid media.

[0004] This invention provides the following technical solution: a red light parameter autonomous control system, comprising: The main light source and the detection light source are used to generate the main red light for treatment or communication and the detection red light for detection, respectively. The light emission module is optically connected to the main light source and the detection light source respectively. It is used to periodically encode and modulate the detection red light, and then combine the encoded and modulated detection red light with the main red light and emit it. The optical detection module is optically connected to the output optical path of the optical emission module. It is used to receive the coded detection light backscattered from the dynamic turbid medium and output multiple scattered light signals. A phase-locked demodulation module, electrically connected to the photodetector module, is used to synchronously demodulate the real-time spatial distribution signal of scattered light intensity from the multi-path scattered light signals; The prediction module, electrically connected to the phase-locked demodulation module, is used to calculate the optical cross-correlation function based on the spatial distribution signal of the scattered light intensity, and generate a turbidity change warning signal when the attenuation of the optical cross-correlation function value exceeds a preset threshold. The differential control module is electrically connected to the phase-locked demodulation module and the prediction module. It is used to generate a spatial error voltage signal based on the difference between the spatial distribution signal of the scattered light intensity and a preset reference spatial distribution signal, and to simulate and superimpose the turbidity change warning signal and the spatial error voltage signal to form the final driving signal. A liquid crystal modulation array is disposed in the optical path of the main red light and electrically connected to the differential control module, for spatially modulating the wavefront phase of the main red light according to the final driving signal; A penetration energy detection module is set at the expected target location of the dynamic turbid medium to receive the main red light after penetrating the dynamic turbid medium and generate a penetration energy feedback signal. A depth-locked controller is electrically connected to the penetration energy detection module and the liquid crystal modulation array. It is used to generate an overall curvature adjustment signal based on the deviation between the penetration energy feedback signal and a preset target energy value, and send it to the liquid crystal modulation array to adjust the overall phase curvature of the liquid crystal modulation array.

[0005] Preferably, the optical emitting module includes: The main red light emitting unit is optically connected to the main light source and is used to collimate and transmit the main red light; A detection light coding emission unit, optically connected to the detection light source, is used to generate the detection red light and perform periodic intensity coding modulation on it using a coding modulator; The beam combining unit optically connects the main red light emitting unit and the probe light encoding emitting unit, and is used to combine the modulated probe red light with the main red light.

[0006] Preferably, the optical detection module includes: The ring detector array unit is optically connected to the output optical path of the light emitting module. It is used to receive the coded detection light backscattered from the dynamic turbid medium and output multiple raw scattered photoelectric signals. The signal conditioning unit is electrically connected to the ring detector array unit and is used to amplify and filter the original scattered photoelectric signals of each channel, and output multiple channels of conditioned scattered light signals. The signal distribution and output unit is electrically connected to the signal conditioning unit and is used to output the multi-channel conditioned scattered light signals to the phase-locked demodulation module.

[0007] Preferably, the phase-locked demodulation module includes: Multiple demodulator units are electrically connected to each output terminal of the optical detection module in a one-to-one correspondence. Each demodulator unit is used to coherently demodulate one of the input scattered light signals at a reference frequency that is the same as the coding modulation frequency of the detected red light, and output the demodulated signal. The spatial distribution synthesis unit is electrically connected to the output terminals of all the demodulator units, and is used to receive all the demodulated signals and integrate them according to the spatial position correspondence to form the spatial distribution signal of the scattered light intensity.

[0008] Preferably, the prediction module includes: The signal buffer unit is electrically connected to the output terminal of the phase-locked demodulation module and is used to continuously store and output the spatial distribution signal of scattered light intensity at historical times. The analog optical correlation operation unit has two input terminals and one output terminal. Its first input terminal is electrically connected to the output terminal of the signal buffer unit to receive historical signals, and its second input terminal is electrically connected to the real-time output terminal of the phase-locked demodulation module to receive real-time signals. It is used to calculate the normalized cross-correlation function between the real-time signal and the historical signal and output the function value. The warning signal generation unit is electrically connected to the output terminal of the analog optical correlation calculation unit. It is used to compare the normalized cross-correlation function value with the preset threshold, and generate and output the turbidity change warning signal when the function value is lower than the threshold.

[0009] Preferably, the differential control module includes: A reference signal generation unit is used to store and output an electrical reference signal corresponding to the preset reference spatial distribution signal; The analog differential computing array has multiple parallel differential computing channels. Each differential computing channel has two input terminals and one output terminal. The first input terminal of each differential computing channel is electrically connected to the corresponding spatial position output terminal of the phase-locked demodulation module to receive a portion of the scattered light intensity spatial distribution signal. The second input terminal of each differential computing channel is electrically connected to the corresponding output terminal of the reference signal generation unit to receive a portion of the electrical reference signal. Each differential computing channel is used to calculate the difference between the two received signals and output it as the error voltage of the corresponding spatial position. The feedforward fusion node has multiple fusion input ports and one drive output port. Each fusion input port is electrically connected to the output of each channel of the analog differential computing array to receive the error voltage, and is additionally electrically connected to the output of the prediction module to receive the turbidity change warning signal. It is used to simulate superimpose the turbidity change warning signal with all the error voltages, and output the final drive signal from the drive output port to the liquid crystal modulation array.

[0010] Preferably, the liquid crystal modulation array comprises: A liquid crystal phase modulation panel, the surface of which is divided into multiple independently controllable micro-units, is disposed in the optical path of the main red light; The high-speed drive circuit is electrically connected to the output of the differential control module and the output of the depth lock controller, and is independently connected to each micro-unit of the liquid crystal phase modulation panel. The high-speed drive circuit is used to independently generate and apply a drive voltage to each micro-unit according to the final drive signal to control its phase delay, and synchronously adjust the reference voltage bias applied to all micro-units according to the overall curvature adjustment signal.

[0011] Preferably, the penetration energy detection module includes: A narrowband optical filter is disposed in the receiving optical path at the expected target location to filter out the detection red light and ambient stray light, allowing only the main red light to pass through; The photoelectric conversion unit is optically connected to the output optical path of the narrowband optical filter and is used to convert the transmitted main red light into an electrical signal as the output of the penetration energy feedback signal.

[0012] Preferably, the depth lock controller is an analog integral controller, the input terminal of the depth lock controller is electrically connected to the output terminal of the penetration energy detection module to receive the penetration energy feedback signal, and the output terminal of the depth lock controller is electrically connected to the liquid crystal modulation array to output the overall curvature adjustment signal.

[0013] Preferably, the differential control module, the liquid crystal modulation array, the phase-locked demodulation module, and the prediction module constitute a fast-response control loop; the depth-locked controller, the liquid crystal modulation array, and the penetration energy detection module constitute a slow-response control loop; the fast-response control loop is used to adjust the wavefront energy distribution of the main red light in real time based on the spatial distribution signal of the scattered light intensity and the turbidity change early warning signal; the slow-response control loop is used to stably adjust the overall penetration depth of the main red light based on the penetration energy feedback signal; the fast-response control loop and the slow-response control loop operate in parallel and independently, and work together on the liquid crystal modulation array.

[0014] The present invention has the following beneficial effects: 1. In this invention, a fast-response control loop consisting of a phase-locked demodulation module, a prediction module, a differential control module, and a liquid crystal modulation array, and a slow-response control loop consisting of a penetration energy detection module, a depth-locked controller, and a liquid crystal modulation array, are constructed. These two loops operate independently and collaboratively on the same actuator. The fast loop adjusts the wavefront energy distribution in real time based on the spatial scattering field signal to avoid local high-scattering regions, while the slow loop adjusts the overall beam curvature based on penetration energy feedback to lock the target energy. This achieves synchronous adaptive optimization of red light spatial transmission efficiency and final performance in a dynamically turbid medium.

[0015] 2. In this invention, the prediction module calculates the optical cross-correlation function based on real-time and historical spatial distribution signals of scattered light intensity and generates a turbidity change early warning signal. The differential control module then simulates and superimposes this early warning signal with a feedback signal based on spatial error. This design enables the system not only to compensate for existing medium scattering changes but also to provide feedforward pre-compensation for impending scattering abrupt changes, significantly improving the system's response speed and control stability in the face of rapid and sudden changes in the medium's state, and enhancing the system's robustness.

[0016] 3. In this invention, the penetration energy detection module uses a narrowband optical filter to accurately separate the main red light and generate a feedback signal. The depth locking controller uses an analog integrator to process the deviation between this signal and the target value and outputs an overall curvature adjustment signal. This slow-response closed-loop does not rely on complex algorithms and achieves zero steady-state error locking of the total energy reaching the target position through pure analog circuitry. This ensures that regardless of how slowly the overall turbidity of the medium drifts, the dose or signal intensity of the red light energy acting on the target can be automatically maintained constant, guaranteeing the final performance of the application. Attached Figure Description

[0017] Figure 1 This is a framework diagram of a red light parameter autonomous control system proposed in this invention. Detailed Implementation

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1: In the first embodiment of the present invention, the present invention provides a system for autonomous adjustment of red light parameters, such as... Figure 1 As shown, it includes: The main light source and the detection light source are used to generate the main red light for treatment or communication and the detection red light for detection, respectively. Specifically, the main light source is responsible for generating the core beam used for final treatment or information transmission. It is implemented using a semiconductor laser that emits red light of a specific wavelength, such as a laser diode with a wavelength of 660 nanometers. This light source has high and stable output power, and the emitted light, called the main red light, is the effective energy carrier for the system's final function. The probe light source is responsible for generating a probe beam for sensing the state of the medium. It is implemented using another semiconductor laser that emits a slightly different wavelength from the main red light, such as a laser diode with a wavelength of 670 nanometers. Its output power is lower, and crucially, its output light intensity is subjected to periodic intensity modulation at a specific frequency, such as a 1 kilohertz sine wave modulation. The emitted light after modulation is called probe red light, which acts as a sensitive probe for the system to sense changes in the external medium. The main light source and the probe light source work together to form the foundation of the system. The main red light serves as the action beam, and the probe red light as the sensing beam; the slight difference in their wavelengths provides the conditions for subsequent optical separation, while the unique modulation characteristics of the probe red light enable it to be accurately identified and extracted from complex background light. For example, in a phototherapy application, a stable 660 nm main red light is used to irradiate the tissue, while a modulated 670 nm weak probe light is emitted at the same time to monitor changes in the tissue's scattering characteristics in real time, thus providing a basis for the system's autonomous regulation.

[0020] The light emission module is optically connected to the main light source and the detection light source respectively. It is used to periodically encode and modulate the detection red light, and then combine the encoded and modulated detection red light with the main red light and emit it. Furthermore, the optical emitting module includes: The main red light emitting unit is optically connected to the main light source and is used to collimate and transmit the main red light; A detection light coding emission unit, optically connected to the detection light source, is used to generate the detection red light and perform periodic intensity coding modulation on it using a coding modulator; The beam combining unit optically connects the main red light emitting unit and the probe light encoding emitting unit, and is used to combine the modulated probe red light with the main red light.

[0021] Specifically, the light emission module is the key front end of the system that shapes, modulates, and synthesizes the raw light beam generated by the light source to form the final emitted light. It ensures that the primary red light and the probe red light enter the test medium together with the correct spatial relationship and signal characteristics. This module is specifically completed by three units working sequentially, and its technical implementation is as follows: The main red light emitting unit is directly connected to the main light source. Its core task is to convert the raw laser beam emitted by the main light source, which typically has a certain divergence angle, into a collimated, parallel beam with a suitable spot size and uniform energy distribution. This is mainly achieved through a lens group consisting of a collimating lens and a spatial filter. The focal length f of the collimating lens... c The aperture and divergence angle of the main light source need to be selected to convert the diverging light into parallel light. Subsequently, the beam may pass through a beam expander system consisting of two lenses to adjust its beam diameter D. m This process ensures the beam is matched to the apertures of subsequent beam combining units and liquid crystal modulation arrays. For example, for a main light source with an output beam diameter of 2 mm and a divergence angle of 10 milliradians, a collimating lens with a focal length of 20 mm can be selected for initial collimation. Then, a beam expander system with a 3x magnification is used to expand the beam diameter to 6 mm to meet the light transmission requirements of subsequent optical components. The main red light processed by this unit is a parallel beam with good collimation, uniform spot size, and controllable diameter, providing an ideal optical input for subsequent wavefront modulation.

[0022] The probe light encoding and emitting unit is connected to the probe light source and undertakes the crucial task of encoding the probe red light signal. Its implementation involves two key steps: First, beam collimation, similar in principle to the main red light emitting unit, uses a lens group to collimate the light emitted from the probe light source into parallel light. Next, this collimated continuous probe light is incident on an encoding modulator. This encoding modulator is typically an external modulator based on electro-optic or acousto-optic effects, whose transmittance or diffraction efficiency is determined by an externally applied periodic electrical signal V. mod (t) Control. Modulation signal V mod (t) is generated by a low-frequency signal generator in the system, and its form is usually a sine wave, square wave, or pseudo-random binary sequence. Taking sinusoidal intensity modulation as an example, the transfer function of the modulator makes its output light intensity I... p (t) and input light intensity I p0 and driving voltage V mod (t) exhibits a linear relationship. Assuming the modulator operates at the bias point, its output light intensity can be expressed as: Among them, f m The modulation frequency ϕ0 and the initial phase are both set by the signal generator. I in the formula... p0 The power of the probe light source is determined by this unit. Through this unit, the originally continuous probe light is given specific, periodic intensity variation characteristics that can be recognized by subsequent circuits; this is known as "encoding." For example, the system sets f... m =10kHz, the signal generator produces a 10kHz sinusoidal voltage to drive the modulator, causing the output detection red light intensity to oscillate sinusoidally at a frequency of 10kHz. This encoded detection red light beam carries the frequency f in its intensity envelope.m The phase ϕ0 information becomes the sole identification basis for subsequent phase-locked demodulation modules to extract signals and suppress noise.

[0023] The beam combining unit receives the processed beams from the two units mentioned above—the collimated primary red light and the coded and modulated probe red light—and combines them into the same optical path. This is typically achieved using a dichroic mirror or polarizing beam splitter with high transmittance for both wavelengths. Taking a dichroic mirror as an example, it is designed to transmit high transmittance for the primary red light wavelength λ. m High transmittance, and also suitable for detecting red light wavelength λ p High reflectivity. In the optical path arrangement, the collimated primary red light is incident perpendicularly through the dichroic mirror, while the coded probe red light is incident at a certain angle onto the reflecting surface of the dichroic mirror. After being reflected, it propagates collinearly with the transmitted primary red light, thus synthesizing a composite beam containing two wavelengths and two intensity modes simultaneously. After beam combining, it is necessary to ensure that the two beams are completely overlapped in space, have the same direction, and match in spot size to avoid different propagation paths in the medium. The output of this unit is the optical signal that the system ultimately emits into the dynamically turbid medium. It simultaneously carries the dual functions of energy transmission and state sensing, providing a unified and well-defined excitation source for the entire closed-loop control system.

[0024] The optical detection module is optically connected to the output optical path of the optical emission module. It is used to receive the coded detection light backscattered from the dynamic turbid medium and output multiple scattered light signals. Furthermore, the optical detection module includes: The ring detector array unit is optically connected to the output optical path of the light emitting module. It is used to receive the coded detection light backscattered from the dynamic turbid medium and output multiple raw scattered photoelectric signals. The signal conditioning unit is electrically connected to the ring detector array unit and is used to amplify and filter the original scattered photoelectric signals of each channel, and output multiple channels of conditioned scattered light signals. The signal distribution and output unit is electrically connected to the signal conditioning unit and is used to output the multi-channel conditioned scattered light signals to the phase-locked demodulation module.

[0025] Specifically, the optical detection module is the front-end sensor of the system that senses the state information of the external dynamic turbid medium. Its core function is to capture the coded probe light carrying information about the scattering characteristics of the medium, which is backscattered from it, and convert it into multiple electrical signals that can be processed by subsequent circuits. The implementation of this module ensures that the system can analyze changes in the scattered field from a spatial perspective.

[0026] The ring detector array unit is the core of the module's optical sensing. It is implemented by arranging multiple identical silicon photodiodes or photomultiplier tubes at equal intervals around the emission optical axis on a circular plane. This ring plane is perpendicular to the emission optical path, and its center is typically located in the paraxial region of the emitted beam to specifically receive large-angle backscattered light rather than specular reflection light. Each detector unit has a small receiving field of view, collectively forming a ring field of view around the optical axis. When the modulated probe red light is scattered in a dynamically turbid medium, some photons return in various directions and are received by the individual units in the ring array. Each detector unit encodes the received probe light intensity I, whose intensity varies with time. s,i (t), where the subscript i represents the i-th detector unit, converted to a weak photocurrent I proportional to it. elec,i (t). For example, in a scenario where changes in tissue blood flow are monitored, when a blood vessel constricts at a certain location, causing an increase in the local scattering coefficient, the intensity of the backscattered light received by the detector unit at the corresponding spatial angle will undergo a specific change, which is initially converted into an electrical signal.

[0027] The signal conditioning unit is directly connected to the output of each detector unit. Its core task is to convert the weak, noisy photocurrent signal into a clean, appropriately amplitude voltage signal. This unit performs the same processing on each signal in parallel. First, a transimpedance amplifier converts the photocurrent I... elec,i (t) is converted into a voltage signal V raw,i (t), whose conversion relationship is determined by the feedback resistor R f Decide: Among them, R f The resistance value is selected based on the expected intensity of the probe light and the input range of subsequent circuits. The voltage signal then passes through a bandpass filter. The center frequency f of the filter... c Precisely set to the modulation frequency f of the probe light m Its bandwidth B is narrow enough to preserve the modulated signal while suppressing ambient light noise, main red light scattering noise, and circuit thermal noise to the greatest extent. This filtering process can be regarded as a signal purification in the frequency domain. After amplification and filtering, the original, noisy current signal I... elec,i (t) is converted into a relatively clean, amplified analog voltage signal V. cond,i (t), this signal mainly retains the signal caused by medium scattering and the signal with encoding frequency f. m Related intensity modulation information.

[0028] The signal distribution and output unit serves as the interface between the module and the subsequent phase-locked demodulation module. It is implemented as a stable connection architecture of a multi-channel buffered driver circuit or an analog multiplexer. This unit receives multiple parallel voltage signals V from the signal conditioning unit.cond,i (t), its core function is to provide a low-output-impedance buffer drive for each signal, ensuring that the signal is not attenuated or introduced with crosstalk during transmission to the phase-locked loop demodulation module, and achieving correct channel correspondence. For example, in a ring array, the i-th detector corresponding to the azimuth angle θ has a conditioned signal V. cond,i (t) After passing through this unit, the signal is stably transmitted to the demodulator unit in the phase-locked demodulation module, which is responsible for processing the signal in the θ direction. The output of this unit is the final product of the photodetector module, the multi-path scattered light signal S. det,i These signals (t) are spatially discrete and temporally synchronized, collectively characterizing the instantaneous state of the backscattered light intensity at different spatial angles, providing a direct input for subsequent extraction of the spatial distribution of the scattered field.

[0029] A phase-locked demodulation module, electrically connected to the photodetector module, is used to synchronously demodulate the real-time spatial distribution signal of scattered light intensity from the multi-path scattered light signals; Furthermore, the phase-locked demodulation module includes: Multiple demodulator units are electrically connected to each output terminal of the optical detection module in a one-to-one correspondence. Each demodulator unit is used to coherently demodulate one of the input scattered light signals at a reference frequency that is the same as the coding modulation frequency of the detected red light, and output the demodulated signal. The spatial distribution synthesis unit is electrically connected to the output terminals of all the demodulator units, and is used to receive all the demodulated signals and integrate them according to the spatial position correspondence to form the spatial distribution signal of the scattered light intensity.

[0030] Specifically, the phase-locked demodulation module is the core of the system's signal extraction. Its task is to accurately and synchronously recover the true intensity information of the coded probe light in various spatial directions from multi-path scattered light signals carrying strong noise, thereby constructing the spatial distribution of the backscattered light intensity of the medium in real time. This module achieves this through two steps: parallel coherent detection and spatial information integration.

[0031] Multiple demodulator units operate in parallel, each independently processing a signal from a specific spatial channel of the photodetector module. The core of this implementation is a combination of an analog multiplier or mixer and a low-pass filter. Each demodulator unit receives two inputs: one is the i-th scattered light signal S from the photodetector module. det,i (t), this signal is a modulated signal whose amplitude is attenuated and mixed with noise after being scattered by the medium, and its form can be modeled as S. det,i (t)=A i sin(2πf m t+ϕ i )+n i (t), where A iThe amplitude of the signal to be determined reflects the intensity of scattered light in that direction, ϕ. i For the phase delay caused by propagation, n i (t) represents noise; the other input is a local reference signal R(t) = sin(2πf) from the system's global modulation signal generator, which is in phase and originate from the same source as the transmitter's probe light modulation. m t). The demodulator first converts the input signal S det,i Multiply (t) with the reference signal R(t): After the multiplication operation, signal S det,i Useful amplitude information A in (t) i and phase information ϕ i It is converted into the difference frequency (DC) term and the harmonic term, while the noise n i (t) is then modulated to the reference frequency and its sidebands. Following this, the multiplier output M... i (t) is fed into a cutoff frequency much lower than 2f m A low-pass filter. This filter completely removes all high-frequency components, including noise, retaining only the DC component. Where G is the overall gain of the multiplier and filter. During system design, optical path calibration or the selection of a zero-difference detection architecture can be used to adjust the ϕ of each channel. i Keep it within a very small range, at which point cos(ϕ) i Therefore, the output voltage V is approximately 1. demod,i That is, the amplitude A of the original scattered light i That is, it is proportional to the instantaneous scattered light intensity in that spatial direction. For example, in applications monitoring particle concentration in turbid water flow, when particles locally aggregate in a certain direction due to eddies, the corresponding DC voltage V output by the demodulator unit will be proportional to this. demod,i This will increase significantly, accurately reflecting the enhancement of backscattering in that direction. The V output of each demodulator unit... demod,i This is the "demodulated signal," which represents the scattered light intensity information at a specific spatial angle after noise reduction and amplitude extraction.

[0032] The spatially distributed synthesis unit receives the demodulated signal V output from all N demodulator units. demod,1 V demod,2 ,...,V demod,N In essence, it is a data aggregation and formatting interface. Internally, this unit contains a multi-channel analog sample-and-hold circuit that synchronously captures the V signals of all channels under unified clock control. demod,iThe values ​​are captured and held within a short sampling window. These synchronously captured analog voltage values ​​can then be read out sequentially using an analog multiplexer or directly converted to digital values ​​using a set of parallel analog-to-digital converters. Regardless of the method used, the output is an ordered set of data, where each data point explicitly corresponds to a specific spatial azimuth angle θ within the ring detector array. i This ordered discrete data sequence or vector, reflecting the variation of scattered light intensity with spatial angle θ, is defined as the spatial distribution signal I of scattered light intensity. s (θ). For example, if the ring array has 16 detectors, then I s (θ) is a sequence containing 16 elements, where the value of the i-th element is V. demod,i The corresponding angle θ i =(i−1)×22.5. This spatially distributed signal I s (θ) is the direct and clean input for the subsequent prediction module and differential control module to perform time-domain analysis and spatial difference calculation. It transforms the dynamic spatial changes of the medium state into a deterministic electrical signal that the system can process.

[0033] The prediction module, electrically connected to the phase-locked demodulation module, is used to calculate the optical cross-correlation function based on the spatial distribution signal of the scattered light intensity, and generate a turbidity change warning signal when the attenuation of the optical cross-correlation function value exceeds a preset threshold. Furthermore, the prediction module includes: The signal buffer unit is electrically connected to the output terminal of the phase-locked demodulation module and is used to continuously store and output the spatial distribution signal of scattered light intensity at historical times. The analog optical correlation operation unit has two input terminals and one output terminal. Its first input terminal is electrically connected to the output terminal of the signal buffer unit to receive historical signals, and its second input terminal is electrically connected to the real-time output terminal of the phase-locked demodulation module to receive real-time signals. It is used to calculate the normalized cross-correlation function between the real-time signal and the historical signal and output the function value. The warning signal generation unit is electrically connected to the output terminal of the analog optical correlation calculation unit. It is used to compare the normalized cross-correlation function value with the preset threshold, and generate and output the turbidity change warning signal when the function value is lower than the threshold.

[0034] Specifically, the prediction module is a key component of the system's proactive control. Its core function is to predict abrupt changes in the turbidity properties of the medium by analyzing the temporal correlation changes in the spatial distribution signal of scattered light intensity, and to generate early warning signals in advance. This module uses fully analog circuitry to perform time-domain correlation analysis and threshold determination of the signal at the hardware level.

[0035] The signal buffer unit is responsible for providing the spatial distribution signal of historical time points for related operations. It can be implemented using a set of parallel analog delay line arrays. The real-time scattered light intensity spatial distribution signal I from the phase-locked demodulation module... s (t), as an analog voltage vector containing multiple channels, is simultaneously fed into this unit. This unit consists of N independent analog delay lines with a fixed time delay τ. Each delay line corresponds to a spatial channel. When the real-time signal vector I... s (t) Upon input, each delay line delays the voltage signal of its corresponding channel by a fixed time τ before outputting. Therefore, the output of this unit is a historical spatial distribution signal vector delayed by τ in time, denoted as I. s (t−τ). This historical signal I s (t−τ) is continuously provided to subsequent computation units as a reference to the current real-time signal I. s (t) is the benchmark for comparison.

[0036] The analog optical correlation computing unit is the core computing component of this module, and its function is to calculate the current signal I in real time. s (t) and historical signal I s The normalized cross-correlation coefficient between (t−τ) is implemented in the analog domain, and its operation can be represented as an integral evaluation of signal similarity over a short time window T. The core of its implementation circuit consists of an analog multiplier array, an integrator, and a division circuit for normalization. For the i-th spatial channel, the multiplier calculates the real-time signal I. s,i (t) and the delayed signal I s,i The product of (t−τ) is then used. The product signals from all channels are subsequently summed to obtain a composite product signal reflecting the overall spatial distribution similarity. This composite product signal is then fed into an analog integrator with a time constant of T for smoothing, yielding an unnormalized cross-correlation value. Simultaneously, for normalization, the circuit also needs to calculate the real-time signal I. s (t) and historical signal I s (t−τ) represents their respective power. This is achieved through two additional parallel circuits: one calculates I... s (t) Sum of squares of each channel value and integrate, then calculate I on the other path. s The sum of squares of the channel values ​​(t−τ) is integrated, and then the geometric mean of these two integrals is calculated as the normalized denominator. Finally, the normalized cross-correlation function value C(τ) is obtained by the following equation: In this formula, the calculation of the numerator (summation within the integral) is performed by the aforementioned multiplier array and adder, and its output is then passed through an integrator; the two terms in the denominator are generated by the corresponding squaring, summing, and integrating circuits, respectively. Finally, an analog divider circuit performs the division operation between the numerator and denominator, outputting an analog voltage value V proportional to the cross-correlation coefficient C(τ). C For example, when the scattering characteristics of the medium are stable, the real-time signal is highly similar to the historical signal, C(τ) is close to 1, and V... C Output high voltage; when the medium begins to change rapidly, the signal similarity decreases rapidly, the C(τ) value decreases, and V C The voltage subsequently decreases. The warning signal generation unit makes a judgment based on the result VC of the relevant calculation. This is implemented as an analog voltage comparator with hysteresis characteristics. One input of this comparator receives V... C The other input is connected to an adjustable voltage source to set the warning threshold V. th Threshold V th A critical value corresponding to a normalized cross-correlation coefficient, determined through experimental calibration, is used to distinguish between normal, slow fluctuations in the medium and drastic changes that will soon affect the stability of the optical path. When V C Higher than V th When the comparator outputs a low level, it indicates a stable state with no warning. However, if a change in the medium causes C(τ) to decay and V... C Below V th The comparator output immediately flips to a high level. This high-level signal is the turbidity change warning signal V. alert The warning signal V alert The data is directly fed to the feedforward fusion node of the differential control module, allowing it to apply a correction bias to the wavefront modulation before significant distortion actually occurs in the scattered field. This significantly improves the system's response speed and stability to sudden changes in dynamic media. For example, in underwater optical communication, when the prediction module detects a rapid decay in the correlation of the scattered field caused by water agitation and triggers an early warning, the system can adjust the transmitted wavefront in advance to effectively counteract the impending signal degradation.

[0037] The differential control module is electrically connected to the phase-locked demodulation module and the prediction module. It is used to generate a spatial error voltage signal based on the difference between the spatial distribution signal of the scattered light intensity and a preset reference spatial distribution signal, and to simulate and superimpose the turbidity change warning signal and the spatial error voltage signal to form the final driving signal. Furthermore, the differential control module includes: A reference signal generation unit is used to store and output an electrical reference signal corresponding to the preset reference spatial distribution signal; The analog differential computing array has multiple parallel differential computing channels. Each differential computing channel has two input terminals and one output terminal. The first input terminal of each differential computing channel is electrically connected to the corresponding spatial position output terminal of the phase-locked demodulation module to receive a portion of the scattered light intensity spatial distribution signal. The second input terminal of each differential computing channel is electrically connected to the corresponding output terminal of the reference signal generation unit to receive a portion of the electrical reference signal. Each differential computing channel is used to calculate the difference between the two received signals and output it as the error voltage of the corresponding spatial position. The feedforward fusion node has multiple fusion input ports and one drive output port. Each fusion input port is electrically connected to the output of each channel of the analog differential computing array to receive the error voltage, and is additionally electrically connected to the output of the prediction module to receive the turbidity change warning signal. It is used to simulate superimpose the turbidity change warning signal with all the error voltages, and output the final drive signal from the drive output port to the liquid crystal modulation array.

[0038] Specifically, the differential control module is the core of the system's control strategy generation. Its function is to compare the sensed real-time medium state with the ideal state and, in conjunction with forward-looking early warning information, generate the final driving signal for control wavefront modulation. This module achieves high-bandwidth, low-latency analog feedback control through parallel spatial differential computation and multi-signal fusion.

[0039] The reference signal generation unit is responsible for providing the reference state upon which system control is based. Its implementation can be an array of circuits consisting of non-volatile memory and a digital-to-analog converter, or a fixed analog voltage network set by a network of precision potentiometers. This unit stores and outputs a preset reference spatial distribution signal I. ref (θ), its electrical behavior is a set of values ​​related to the spatial angle θ. i The corresponding DC reference voltage V ref,i This reference signal I ref (θ) characterizes the spatial distribution of scattered light intensity that the lock-in demodulation module should measure under "ideal" or "desired" media conditions. Its value is typically obtained through calibration measurements of the system under standard or initialized media conditions and is embedded as a hardware parameter. For example, in a system used for superficial tissue phototherapy, the reference signal I... ref (θ) can be set as a relatively uniform backscattering distribution in an isotropic structure, as measured in a homogeneous phantom tissue.

[0040] The analog differential computing array is the core of the entire module for spatial parallel error calculation. It is implemented as a series of parallel differential amplifiers or instrumentation amplifier circuits. The i-th differential computing channel in the array corresponds to a specific spatial angle θ. iThe first input of this channel receives the voltage value V at this angle from the real-time scattered light intensity signal from the phase-locked demodulation module. s,i That is I s (θ i The first input terminal receives the instantaneous value of the reference signal; the second input terminal receives the reference voltage V corresponding to the angle provided by the reference signal generation unit. ref,i Each differential amplifier subtracts the voltages at its two input terminals and multiplies them by a fixed gain G. d Its output is the space error voltage signal V. err,i Calculated by the following formula: Among them, G d V represents the gain of the differential amplifier, a fixed design parameter. s,i It is a real-time measurement value, V ref,i This is a preset baseline value. Therefore, V err,i It directly quantifies the degree of deviation between the current scattering state of the medium in the i-th spatial direction and the ideal state. If V s,i >V ref,i This indicates that backscattering is enhanced in this direction, then V err,i A positive value indicates a negative value; conversely, a negative value indicates a negative value. All channels operate in parallel, simultaneously generating a set of error voltage vectors V that correspond one-to-one with the spatial angle. err (θ). This vector directly reflects the spatial pattern of the distortion of the medium's scattering field.

[0041] The feedforward fusion node is responsible for integrating the feedback signal based on the current error with the early warning signal based on future changes. It is implemented as a multi-input analog adder network. This node has multiple fusion input ports, each connected to the output of a channel in the analog differential computing array, receiving all N spatial error voltages V. err,i In addition, it has an extra warning input port to receive a turbidity change warning signal V from the prediction module. alert V alert The voltage is 0V when there is no warning and a fixed positive voltage value when a warning is triggered. The function of the fusion node is to superimpose the warning signal onto each spatial error voltage with a certain weight. For the i-th channel, its final driving voltage V drive,i The calculation is implemented in analog circuits using a resistor network, and its mathematical expression is: V drive,i =V err,i +k⋅V alert Here, k is a preset, adjustable gain factor that determines the strength of the contribution of the warning signal to the final drive signal. When V alert When =0, V drive,i =V err,i The system only performs routine feedback compensation. Once the warning is triggered, the final drive signal V of all space channels...drive,i Each of these will add a global positive bias k⋅V to the existing error voltage. alert This global bias is equivalent to commanding the liquid crystal modulation array during the warning period to generally generate a wavefront that avoids the approaching high-scattering region. For example, in underwater communication scenarios, when the prediction module issues a warning due to the detection of distant disturbances, the driving signal output by the fusion node will cause the main lobe of the beam to be pre-adjusted towards the direction of weaker predicted scattering, thereby initiating avoidance adjustments before the disturbance actually reaches the optical path, significantly improving the robustness of the link. Finally, the driving voltage vector V output by the fusion node... drive (θ) is fed in parallel into the high-speed driving circuit of the liquid crystal modulation array, directly controlling the phase modulation amount of each micro-unit.

[0042] A liquid crystal modulation array is disposed in the optical path of the main red light and electrically connected to the differential control module, for spatially modulating the wavefront phase of the main red light according to the final driving signal; Furthermore, the liquid crystal modulation array includes: A liquid crystal phase modulation panel, the surface of which is divided into multiple independently controllable micro-units, is disposed in the optical path of the main red light; The high-speed drive circuit is electrically connected to the output of the differential control module and the output of the depth lock controller, and is independently connected to each micro-unit of the liquid crystal phase modulation panel. The high-speed drive circuit is used to independently generate and apply a drive voltage to each micro-unit according to the final drive signal to control its phase delay, and synchronously adjust the reference voltage bias applied to all micro-units according to the overall curvature adjustment signal.

[0043] Specifically, the liquid crystal modulation array is the core actuator of this system. Its function is to precisely and rapidly reshape the wavefront phase of the main red light according to control commands, thereby actively regulating the propagation behavior of the beam in a dynamic turbid medium. Through the cooperation of two key internal components, the array transforms electrical commands from the differential control module and the depth-locked controller into spatial modulation and overall shaping of the beam wavefront.

[0044] Liquid crystal phase modulation panels are the physical basis for wavefront spatial modulation. Their implementation typically employs a pure phase spatial light modulator based on nematic liquid crystals. This panel consists of millions of individually addressed pixel units arranged in a two-dimensional array, precisely positioned within the collimated and expanded main red light path. Each micro-unit contains a transparent electrode, an alignment layer, and a liquid crystal molecule layer. When a driving voltage V is applied to a micro-unit... unit When this happens, the orientation of the liquid crystal molecules changes, thereby altering the effective refractive index n of the liquid crystal layer at their location. eff (V unitSince the optical path difference is proportional to the product of the refractive index and the thickness, the phase ϕ of the transmitted light after passing through this unit will change. unit A voltage-dependent delay Δϕ(V) will occur relative to the reference phase when no voltage is applied. unit The relationship between this phase delay and the driving voltage is non-linear, but its lookup table or fitting function can be obtained through pre-calibration. For a liquid crystal layer with thickness d, at a specific wavelength λ... m The maximum phase modulation amount is typically designed to be greater than 2π to ensure the generation of any desired wavefront shape. This is achieved when the final drive signal V, which characterizes the spatial error distribution, comes from the differential control module. drive When (θ) is applied to the corresponding microarray, the panel generates a beam cross-section that is similar to V. drive The phase distribution Φ(x,y) corresponding to the (θ) spatial mode is the "tuned wavefront" designed to compensate for or pre-distort the effects of scattering in the medium.

[0045] The high-speed drive circuit serves as a bridge connecting the control logic and the LCD panel, responsible for efficiently and accurately converting two types of control signals into specific voltages applied to each microcell. This circuit comprises two main parts: a multi-channel high-speed digital-to-analog converter and drive array for applying spatial modulation signals; and a global bias voltage generation circuit for applying overall curvature adjustment.

[0046] For the spatial modulation section, the drive circuit receives the final drive signal V, which is transmitted in parallel from the differential control module. drive,i Each 'i' corresponds to a set of microcells at specific spatial locations on the panel. Since LCD panels typically use digital addressing, the driving circuit first converts the analog V... drive,i The values ​​are converted into digital values. Then, a field-programmable gate array (FPGA) converts these digital values ​​into specific pixel driving data for each microcell according to a pre-defined voltage-phase delay relationship. This data is written at high speed to the driver chip of the LCD panel at a frame rate, thereby generating the spatial voltage distribution map V. drive (θ) is "plotted" on the panel in real time to generate the desired wavefront Φ(x,y). For example, when the system identifies enhanced scattering in the upper right region of the medium, the corresponding region's V... drive,i When positive, the driving circuit will cause the micro-units in that area of ​​the panel to produce a specific phase delay, guiding the beam energy to propagate more towards the lower left.

[0047] For the overall curvature adjustment section, the drive circuit receives the overall curvature adjustment signal V from the depth-locking controller. curv This is a single analog voltage value. This signal is fed into a global bias adder. The adder converts this global bias voltage V...bias =k curv ⋅V curv , where k curv This is a scaling factor, which is analogously or digitally superimposed on the fundamental driving voltage of each microcell calculated from the spatial modulation signal. Its mathematical essence is: V unit,final [x,y]=V unit,base [x,y]+V bias ; Here, V unit,base [x,y] is derived from the spatial error signal V. drive The underlying voltage distribution obtained by (θ) mapping is used to generate a fine wavefront structure for compensated scattering; while the superimposed global bias V bias This is equivalent to adding the same linear phase ramp to all micro-units. In wavefront optics, a global linear phase ramp corresponds to changing the beam's propagation direction, while a global quadratic phase distribution corresponds to changing the beam's convergence / divergence. The depth-locked controller outputs V... curv To adjust V bias This allows for the dynamic and holistic adjustment of the average radius of curvature of the wavefront generated by the LCD panel. For example, when the penetration energy detection module detects that the received light intensity is lower than the target value, the depth lock controller outputs an increased V. curv The drive circuit increases V accordingly. bias This makes the wavefront generated by the LCD panel more convergent, thereby increasing the penetration ability of the light beam in the medium until the received light intensity returns to the target value.

[0048] Ultimately, the high-speed drive circuit, through its parallel data paths and global addition mechanism, fuses the instructions from the "fast loop" and the "slow loop," and converts them into the final drive voltage V applied to each microcell of the liquid crystal phase modulation panel. unit,final [x,y], thereby generating a complex wavefront in real time that can both avoid local strong scattering regions and ensure that the overall penetration depth reaches the target, realizing fully autonomous, multi-dimensional coordinated control of the main red light transmission parameters.

[0049] A penetration energy detection module is set at the expected target location of the dynamic turbid medium to receive the main red light after penetrating the dynamic turbid medium and generate a penetration energy feedback signal. Furthermore, the penetration energy detection module includes: A narrowband optical filter is disposed in the receiving optical path at the expected target location to filter out the detection red light and ambient stray light, allowing only the main red light to pass through; The photoelectric conversion unit is optically connected to the output optical path of the narrowband optical filter and is used to convert the transmitted main red light into an electrical signal as the output of the penetration energy feedback signal.

[0050] Specifically, the penetration energy detection module is another key sensor for achieving closed-loop control of the system. Its function is to quantitatively evaluate the actual energy level of the main red light after transmission through the dynamic turbid medium to the target area, and convert this physical quantity into an electrical signal that can be processed by the controller. This module directly reflects the final energy transmission efficiency of the system and provides the core feedback input for the depth-locking controller.

[0051] Narrowband optical filters are the primary optical components for achieving selective detection in the module. They are implemented as a single filter targeting the dominant red light wavelength λ. m An optimized interference filter is used. This filter is typically manufactured using multilayer dielectric film coating technology, and its central transmission wavelength is precisely matched to the emission wavelength λ of the main light source. m It has a narrow transmission bandwidth. The filter is precisely installed at the entrance of the receiving optical path at the intended target location. When a composite beam containing the main red light, the probe red light scattering component, and ambient stray light arrives, the narrowband filter, due to its wavelength selectivity, only allows the center wavelength to be within λ. m Light components within a very narrow range are efficiently transmitted, while the probe red light of different wavelengths and most ambient light are strongly attenuated or reflected. This design ensures that the subsequent photoelectric conversion unit responds only to the energy of the main red light, effectively avoiding signal crosstalk from inside and outside the system, thus guaranteeing the feedback signal E. received Purity and accuracy.

[0052] The photoelectric conversion unit is responsible for linearly converting the light power transmitted through the filter into an electrical signal. It is typically implemented using a silicon photodiode or photomultiplier tube with a large photosensitive surface. This device is optically placed close to the exit surface of the narrow-band optical filter to ensure the collection of all transmitted photons. The photodiode operates in the linear response region, and the photocurrent it generates is I... pd The power P of the main red light incident on its photosensitive surface received It is directly proportional to the responsivity R of the photodiode, measured in A / W. That is: I pd =R⋅P received ; Wherein, the responsivity R is the device's response at wavelength λ. m The inherent parameters under this condition. Subsequently, this weak photocurrent I... pd The signal is converted into a voltage signal by a low-noise transimpedance amplifier circuit. The transimpedance amplifier is connected through a precision feedback resistor R. f Achieve current-to-voltage conversion, output voltage V out =−Ipd ⋅R f This voltage signal V out The signal is then passed through a programmable gain amplifier for amplitude adjustment to match the input voltage range of the subsequent controller, ultimately outputting a stable analog voltage signal that is proportional to the main red light power received at the target location, i.e., the penetration energy feedback signal V. feedback For example, at the receiver of underwater optical communication, regardless of water turbidity or fluctuation, this module consistently and accurately measures the power of the 660 nm main optical signal carrying the information, while ignoring the 670 nm modulated light used for detection and other background light. The entire module outputs V... feedback The signal is transmitted directly and continuously to the input of the depth-locking controller. This feedback signal is used by the depth-locking controller to determine whether the current penetration energy has reached the preset target value E. target The only direct evidence. By comparing V feedback With E target The controller generates a correction signal, thus forming an independent closed-loop control loop aimed at stabilizing the penetration energy. The implementation of this module enables the system not only to optimize energy distribution in space to avoid scattering, but also to precisely guarantee the final energy dose or communication signal strength applied to the target, achieving a direct closed-loop guarantee for the ultimate effect of red light transmission.

[0053] A depth-locked controller is electrically connected to the penetration energy detection module and the liquid crystal modulation array. It is used to generate an overall curvature adjustment signal based on the deviation between the penetration energy feedback signal and a preset target energy value, and send it to the liquid crystal modulation array to adjust the overall phase curvature of the liquid crystal modulation array.

[0054] Furthermore, the depth lock controller is an analog integral controller. The input terminal of the depth lock controller is electrically connected to the output terminal of the penetration energy detection module to receive the penetration energy feedback signal, and the output terminal of the depth lock controller is electrically connected to the liquid crystal modulation array to output the overall curvature adjustment signal.

[0055] Specifically, the depth-locked controller is the core of the closed-loop regulation for the stable final energy transmission of the system. Its function is to convert the deviation between the actual energy received at the target position and the desired energy into continuous adjustment commands for the overall convergence state of the beam. This controller is implemented with purely analog circuitry and is specifically responsible for responding to the slower but crucial energy balance requirements, ensuring that the total energy dose or signal intensity of the main red light acting on the target remains constant after undergoing complex dynamic scattering.

[0056] The core of the depth-locked controller is an analog integrator controller. Its implementation circuitry is based on a high-precision, low-drift operational amplifier configured as a classic integrator. The controller's input receives the penetration energy feedback signal V from the penetration energy detection module. feedback (t), which is an analog voltage proportional to the main red light power received at the target. The system's preset target energy value E target Also provided as a reference voltage, typically generated by a highly stable precision voltage reference source. These two voltage signals are compared in a differential preamplifier to generate the instantaneous error signal e(t): e(t) = E target -V feedback (t); The error signal e(t) directly reflects the gap between the current penetration energy and the expected value: e(t)>0 indicates insufficient energy, and e(t)<0 indicates excess energy.

[0057] This error signal e(t) is then fed into the circuit via an operational amplifier and an integrating resistor R. i and integrating capacitor C i The core integrator circuit constitutes the circuit. The integrator performs continuous time integration on the input voltage, and its output voltage V... curv (t), i.e., the overall curvature adjustment signal, is determined by the following formula: in, V represents the gain of the integrator controller, where the negative sign is determined by the circuit's inverting configuration and can be processed by subsequent circuitry. curv (0) represents the initial output voltage. In this formula, parameter E... target It is a preset constant voltage, V feedback e(t) is a real-time measured value, and the two are obtained by analog subtraction circuit to obtain e(t). Integral gain K i The selected resistor R i and capacitor C i The value of V is determined by the value of ε, which is designed to be relatively small, making the controller's response speed much slower than the fast-response loop responsible for spatial wavefront shaping, thus avoiding mutual interference between the two loops. The characteristic of integral control is that as long as the error e(t) is not zero, the controller's output V... curv (t) will continue to change in the direction of reducing the error until the error is eliminated, which enables the system to eventually transmit the penetration energy V. feedback (t) Locked at the target value E without steady-state error target superior.

[0058] The analog voltage signal V output by the integral controller curv (t), directly used as the overall curvature adjustment signal, is transmitted to the high-speed drive circuit of the liquid crystal modulation array. The drive circuit transmits this global voltage signal V...curv (t), proportionally converted to a global bias voltage V bias =k curv ⋅V curv (t), and superimposed on the base driving voltage of all liquid crystal microcells. This operation is equivalent to superimposing a global, controllable quadratic phase surface on top of the finely compensated wavefront generated by the fast loop. Changing V curv (t) means that the curvature of this global secondary phase surface is changed, thereby changing the effective numerical aperture of the main red light beam, that is, the degree of its convergence or divergence.

[0059] For example, in a deep tissue phototherapy application, the preset E target The required safe and effective energy density for treatment. Increased blood flow to the tissue surface leads to enhanced superficial scattering, increasing the primary red light energy V that penetrates to the deep treatment area. feedback When (t) decreases, the error e(t) becomes positive. The integral controller then integrates it in the positive direction and outputs V. curv (t) begins to increase slowly. This increase in voltage commands the liquid crystal modulation array to generate a converging wavefront with a larger overall curvature, making the beam more focused and thus increasing its penetration ability in turbid tissues, "pushing" more energy towards deeper targets. As V... feedback (t) rebounded and approached E due to this adjustment. target As the error e(t) decreases, the integration speed slows down, and eventually when V... feedback (t)=E target At that time, V curv (t) stabilizes at a new value, and the system has reached energy balance again under the new medium conditions.

[0060] Therefore, the depth-locking controller, together with the penetration energy detection module and the liquid crystal modulation array, constitutes an independent, slow, but precise energy-locking closed loop. It does not concern itself with the spatial distribution of energy, but only ensures that the total effective arrival energy remains constant. This dual-loop parallel architecture, where each loop performs its specific function, enables the system to simultaneously optimize both the "spatial efficiency" and "final performance" of the transmission process. This is a key innovative design for achieving fully autonomous, adaptive, and stable control of red light parameters in dynamically turbid media.

[0061] Furthermore, the differential control module, the liquid crystal modulation array, the phase-locked demodulation module, and the prediction module constitute a fast-response control loop; the depth-locked controller, the liquid crystal modulation array, and the penetration energy detection module constitute a slow-response control loop; the fast-response control loop is used to adjust the wavefront energy distribution of the main red light in real time based on the spatial distribution signal of the scattered light intensity and the turbidity change early warning signal; the slow-response control loop is used to stably adjust the overall penetration depth of the main red light based on the penetration energy feedback signal; the fast-response control loop and the slow-response control loop operate in parallel and independently, and work together on the liquid crystal modulation array.

[0062] Specifically, the aforementioned modules do not operate in isolation, but are integrated into two functionally defined, parallel, independent, and ultimately collaborative control loops through a sophisticated system architecture design: a fast-response control loop and a slow-response control loop. This dual-loop architecture is the core innovative mechanism by which this system achieves parallel and adaptive regulation of "spatial energy distribution optimization" and "overall penetration depth stability," ensuring that the system possesses both rapid response capability and ultimate effectiveness when facing dynamic turbid media.

[0063] The fast-response control loop is a high-bandwidth feedback and feedforward combined path operating at millisecond speeds. Its core objective is to sense and compensate for spatial energy distribution distortions caused by dynamic scattering of the medium in real time. The signal flow of this loop originates from the phase-locked demodulation module and the prediction module. The phase-locked demodulation module extracts the spatial distribution signal I of the scattered light intensity from the backscattered probe light in real time. s (θ), this signal directly reflects a snapshot of the medium's scattering characteristics in a spatial angle. The prediction module then analyzes I... s The temporal correlation of (θ) generates an early warning signal V of turbidity change before its abrupt change. alert It provides feedforward information.

[0064] These sensing signals are transmitted to the differential control module. This module will then transmit the spatial distribution I measured in real time. s (θ) and the preset reference distribution I characterizing the state of an ideal homogeneous medium ref (θ) is compared to generate a set of spatial error voltage signals V. err (θ) quantifies the enhancement or reduction of scattering in each direction. Subsequently, the differential control module will use V, which characterizes the current spatial error. err (θ) and the early warning signal V from the prediction module, which characterizes future trends. alert Simulation overlay and fusion are performed to form the final space driving signal V. drive (θ). This fusion process enables control commands to include both feedback corrections for existing distortions and preventative adjustments for impending disturbances.

[0065] Ultimately, Vdrive (θ) is applied at high speed to the liquid crystal modulation array. Based on this signal, the array generates a corresponding phase delay distribution across its millions of micro-cells, thereby reshaping the wavefront phase of the dominant red light in real time and with precision. For example, when a sudden increase in scattering is detected in a certain region of the medium, or when it is predicted that the region will become cloudy, the fast loop will instruct the liquid crystal modulation array to generate a phase pattern in the corresponding spatial region that deflects the beam energy, actively avoiding the highly scattering region and redirecting the energy to a channel with weaker scattering. The entire loop is implemented using both optical and analog circuitry, achieving extremely high response speeds specifically designed to handle rapid spatial fluctuations in the medium's scattering field.

[0066] The slow-response control loop is a narrow-bandwidth feedback path operating at speeds ranging from hundreds of milliseconds to seconds. Its core objective is to ensure that the total energy of the dominant red light reaching the target location remains constant regardless of how slowly the overall turbidity of the medium changes. The loop originates from a penetration energy detection module, which directly measures and outputs a feedback signal V proportional to the dominant red light power reaching the target. feedback The signal is sent to the depth lock-in controller, which is an analog integral controller. The controller will V feedback With a preset target voltage value E representing the desired energy dose target The comparison is continuously performed, and the deviation e(t) between the two is integrated over time to output the overall curvature adjustment signal V. curv (t). The characteristics of integral control determine that its response is slow and cumulative, but it can eventually eliminate steady-state error, that is, ensure V. feedback Converges exactly at E target .

[0067] V curv (t) is also sent to the liquid crystal modulation array. Unlike the fast loop which applies a fine spatial pattern, the slow loop's command is converted into a global bias voltage, superimposed on all microcells of the array. This operation is equivalent to superimposing a global, controllable quadratic phase surface onto the entire beam wavefront, thereby changing the overall convergence of the beam. For example, when the overall turbidity of the medium slowly increases, leading to a decrease in the penetration energy V... feedback During descent, the slow loop integrator outputs V. curv As (t) gradually increases, the liquid crystal modulation array is instructed to increase the overall convergence of the beam, making it more penetrating, thereby "pushing" the decreasing energy back to the target value. This loop does not care how the energy is distributed in space, but only focuses on maintaining a constant total energy at the final point of action.

[0068] The fast and slow response control loops are parallel and independent in both time and function. The fast loop processes high-frequency spatial signals and has a high control bandwidth; the slow loop processes low-frequency energy scalar signals and has a low control bandwidth. This bandwidth separation design effectively avoids oscillations caused by the coupling between the two loops.

[0069] Their convergence point ultimately converges at the liquid crystal modulation array. The array's high-speed drive circuit simultaneously receives the spatial drive signal V from the fast loop. drive (θ) and the overall curvature adjustment signal V from the slow loop curv (t). The driving circuit integrates the two to generate the final control voltage applied to each liquid crystal microcell. From the perspective of wavefront synthesis, the final phase distribution Φ generated by the liquid crystal modulation array total (x,y) is a fine phase pattern Φ generated by a fast loop for achieving spatial energy redistribution. fast (x,y), and the fundamental quadratic phase surface Φ generated by the slow loop to adjust the overall penetration capability. slow Linear superposition of (x,y): Φ total (x,y)=Φ fast (x,y)+Φ slow (x,y); For example, in underwater wireless optical communication applications, the fast loop cancels out instantaneous beam drift and distortion caused by turbulence and suspended particles in real time, ensuring the beam spot remains stable on the receiver; simultaneously, the slow loop slowly adjusts the initial divergence angle of the beam based on the average optical power measured by the receiver to compensate for the slow changes in overall water turbidity caused by tides, maintaining the long-term stability of the communication link budget. Through this dual-loop collaborative architecture of "fast loop adjusting distribution, slow loop locking energy," the system achieves multi-level, adaptive, and autonomous optimization and control of red light transmission parameters in dynamically turbid media.

[0070] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A red light parameter autonomous control system, characterized in that, include: The main light source and the detection light source are used to generate the main red light for treatment or communication and the detection red light for detection, respectively. The light emission module is optically connected to the main light source and the detection light source respectively. It is used to periodically encode and modulate the detection red light, and then combine the encoded and modulated detection red light with the main red light and emit it. The optical detection module is optically connected to the output optical path of the optical emission module. It is used to receive the coded detection light backscattered from the dynamic turbid medium and output multiple scattered light signals. A phase-locked demodulation module, electrically connected to the photodetector module, is used to synchronously demodulate the real-time spatial distribution signal of scattered light intensity from the multi-path scattered light signals; The prediction module, electrically connected to the phase-locked demodulation module, is used to calculate the optical cross-correlation function based on the spatial distribution signal of the scattered light intensity, and generate a turbidity change warning signal when the attenuation of the optical cross-correlation function value exceeds a preset threshold. The differential control module is electrically connected to the phase-locked demodulation module and the prediction module. It is used to generate a spatial error voltage signal based on the difference between the spatial distribution signal of the scattered light intensity and a preset reference spatial distribution signal, and to simulate and superimpose the turbidity change warning signal and the spatial error voltage signal to form the final driving signal. A liquid crystal modulation array is disposed in the optical path of the main red light and electrically connected to the differential control module, for spatially modulating the wavefront phase of the main red light according to the final driving signal; A penetration energy detection module is set at the expected target location of the dynamic turbid medium to receive the main red light after penetrating the dynamic turbid medium and generate a penetration energy feedback signal. A depth-locked controller is electrically connected to the penetration energy detection module and the liquid crystal modulation array. It is used to generate an overall curvature adjustment signal based on the deviation between the penetration energy feedback signal and a preset target energy value, and send it to the liquid crystal modulation array to adjust the overall phase curvature of the liquid crystal modulation array.

2. The autonomous red light parameter control system according to claim 1, characterized in that, The optical emitting module includes: The main red light emitting unit is optically connected to the main light source and is used to collimate and transmit the main red light; A detection light coding emission unit, optically connected to the detection light source, is used to generate the detection red light and perform periodic intensity coding modulation on it using a coding modulator; The beam combining unit optically connects the main red light emitting unit and the probe light encoding emitting unit, and is used to combine the modulated probe red light with the main red light.

3. The autonomous red light parameter control system according to claim 1, characterized in that, The optical detection module includes: The ring detector array unit is optically connected to the output optical path of the light emitting module. It is used to receive the coded detection light backscattered from the dynamic turbid medium and output multiple raw scattered photoelectric signals. The signal conditioning unit is electrically connected to the ring detector array unit and is used to amplify and filter the original scattered photoelectric signals of each channel, and output multiple channels of conditioned scattered light signals. The signal distribution and output unit is electrically connected to the signal conditioning unit and is used to output the multi-channel conditioned scattered light signals to the phase-locked demodulation module.

4. The autonomous red light parameter control system according to claim 1, characterized in that, The phase-locked demodulation module includes: Multiple demodulator units are electrically connected to each output terminal of the optical detection module in a one-to-one correspondence. Each demodulator unit is used to coherently demodulate one of the input scattered light signals at a reference frequency that is the same as the coding modulation frequency of the detected red light, and output the demodulated signal. The spatial distribution synthesis unit is electrically connected to the output terminals of all the demodulator units, and is used to receive all the demodulated signals and integrate them according to the spatial position correspondence to form the spatial distribution signal of the scattered light intensity.

5. The autonomous red light parameter control system according to claim 1, characterized in that, The prediction module includes: The signal buffer unit is electrically connected to the output terminal of the phase-locked demodulation module and is used to continuously store and output the spatial distribution signal of scattered light intensity at historical times. The analog optical correlation operation unit has two input terminals and one output terminal. Its first input terminal is electrically connected to the output terminal of the signal buffer unit to receive historical signals, and its second input terminal is electrically connected to the real-time output terminal of the phase-locked demodulation module to receive real-time signals. It is used to calculate the normalized cross-correlation function between the real-time signal and the historical signal and output the function value. The warning signal generation unit is electrically connected to the output terminal of the analog optical correlation calculation unit. It is used to compare the normalized cross-correlation function value with the preset threshold, and generate and output the turbidity change warning signal when the function value is lower than the threshold.

6. The autonomous red light parameter control system according to claim 1, characterized in that, The differential control module includes: A reference signal generation unit is used to store and output an electrical reference signal corresponding to the preset reference spatial distribution signal; The analog differential computing array has multiple parallel differential computing channels. Each differential computing channel has two input terminals and one output terminal. The first input terminal of each differential computing channel is electrically connected to the corresponding spatial position output terminal of the phase-locked demodulation module to receive a portion of the scattered light intensity spatial distribution signal. The second input terminal of each differential computing channel is electrically connected to the corresponding output terminal of the reference signal generation unit to receive a portion of the electrical reference signal. Each differential computing channel is used to calculate the difference between the two received signals and output it as the error voltage of the corresponding spatial position. The feedforward fusion node has multiple fusion input ports and one drive output port. Each fusion input port is electrically connected to the output of each channel of the analog differential computing array to receive the error voltage, and is additionally electrically connected to the output of the prediction module to receive the turbidity change warning signal. It is used to simulate superimpose the turbidity change warning signal with all the error voltages, and output the final drive signal from the drive output port to the liquid crystal modulation array.

7. The autonomous red light parameter control system according to claim 1, characterized in that, The liquid crystal modulation array includes: A liquid crystal phase modulation panel, the surface of which is divided into multiple independently controllable micro-units, is disposed in the optical path of the main red light; The high-speed drive circuit is electrically connected to the output of the differential control module and the output of the depth lock controller, and is independently connected to each micro-unit of the liquid crystal phase modulation panel. The high-speed drive circuit is used to independently generate and apply a drive voltage to each micro-unit according to the final drive signal to control its phase delay, and synchronously adjust the reference voltage bias applied to all micro-units according to the overall curvature adjustment signal.

8. The autonomous red light parameter control system according to claim 1, characterized in that, The penetration energy detection module includes: A narrowband optical filter is disposed in the receiving optical path at the expected target location to filter out the detection red light and ambient stray light, allowing only the main red light to pass through; The photoelectric conversion unit is optically connected to the output optical path of the narrowband optical filter and is used to convert the transmitted main red light into an electrical signal as the output of the penetration energy feedback signal.

9. The autonomous red light parameter control system according to claim 1, characterized in that, The depth lock controller is an analog integral controller. The input terminal of the depth lock controller is electrically connected to the output terminal of the penetration energy detection module to receive the penetration energy feedback signal, and the output terminal of the depth lock controller is electrically connected to the liquid crystal modulation array to output the overall curvature adjustment signal.

10. The autonomous red light parameter control system according to claim 1, characterized in that, The differential control module, the liquid crystal modulation array, the phase-locked demodulation module, and the prediction module constitute a fast-response control loop; the depth-locked controller, the liquid crystal modulation array, and the penetration energy detection module constitute a slow-response control loop; the fast-response control loop is used to adjust the wavefront energy distribution of the main red light in real time based on the spatial distribution signal of the scattered light intensity and the turbidity change early warning signal; the slow-response control loop is used to stably adjust the overall penetration depth of the main red light based on the penetration energy feedback signal; the fast-response control loop and the slow-response control loop operate in parallel and independently, and work together on the liquid crystal modulation array.

Images