Freepoint communication system and method for automobile headlamp based on stroboscopic modulation

Abstract

The application relates to the technical field of vehicle lamp communication, in particular to a strobe modulation-based automobile headlamp free light communication system and method, camera collects road condition information; a vehicle-mounted computer receives the road condition information and converts the same into an original bit stream, an anti-interference encoder generates an intermediate signal with a check code, a time-frequency interleaver is used for interleaving, a driving circuit is used for modulating and processing the interleaved intermediate signal, a vehicle lamp receives the modulated signal and emits a light signal in a strobe mode; a photoelectric sensor of a receiving end captures the light signal emitted by the vehicle lamp, a noise reduction and filtering module is used for wavelet noise reduction and filtering processing, a decision decoding module is used for frame positioning, demodulation and check generation of the processed light signal to generate a restored bit stream, the vehicle-mounted computer restores the restored bit stream into the road condition information, and a voice broadcast module broadcasts the road condition information in a voice mode. The application can meet the stable communication demand between vehicles and between a vehicle and a sensing device in a closed or semi-closed environment with seriously disturbed electromagnetic signals.

Description

Technical Field

[0001] This application relates to the field of automotive headlight communication technology, and in particular to a free-beam communication system and method for automotive headlights based on strobe modulation. Background Technology

[0002] With the development of intelligent connected vehicles, vehicle-to-vehicle communication has gradually evolved from traditional one-way prompting to a multimodal, real-time, high-bandwidth information interaction system. To improve road traffic safety and collaborative efficiency, various communication technologies are widely used in in-vehicle systems, especially radio frequency communication technology, which plays a crucial role in dynamic collaboration between vehicles and between vehicles and infrastructure. Simultaneously, to enhance vehicle perception and reaction speed, some intelligent driver assistance systems are attempting to integrate camera recognition, voice broadcasting, and communication functions to meet the needs of information acquisition and rapid response in complex traffic environments.

[0003] Currently, commonly used vehicle-to-everything (V2X) communication technologies primarily rely on radio frequency (RF) modules for information exchange, including 5G cellular-based vehicle-to-everything (V2X) architectures and short-range dedicated communication systems based on the DSRC protocol. These systems transmit commands through external base stations or onboard modules and combine them with sensors such as cameras and radar to acquire environmental information for autonomous driving assistance and traffic cooperative warnings. In addition, some vehicle systems incorporate visible light communication (VLC) technology, using vehicle lights for low-rate data transmission. Some solutions utilize the on / off state of LED lights for simple encoding, used for basic information transmission between vehicles.

[0004] However, existing radio frequency communication systems face serious challenges in enclosed environments with weak or completely shielded electromagnetic signals, such as tunnels, mountainous areas, and underground parking lots. Tunnels and mountainous areas naturally shield radio signals, while the high-density building structures and metal reflections in underground spaces make communication links unstable, preventing emergency information such as accident warnings and route guidance from being transmitted to vehicles in real time. Furthermore, current vehicle lighting systems primarily serve an illumination function and lack sufficient bandwidth and modulation capabilities to support reliable communication tasks in extreme environments. Although VLC technology can theoretically achieve low-interference transmission, in practical applications it is often affected by ambient light interference, making it difficult to maintain a stable link, leading to a decrease in signal-to-noise ratio and an increase in bit error rate. In summary, existing technologies are insufficient to meet the requirements for stable communication between vehicles and between vehicles and sensing devices in enclosed or semi-enclosed environments with severe electromagnetic signal interference. Summary of the Invention

[0005] This application provides a free-beam communication system and method for automotive headlights based on flicker modulation, which can meet the requirements for stable communication between vehicles and between vehicles and sensing devices in enclosed or semi-enclosed environments with severe electromagnetic signal interference. This application provides the following technical solution:

[0006] In a first aspect, this application provides a free-beam communication system for automotive headlights based on strobe modulation, comprising a transmitter and a receiver:

[0007] The transmitting end includes a camera, an on-board computer, an anti-interference encoder, a time-frequency interleaver, a driving circuit, and vehicle lights. The camera is used to collect road condition information of the vehicle's environment. The on-board computer is used to receive the road condition information sent by the camera and convert it into a raw bit stream. The anti-interference encoder is used to encapsulate the raw bit stream and perform redundant encoding to generate an intermediate signal with a checksum. The time-frequency interleaver is used to interleave the intermediate signal. The driving circuit is used to modulate the interleaved intermediate signal. The vehicle lights are used to receive the modulated signal and emit light signals in a strobe manner.

[0008] The receiving end includes a photoelectric sensor, a noise reduction and filtering module, a decision decoding module, an on-board computer, and a voice broadcasting module. The photoelectric sensor is used to capture the light signal emitted by the vehicle lights. The noise reduction and filtering module is used to perform wavelet noise reduction and filtering on the light signal. The decision decoding module is used to perform frame positioning, demodulation, and verification on the processed light signal to generate a restored bitstream. The on-board computer is used to restore the restored bitstream to road condition information. The voice broadcasting module is used to broadcast the road condition information via voice.

[0009] In one specific implementation, the anti-interference encoder divides the original bit stream into optical signal frames in units of 64 bytes. In each optical signal frame, each bit is represented by the on / off state of an LED bead. Logic 1 is represented by a light that is continuously lit for 15ms, and logic 0 is represented by a light that is continuously off for 15ms.

[0010] Each frame of data is marked with a start frame and an end frame. At the end of each frame, a special light pattern that flashes rapidly for three consecutive 30ms is introduced as a dynamic length declaration mark. After the start frame, three consecutive 30ms bright light pulses are inserted as a clock reference light signal for the receiver to calibrate the sampling timing.

[0011] A 10ms dark time slot is forcibly inserted after every 128 bits of data to break up sampling drift;

[0012] The CRC-8 algorithm is used to calculate the CRC-8 check code for each frame of data, and the check code calculated for each frame of data is converted into an optical feature pattern consisting of three consecutive 20-millisecond flashing pulses, which is then appended to the end of each frame of data.

[0013] Each processed optical signal is arranged in a uniform structure, including the start frame, sampling clock reference signal, effective data bit area, dynamic length declaration mark, CRC check code and termination frame.

[0014] In one specific implementation, the noise reduction and filtering module uses the Sym4 wavelet basis to perform five-level decomposition on the optical signal, removes fluorescent lamp flicker and sunlight interference through an adaptive soft thresholding method, and retains the headlight flicker signal characteristics between 5kHz and 7.5kHz; and filters out unwanted frequencies in the optical signal through an 80th-order FIR bandpass filter with a filtering range of 4.8kHz to 8kHz.

[0015] In one specific implementation, the decision decoding module determines the start and end positions of each frame in the received preprocessed optical signal and searches for a synchronization signal in the decoded data stream of the optical signal.

[0016] The synchronization signal consists of eight consecutive 50-millisecond high-frequency flashing pulses. The boundary of each frame of data is determined by the alternating flashing of the start frame and the end frame. A sliding window matching algorithm is introduced for real-time synchronization mark detection.

[0017] A sliding window is set in the bit stream, and the window is gradually slid in chronological order. At each sliding position, the bit sequence in the window is matched with the synchronization symbol pattern. The matching similarity score is calculated by a preset pattern determination function. If the matching score is higher than the threshold, it is determined to be the start frame or end frame position.

[0018] In a specific feasible implementation, after frame synchronization and positioning are completed in the decision decoding module, the FSK coherent demodulation stage is entered. The preprocessed optical signal is converted into a stable binary bit stream, and the received signal is divided into 100ms units. The continuous light intensity sampling data is divided into multiple independent time slots, and temporal correlation analysis is performed with two locally preset types of optical pulse templates in each time slot. In each 100ms time slot, two types of local reference templates are constructed respectively: one is... One represents a 5kHz constant brightness waveform template, and the other is... This represents the 7.5kHz scintillation mode template, and the light intensity signal segment sampled within each time slot is denoted as... The correlation between these two types of templates and the templates mentioned above was calculated, resulting in two sets of correlation scores:

[0019] ;

[0020] ;

[0021] in, Indicates the first The received light intensity value at each sampling point; , These represent the 5kHz and 7.5kHz templates at the [missing information - likely a specific digit or frequency]. Standard light intensity value at each sampling point This represents the total number of light intensity sampling points within a 100ms time slot. , The similarity score between the received signal and the two templates in the current time slot;

[0022] After calculating the two sets of correlation scores, compare the correlation values ​​between them. If If the value is positive, the bit represented by the time slot is determined to be logic 1; otherwise, it is determined to be logic 0. Through the decision process, the stroboscopic flash signal sequence is restored to a binary bit stream.

[0023] In one specific implementation scheme, after the bitstream is restored, the decision decoding module enters the data structure parsing stage to reconstruct the structured data frame.

[0024] Read two consecutive bytes after the start frame as the frame length field, and reconstruct the structured byte frame structure from the optical pulse bit stream that has completed frame synchronization;

[0025] For each reconstructed data frame, a CRC code is calculated to obtain the local verification result at the receiving end. At the same time, the original CRC check code attached by the transmitting end is read from the end of the data frame and compared with the locally calculated value. If the two match, the frame is determined to have passed the verification; otherwise, it is marked as a transmission distortion frame and enters the error handling process.

[0026] For frames that pass the CRC check, the decision decoding module further performs bit energy analysis, recording the coherent demodulation energy difference within each bit slot, i.e. The absolute value of the difference is considered to be in an energy ambiguity state when it is lower than the set threshold of 0.25 volts. It is marked as an ambiguous bit. In text reconstruction, the character mapped by this type of bit will be replaced with a special symbol to indicate that there is uncertainty in the receiver's information. The average bit error rate of the communication link is continuously counted. When more than three low-energy bits are detected in the same frame, the light intensity adaptive adjustment mechanism is automatically triggered to dynamically increase the light intensity of the LED transmitter or modify the modulation parameters.

[0027] For frames that fail CRC check, a forward error correction algorithm is executed, using a Hamming code-based error correction mechanism to locate and repair single-bit errors in the data area. If the repair is successful, the frame can still be received and used. If the repair fails, a selective retransmission request will be sent to the frame, requesting only the retransmission of the current distorted frame.

[0028] In one specific implementation scheme, in the voice broadcast module, the vehicle computer inputs the parsed road condition information into the speech synthesis coprocessor, which converts Chinese characters into phoneme sequences with tones through a Chinese character to Pinyin engine. The phoneme sequences are input into the speech synthesis chip, which uses an improved TD-PSOLA algorithm to generate audio signals. The generated audio stream is pre-emphasized and filtered. The filtered audio signal is then processed by the vehicle audio processor for dynamic range compression and multi-band spectral subtraction noise reduction. The processed audio signal is then broadcast directionally through the headrest directional speakers using the HRTF algorithm.

[0029] When the ambient noise inside the vehicle exceeds a preset threshold, the seat vibration device is automatically activated to issue a synchronous warning.

[0030] Secondly, this application provides a free-beam communication method for automotive headlights based on flicker modulation, employing the following technical solution:

[0031] A flicker-modulated free-beam communication method for automotive headlights, applied to the flicker-modulated free-beam communication system for automotive headlights described in the first aspect, includes:

[0032] The vehicle collects road condition information of the environment in which the vehicle is located through a camera and sends it to the vehicle computer. The vehicle computer receives the road condition information and converts it into a raw bit stream. The anti-interference encoder performs structural encapsulation and redundant encoding processing on the raw bit stream to generate an intermediate signal with a check code. The time-frequency interleaver performs interleaving processing on the intermediate signal. The driving circuit modulates the interleaved intermediate signal. The vehicle lights receive the modulated signal and emit light signals in a strobe manner.

[0033] The photoelectric sensor captures the light signal emitted by the vehicle headlights, the noise reduction and filtering module performs wavelet noise reduction and filtering on the light signal, the decision decoding module performs frame positioning, demodulation and verification on the processed light signal to generate a restored bitstream, the on-board computer restores the restored bitstream to road condition information, and the voice broadcast module broadcasts the road condition information via voice.

[0034] Thirdly, this application provides an electronic device, the device including a processor and a memory; the memory stores a program, the program being loaded and executed by the processor to implement a free-beam communication method for automotive headlights based on strobe modulation as described in the second aspect.

[0035] Fourthly, this application provides a computer-readable storage medium storing a program that, when executed by a processor, is used to implement a free-beam communication method for automotive headlights based on strobe modulation as described in the second aspect.

[0036] In summary, the beneficial effects of this application include at least the following:

[0037] (1) The closed-scene dedicated optical communication architecture of this application adopts dynamic anti-interference transmission technology, combined with block interleaving encapsulation and dual-frequency FSK modulation (logic 0 and logic 1 correspond to 5kHz and 7.5kHz strobe respectively), to achieve single frame transmission time compression to 0.3 seconds and communication bit error rate of less than 10%. -5 It can adapt to environments with intermittent and complex signals, such as tunnels and mountainous areas. Regarding ambient light suppression technology, the system utilizes Sym4 wavelet basis functions for five-level decomposition wavelet noise reduction, combined with an 80th-order FIR bandpass filter (operating frequency band 4.8kHz to 8kHz), effectively suppressing tunnel lighting flicker (100Hz) and other ambient light interference, significantly improving the signal-to-noise ratio of the effective optical signal to over 18dB. The overall technology significantly enhances the system's anti-interference capability and communication stability in harsh, enclosed environments, ensuring highly reliable optical communication between vehicles and between vehicles and sensing devices.

[0038] (2) By using low-power hardware reuse design, the vehicle headlight drive is effectively optimized. PWM dimming technology is used to achieve adaptive brightness control, maintaining a 100% duty cycle in strong light environments without the need for additional RF hardware support, significantly reducing system complexity and energy consumption. At the same time, a chain-like multi-hop communication network based on "camera-vehicle-vehicle" is constructed, which can quickly cover traffic flow within a 200-meter radius behind the vehicle within 1 second, forming an efficient emergency communication closed loop in a closed environment, improving the system's response speed and communication reliability, and meeting the low-power stable communication requirements in complex environments.

[0039] The transmitter controls the headlights to perform strobe modulation at a specific frequency (5kHz to 7.5kHz) via a high-speed drive module. FSK modulation is used to encode the digital signal into a sequence of light pulses, enabling covert and efficient transmission of optical communication signals. The receiver captures the light signal using a high-sensitivity BPW34S photoelectric sensor. Combined with multi-layer noise reduction processing based on the Sym4 wavelet basis and an 80th-order bandpass filter, ambient light and other interference are effectively filtered out, improving the signal-to-noise ratio. The decision decoding module employs frame synchronization positioning, time-domain correlated FSK coherent demodulation, and dynamic byte alignment algorithms to accurately recover the data frame structure. CRC checksum and bit energy analysis ensure data integrity and communication robustness, and forward error correction and retransmission mechanisms further reduce the bit error rate. The onboard computer converts the restored bitstream into road condition information and, combined with an optimized speech synthesis module, provides real-time voice broadcasting. The overall solution effectively avoids the impact of electromagnetic interference on communication, ensuring stable and reliable communication between vehicles and between vehicles and sensing devices in complex environments, thus improving the safety and responsiveness of the intelligent transportation system.

[0040] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, the preferred embodiments of this application are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0041] Figure 1 This is an overall architecture diagram of the automotive headlight free-beam communication system based on strobe modulation in this application embodiment.

[0042] Figure 2 This is a system architecture diagram of the transmitter in an embodiment of this application.

[0043] Figure 3 This is a schematic diagram of the encoding process of the original bit stream in the embodiments of this application.

[0044] Figure 4 This is a system architecture diagram of the receiving end in an embodiment of this application.

[0045] Figure 5 This is a schematic diagram of the voice playback process in an embodiment of this application.

[0046] Figure 6 This is a use case illustration of a specific embodiment in this application. Figure 1 .

[0047] Figure 7 This is a schematic diagram of a use case of binary encoding that poses a danger ahead, as described in the embodiments of this application.

[0048] Figure 8 This is a waveform diagram of signal modulation in an embodiment of this application.

[0049] Figure 9 This is a use case illustration of a specific embodiment in this application. Figure 2 .

[0050] Figure 10 This is a waveform comparison diagram of the noisy signal and the denoised signal in the embodiments of this application.

[0051] Figure 11 This is a schematic diagram of the waveform after signal filtering in an embodiment of this application.

[0052] Figure 12 This is a use case illustration of a specific embodiment in this application. Figure 3 .

[0053] Figure 13 This is a block diagram of an electronic device for free-beam communication of automotive headlights based on strobe modulation, as described in this application. Detailed Implementation

[0054] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this application, but are not intended to limit the scope of this application.

[0055] One embodiment of this application provides a free-beam communication system for automotive headlights based on flicker modulation, referring to... Figure 1 The system includes a transmitter and a receiver.

[0056] Reference Figure 2 The above is a system architecture diagram of the transmitter in this embodiment of the application. The transmitter includes a camera, an on-board computer, an anti-interference encoder, a time-frequency interleaver, a drive circuit, and vehicle lights.

[0057] The cameras are used to collect road condition information about the vehicle's environment. Specifically, the cameras are equipped with AI video surveillance technology with advanced algorithms to identify hazards in the vehicle's environment. If a hazardous situation is detected, the camera will edit the information into road condition information and send it to the target vehicle. This road condition information may include warnings such as "falling rocks ahead," "emergency stop," or "accident warning."

[0058] Alternatively, the camera in this application may also be other road sensing units, and this application does not limit the specific equipment used to collect road condition information.

[0059] It should be noted that the interaction between the camera and the target vehicle is the same as the subsequent interaction between vehicles. In practice, a laser generator can also be configured on the camera to function similarly to car lights. The interaction process between the camera and the target vehicle will not be described in detail here.

[0060] The onboard computer receives traffic information from the camera and converts it into a raw bitstream. Specifically, after receiving the traffic information from the camera, the user selects text, voice, or other methods to input the information into the onboard computer. The computer first converts the traffic information into Chinese text, then converts the Chinese text into a byte array using UTF-8 encoding, and splits each byte into 8 bits to form a continuous raw bitstream. This raw bitstream forms the initial headlight flashing signal. .

[0061] The anti-interference encoder is used to perform structural encapsulation and redundant encoding on the original bit stream to generate an intermediate signal with a check code.

[0062] Specifically, refer to Figure 3This is a schematic diagram illustrating the encoding process of the original bitstream in this embodiment. The anti-interference encoder divides the UTF-8 encoded original bitstream into optical signal frames in units of 64 bytes, and each frame is independently encapsulated and transmitted. In each optical signal frame, each bit is represented by the on / off state of an LED, where logic 1 is represented by a light that is continuously on for 15ms and logic 0 is represented by a light that is continuously off for 15ms, forming the basic bit encoding rhythm. To ensure that the receiver can accurately identify the start and end of each frame, the anti-interference encoder adds dedicated start and end frame identifiers to each frame. The start frame is encoded with a fixed synchronization symbol 0x55, which is repeatedly presented by alternating flashing patterns of 5ms on and 5ms off, used to establish a time reference between transmission and reception. The end frame uses a synchronization symbol 0xAA, implemented by alternating patterns of 20ms on and 20ms off, serving as a clear boundary signal for the receiver to identify the end of data. Meanwhile, to ensure the receiver can correctly determine the actual length of valid information contained in each frame, the anti-interference encoder introduces a set of three consecutive 30ms fast-flashing special light patterns at the end of each frame as a dynamic length declaration marker. This marker indicates that the receiver should consider the preceding bit stream as the valid data range, and any subsequent redundant padding or missampled data should be ignored. This mechanism is particularly suitable for data alignment and boundary judgment under variable-length frame structures, effectively avoiding bit errors and frame structure disorder caused by padding bits or channel distortion. After the start frame, the anti-interference encoder inserts three consecutive 30ms bright light pulses as clock reference light signals for the receiver to calibrate the sampling timing, improving the locking accuracy of the main data signal. These three pulses have fixed rhythm and intensity characteristics in the time domain, used by the receiver to establish a phase reference point for decoding sampling. By calculating the average interval of these three consecutive bright light segments, the receiver calibrates the start offset and sampling period of its local sampling timer, thereby achieving accurate synchronous sampling of subsequent main data segments.

[0063] Meanwhile, to further enhance the robustness of the communication protocol, the anti-interference encoder forcibly inserts a 10ms dark time slot after every 128 bits of data to break up sampling drift and prevent time accumulation errors from occurring during the integral sampling at the receiving end.

[0064] Furthermore, preferably, the anti-interference encoder uses the CRC-8 algorithm to calculate the CRC-8 checksum for each frame of data, and converts the calculated checksum into an optical feature consisting of three consecutive 20-millisecond flash pulses, which is then appended to the end of each frame of data. The optical signal formed after framing and protocol encapsulation of the above data is recorded as the intermediate signal. .

[0065] It should be noted that the intermediate signal after the above processing Each frame of optical signal is arranged according to a unified structure, including, in sequence, a start synchronization symbol, a sampling clock reference signal, a valid data bit area, a dynamic length declaration marker, a CRC checksum, and a stop synchronization symbol. This structure uses time-sequenced separation to achieve implicit protocol layering, ensuring that the receiver can accurately complete synchronization, decoding, verification, and data extraction even without out-of-band signaling.

[0066] The time-frequency interleaver is used to interleave intermediate signals. Specifically, the interleaving process scrambles the original bit transmission order so that it can effectively disperse continuous bit errors caused by sudden interference when reordering at the receiving end, thereby improving communication reliability. In addition, the interleaver dynamically adjusts the interleaving depth based on the vehicle speed parameters fed back in real time by the wheel speed sensor, deepening the interleaving at higher vehicle speeds to enhance communication robustness in high-speed driving environments.

[0067] The drive circuit modulates the interleaved intermediate signal to generate a modulated signal. Specifically, the drive circuit takes the interleaved intermediate signal as input and modulates it using Frequency Shift Keying (FSK) modulation to obtain a modulated signal used to control the headlight flashing state. In this embodiment, the FSK modulation method is adapted to a strobe coding scheme recognizable by automotive headlights: logic bit 0 is represented by a 5Hz low-frequency flash, i.e., alternating between on and off for 300ms and 200ms; logic bit 1 is represented by a 7.5Hz high-frequency flash, i.e., alternating between on and off for 67ms. The symbol rate is set to 1bps, i.e., each bit lasts for 1 second, ensuring that the receiver can accurately perceive frequency changes. In addition, to ensure that the receiver can correctly parse the modulated signal, the driving circuit generates a data frame containing multiple functional fields according to a predetermined data structure: the frame header is divided into a start marker, which uses a continuous 3-second high-frequency flash to indicate the start of the frame; followed by a length field, occupying a 4-second time window, which uses optical frequency encoding to represent the effective data length; next is the UTF-8 encoded data payload segment, which is sequentially converted into light pulses of the corresponding frequency; followed by a CRC check field, occupying 2 seconds, which uses a special frequency combination to form an optical feature pattern; finally, a continuous 3-second low-frequency flash constitutes an end marker, indicating the end boundary of the data frame.

[0068] In terms of hardware implementation, the driver circuit uses the L99LD01 LED driver chip from STMicroelectronics. This chip has precise PWM output control capabilities and can stably drive high-power automotive LED lighting assemblies.

[0069] The headlights are used to receive modulation signals and emit light signals in a strobe manner. Specifically, the headlights receive and amplify the modulation signals output by the drive circuit. This is then converted into an optical signal that can be perceived by the receiver in the form of stroboscopic motion. In the system architecture, the modulated signal... As an effective optical communication signal that has undergone encoding and modulation, its original output range is 0-1V TTL level. Directly driving the vehicle lamp load would result in insufficient LED junction voltage, typically requiring 3-3.6V, and the driving current would be below the threshold. The rated current of the vehicle lamp unit is usually 500mA. Therefore, a power amplifier circuit is configured at the end of the driving circuit, employing a complementary symmetrical common-emitter amplifier topology, and an AB class power amplifier stage is constructed based on the L99LD01 driver IC. This topology utilizes the linear amplification characteristics of bipolar transistors to increase the voltage swing and expand the current driving capability, effectively amplifying the TTL level signal to an amplitude range suitable for the LED vehicle lamp load, while maintaining the integrity of the modulation waveform in the time domain.

[0070] In summary, the transmitting system collects road condition information via a camera, which is then converted into a raw bitstream by the onboard computer. An anti-interference encoder encapsulates the bitstream with a frame structure and adds a CRC checksum to ensure data integrity and frame synchronization. A time-frequency interleaver dynamically adjusts the interleaving depth based on vehicle speed, effectively dispersing sudden interference and improving transmission robustness. The driving circuit employs an FSK modulation scheme adapted for automotive headlights, converting digital signals into sequences of light pulses at different frequencies. Simultaneously, a high-performance driver chip and power amplifier circuit ensure stable LED illumination and high-quality strobe output. The overall design fully considers the optical and electrical characteristics of vehicle headlight communication, as well as the vehicle's operating environment, balancing signal transmission accuracy, real-time performance, and anti-interference capabilities to ensure efficient and reliable free-light communication in complex driving environments.

[0071] Reference Figure 4 The diagram below shows the system architecture of the receiving end in this embodiment. The receiving end includes a photoelectric sensor, a noise reduction and filtering module, a decision decoding module, an on-board computer, and a voice broadcasting module.

[0072] The photoelectric sensor is used to capture the light signal emitted by the vehicle headlights. In this application, the BPW34S sensor is selected as the photoelectric sensor. Alternatively, other models can be selected. This application does not limit the specific model of the photoelectric sensor.

[0073] The noise reduction and filtering module is used for wavelet noise reduction and filtering of the optical signal. Specifically, the module first uses the Sym4 wavelet basis to perform a five-level decomposition of the optical signal. The Sym4 wavelet basis is a mathematical tool for multi-scale decomposition and reconstruction of signals, possessing good time-frequency localization characteristics, making it suitable for processing noise and detail information in non-stationary signals. Then, an adaptive soft thresholding method is used to remove fluorescent lamp flicker (approximately 100Hz) and lower-frequency sunlight interference, while retaining the headlight flicker signal characteristics between 5kHz and 7.5kHz. Next, the optical signal passes through an 80th-order FIR bandpass filter with a filtering range of 4.8kHz to 8kHz to further filter out unwanted frequencies and enhance the effective signal. The processed optical signal achieves a signal-to-noise ratio of over 18dB, ensuring effective signal extraction and accurate subsequent decoding.

[0074] The decision decoding module is used to perform frame positioning, demodulation, and verification on the processed optical signal to generate a restored bitstream.

[0075] First, the decision decoding module needs to determine the start and end positions of each frame in the received preprocessed optical signal. Specifically, it searches for a specific synchronization signal in the decoded data stream of the optical signal. The synchronization signal consists of eight consecutive 50-millisecond high-frequency flashing pulses, with the start frame being 0x55 and the end frame being 0xAA. The boundary of each frame is determined by the alternating flashing of these two pulses. In implementation, to improve the robustness of synchronization frame extraction under strong ambient light interference or optical signal distortion, the system introduces a sliding window matching algorithm for real-time synchronization marker detection. The specific method is as follows: a 500-bit sliding window is set in the bit stream, and the window is gradually slid in chronological order. At each sliding position, the bit sequence within the window is matched with a synchronization symbol pattern, and a matching similarity score is calculated using a preset pattern determination function. If the matching score is higher than a threshold (corresponding to a fault tolerance range of ±3 bits), it is determined as the start or end synchronization symbol position. This algorithm allows for local bit misalignment due to channel disturbances or sampling offsets, but it can still achieve effective capture of synchronization symbols in a high bit error rate environment.

[0076] Secondly, after frame synchronization and positioning are completed, the decision decoding module restores the received optical signal to the original digital information and reconstructs the structured data frame accordingly. Specifically, after frame synchronization and positioning are completed, the decision decoding module enters the FSK coherent demodulation stage, converting the preprocessed optical signal into a stable binary bit stream. In implementation, since the transmitter adopts a visible light frequency keying (FSK) modulation method, logic 0 corresponds to 5kHz constant brightness (equivalent to stable light intensity), and logic 1 corresponds to 7.5kHz frequency flicker (equivalent to periodic on / off). Therefore, the system divides the received signal into 100ms units, dividing the continuous light intensity sampling data into multiple independent time windows (slots), and performs time-domain correlation analysis with two locally preset optical pulse templates in each slot to determine the logic bit value represented by that slot. In each 100ms slot, two types of local reference templates are constructed: one is... One represents a 5kHz constant brightness waveform template, and the other is... , representing the 7.5kHz scintillation mode template. The light intensity signal segment sampled within each time slot is denoted as . The correlation scores are then calculated by performing a dot product and summation operation on the two types of templates mentioned above, resulting in two sets of correlation scores:

[0077] ;

[0078] ;

[0079] in, Indicates the first The received light intensity value at each sampling point; , These represent the 5kHz and 7.5kHz templates at the [missing information - likely a specific digit or frequency]. Standard light intensity value at each sampling point The total number of light intensity sampling points within a 100ms time slot (determined by the sampling rate, for example, 1000 points at a 10kHz sampling rate). , The similarity score between the received signal and the two templates in the current time slot.

[0080] After calculating the two sets of correlation scores, the correlation values ​​of the two sets are then compared. Specifically, if the correlation energy of the 7.5kHz frequency signal exceeds the 5kHz reference template by at least 30%, the bit value represented by that time slot is determined to be logic 1; otherwise, it is determined to be logic 0. This method is essentially a template-matching-based FSK coherent demodulation algorithm, which can effectively improve the noise immunity and decision accuracy of demodulation. Through the above decision process, the complete stroboscopic signal sequence can be restored to a stable binary bit stream.

[0081] After completing bitstream reconstruction, the decision decoding module enters the data structure parsing stage. The goal is to accurately reconstruct the structured byte frame structure and extract the valid data payload from the synchronized optical pulse bitstream. Specifically, after frame synchronization, two consecutive bytes (16 bits) are read sequentially after the start character as the frame length field. This field is encoded using the MSB-first (most significant bit first) principle, indicating the number of bytes of payload in this frame. The decoder parses this into an integer L, which serves as a control parameter for the number of bytes read subsequently. Then, the decoder extracts 8 × L bits consecutively from the bitstream in groups of 8 bits, dividing them into L standard bytes. Each byte represents a valid data character, which can then be directly mapped to ASCII or UTF-8 encoded characters to reconstruct the text content within the frame. However, in stroboscopic communication scenarios, due to external ambient light disturbances (such as direct sunlight, vehicle headlight flicker, etc.) and sampling jitter from the receiving sensor, the actual width and period of the light pulse may deviate from the design value, leading to deviations in bit boundary determination and thus disrupting the standard 8-bit byte division, resulting in misaligned code reading. To address this, a dynamic byte alignment algorithm is introduced. By analyzing the actual timing of the sampled waveform, the bit boundaries are automatically adjusted, thereby eliminating the impact of pulse drift on byte segmentation. Finally, based on dynamic boundary adjustment, L bytes of data are extracted, and a complete and clearly defined data frame is constructed by combining the start character and length field. This process ensures that even under conditions of significant light intensity disturbances or analog-to-digital conversion jitter, the byte sequence required by the protocol can be stably restored, effectively improving the system's communication robustness in complex optical environments.

[0082] Optionally, the specific process of the dynamic byte alignment algorithm is as follows: First, analyze the light intensity waveform segment corresponding to each bit, and accurately locate the rising / falling edge position of the light pulse through derivative operation or energy fluctuation detection to form a set of boundary candidate points; then, calculate the bit center back from the boundary candidate points according to the sampling frequency and the bit period defined by the protocol to correct the displacement error caused by sampling drift; then, statistically analyze the period stability and amplitude contrast of the extracted bits to dynamically determine whether the current bit group meets the uniformity requirements of the standard byte composition; when the cumulative bit interval drift exceeds a certain threshold, automatically adjust the grouping starting point to achieve sliding alignment and prevent miscutting or missed cutting.

[0083] Finally, after the reconstruction of the structured data frame is completed, the decision decoding module needs to perform CRC integrity check on each frame, and take measures such as error marking, reconstruction or requesting retransmission for frames that fail the check. The module sequentially performs CRC integrity check, bit energy analysis and text restoration processing to ensure the accuracy and identifiability of each frame of information.

[0084] Specifically, for each reconstructed frame of data, a CRC-8 checksum is first performed to determine if the data has been corrupted during transmission. Specifically, the generator polynomial x is used... 8 +x 4 +x 3 +x 2 +1, using a shift register to perform modulo-2 division, calculates the CRC code for each reconstructed data frame, and obtains the local checksum result at the receiving end. Simultaneously, the original CRC checksum appended by the transmitter (both positions are fixed) is read from the end of the data frame and compared with the locally calculated value. If they match, the frame is considered to have passed the checksum; otherwise, it is marked as a transmission distortion frame and enters the error handling process.

[0085] For frames that pass the CRC check, the decision decoding module further performs bit energy analysis, recording the coherent demodulation energy difference (i.e., ...) within each bit slot. The absolute value of the difference reflects the confidence level of its match with the standard template. When this difference is below a set threshold of 0.25 volts, the bit is considered to be in an energy ambiguity state and is marked as an ambiguity bit. In subsequent text reconstruction, the character mapped by this type of bit will be replaced with the special symbol "".<?> The message indicates uncertainty to the receiver. Furthermore, the average bit error rate of the communication link is continuously monitored. When three or more low-energy bits are detected consecutively within the same frame, an adaptive light intensity adjustment mechanism is automatically triggered to dynamically increase the light intensity of the LED transmitter or modify modulation parameters to improve the signal-to-noise ratio. For frames with failed CRC checks, a forward error correction (FEC) algorithm is first executed, using a Hamming code-based error correction mechanism to locate and repair single-bit errors in the data area. If FEC successfully repairs the error, the frame can still be received and used. If FEC fails to repair the error, a selective retransmission request (ARQ) is sent to the frame, requesting only the retransmission of the currently distorted frame to avoid full-link retransmission and improve communication efficiency. After the above verification, repair, and marking processes, the payload portion (UTF-8 encoded) is extracted and spliced ​​together to form a complete restored bitstream.

[0086] The onboard computer is used to restore the restored bitstream into traffic information. Specifically, after receiving the restored bitstream output by the decision decoding module, the onboard computer performs the following data parsing process to restore the traffic information. Upon receiving the restored bitstream, the onboard computer first groups it into byte sequences according to UTF-8 encoding rules and converts them into corresponding text strings. Then, based on a preset traffic information protocol, it performs simple pattern matching and parsing on the text content, mapping common commands to internal event types. Finally, the onboard computer encapsulates the parsed event types and optional additional parameters into structured traffic information.

[0087] The voice broadcast module is used to broadcast the road condition information in voice. Specifically, referring to Figure 5 , it is a schematic diagram of the voice playback process in the embodiments of this application. The in-vehicle computer inputs the parsed road condition information into the voice synthesis coprocessor. First, the Chinese characters are converted into a sequence of phonemes with tones through a Chinese-to-Pinyin engine (a look-up table algorithm based on the GB2312 code table). For example, "制动" is converted to "zhi4 dong4". In an emergency scenario, the prosody generation module uses DSP for real-time calculation to dynamically adjust the speech rate (increased from the normal 3 syllables per second to 4 syllables per second), the fundamental pitch (increased from 180 Hz to 220 Hz), and the end-of-sentence pitch (enhanced by an additional 30%) to enhance the sense of urgency and clarity of the broadcast. The entire processing flow is scheduled by the NXP S32K148 automotive-grade MCU, and the processing duration does not exceed 5 ms, outputting a phoneme stream with timestamps and a three-dimensional prosody feature vector (pitch, duration, energy). Then, the phoneme sequence is input into the iFlytek XFS5152CE voice synthesis chip, and the chip uses an improved TD-PSOLA algorithm to generate an audio signal. This algorithm first generates a fundamental period template (T0 = 1 / f0) according to the tone parameters and simulates the characteristics of the human vocal tract through a three-stage formant filter (F1 = 800 Hz, F2 = 1200 Hz, F3 = 2500 Hz). For emergency sentences, the chip particularly enhances the energy in the 3000 - 4000 Hz frequency band (+6 dB) to improve the voice penetration and clarity. The synthesis process calls the voice segment library in the IS42S16160D memory to generate an audio signal in 16 kHz sampling rate and 16-bit PCM format. The generated PCM audio signal then enters the ADI SHARC 21489 DSP for pre-emphasis filtering (+6 dB / oct) to ensure clear sound quality, and the total delay is controlled within 15 ms. Subsequently, the audio signal passes through the Cirrus Logic CS47L15 in-vehicle audio processor for dynamic range compression (threshold -20 dBFS, compression ratio of 4:1 in emergency mode) and multi-band spectral subtraction noise reduction processing. A notch filter is specifically implemented for the 200 - 500 Hz frequency band to suppress the engine idle noise, and at the same time, the voice gain is increased within the main noise frequency band to ensure the audibility of the voice in a complex in-vehicle sound field. Finally, the processed audio signal drives the Foster 423027 headrest directional speaker through the ST TDA7803 class-D power amplifier with an efficiency exceeding 92%, and combines the HRTF algorithm to create a spatial sound field exclusive to the driver to achieve high-quality directional broadcast. When it is detected that the in-vehicle ambient noise exceeds 65 dB, the system automatically activates the seat vibrator for synchronous warning to ensure that the driver can receive the warning information in a timely manner even in extreme noise environments such as tunnels.

[0088] In summary, the receiving system of this application integrates a photoelectric sensor, a noise reduction and filtering module, a decision decoding module, an onboard computer, and a voice broadcasting module. These components sequentially achieve high-precision acquisition of optical signals, effective noise suppression, stable decoding, and accurate reconstruction, ultimately transmitting road condition information to the driver in real-time via voice. Multi-layer wavelet noise reduction and bandpass filtering effectively filter out environmental interference, ensuring clear signals. Frame synchronization and FSK coherent demodulation techniques are used to accurately locate and reconstruct digital data frames, while CRC checksum and energy analysis ensure data integrity and robustness. The onboard computer parses road condition information, and the voice module, based on high-quality speech synthesis and dynamic adjustment, delivers clear, natural broadcasts with emergency alerts, significantly improving the reliability of vehicle communication and driving safety.

[0089] The transmitter controls the headlights to perform strobe modulation at a specific frequency (5kHz to 7.5kHz) via a high-speed drive module. FSK modulation is used to encode the digital signal into a sequence of light pulses, enabling covert and efficient transmission of optical communication signals. The receiver captures the light signal using a high-sensitivity BPW34S photoelectric sensor. Combined with multi-layer noise reduction processing based on the Sym4 wavelet basis and an 80th-order bandpass filter, ambient light and other interference are effectively filtered out, improving the signal-to-noise ratio. The decision decoding module employs frame synchronization positioning, time-domain correlated FSK coherent demodulation, and dynamic byte alignment algorithms to accurately recover the data frame structure. CRC checksum and bit energy analysis ensure data integrity and communication robustness, and forward error correction and retransmission mechanisms further reduce the bit error rate. The onboard computer converts the restored bitstream into road condition information and, combined with an optimized speech synthesis module, provides real-time voice broadcasting. The overall solution effectively avoids the impact of electromagnetic interference on communication, ensuring stable and reliable communication between vehicles and between vehicles and sensing devices in complex environments, thus improving the safety and responsiveness of the intelligent transportation system.

[0090] For example, refer to Figure 6, which is a schematic diagram of a use case in a specific embodiment of this application. In a one-way two-lane tunnel, the top monitoring camera detects a cow walking in the lane 30 meters ahead through AI video monitoring technology. Since traditional radio is completely ineffective due to tunnel shielding, the system immediately activates the optical communication emergency response: the camera transmits the coordinates of the falling rock to the vehicle in front in real time, the vehicle-mounted system quickly generates an instruction of "Danger ahead, avoid immediately", and transmits it through high-frequency light pulse coding of the headlight. The vehicle in front sends anti-interference signals in the enhanced light intensity mode to ensure stable transmission in the strong light environment of the tunnel; the rear vehicle captures the signals in real time through an optical sensor, analyzes the instruction within 50 milliseconds and triggers a voice alarm, and at the same time relays and forwards the warning information to the subsequent vehicles, covering the 200-meter traffic flow behind within 1 second, forming a chain-like risk avoidance network. The system breaks through the electromagnetic shielding limit and realizes millisecond-level multi-vehicle collaborative response in a closed scenario that completely relies on optical communication, effectively avoiding chain collisions caused by unexpected situations, highlighting its irreplaceability in extreme environments.

[0091] Specifically, first define the core parameters of the communication system: the start frame 0x55 and the end frame 0xAA, and set the FSK modulation parameters and noise parameters. Calculate the number of samples per bit to be 100, providing a time reference for subsequent signal generation. The text of the image recognized by the camera is converted into a UTF-8 byte sequence through the unicode2native function, and then each byte is expanded into an 8-bit binary stream using dec2bin. For example, the UTF-8 encoding of the Chinese character "在" is 3 bytes, which is converted into a 24-bit binary sequence to form a bits sequence for subsequent processing, such as Figure 7 shown, which is a schematic diagram of a use case of the binary encoding of danger ahead in an embodiment of this application. Taking "Danger ahead" as an example:

[0092] Front: 11100101 10001001 10001101;

[0093] Fang: 11100110 10010110 10111001;

[0094] Wei: 11100101 10001101 10110001;

[0095] Xian: 11101001 10011001 10101001;

[0096] Then process the original data in 64-byte blocks. Add a frame header 0x55, a 1-byte length field (binary representation of the data block length), a data payload, an 8-bit CRC check code, and a frame tail 0xAA to each block. Then customize the crc8 function to implement cyclic redundancy check: adopt a polynomial, calculate the check value byte by byte through shift exclusive OR operations, and append the result as an 8-bit binary to the signal at the end of the data block A sinusoidal carrier wave of corresponding frequency is generated for each bit: 0 corresponds to a 5kHz carrier wave, and 1 corresponds to a 7.5kHz carrier wave. By cyclically concatenating 100 sampling points for each bit, the signal is processed by the onboard computer. Modulated ,like Figure 8 The image shown is a waveform diagram of signal modulation in an embodiment of this application.

[0097] After spectrum shifting, information is encoded via phase change. This is then amplified by a power amplifier to drive the vehicle lights. The final signal output to the headlights is transmitted to the vehicle via a PWM dimming system. Figure 9 As shown. In communication systems based on vehicle headlight signal modulation transmission, ambient light sources (such as road lighting) and light interference noise from adjacent vehicle lights can cause signal waveform distortion and increased bit error rate. To construct a communication model that more closely resembles real-world scenarios, a multi-source optical noise interference mechanism needs to be introduced into the channel simulation to reproduce the true transmission characteristics of the optical communication link under complex lighting conditions. This involves using a clean modulation signal... Add the signal to the band-limited noise and normalize the amplitude to ensure the accuracy of the signal. The dynamic range is consistent with the original signal, avoiding additional errors introduced by amplitude distortion in subsequent processing. The onboard computer calls the wdenoise function to perform a 5-level Sym4 wavelet decomposition: multi-scale analysis of the noisy signal is performed, a general soft threshold is applied to high-frequency detail coefficients to suppress noise components, and the reconstructed signal is obtained. ,like Figure 10 The image shows a waveform comparison between the noisy signal and the denoised signal.

[0098] The onboard computer then applies an 80th-order FIR bandpass filter, followed by a zero-phase filter function to perform a second filtering of the noise-reducing signal, eliminating out-of-band noise remaining from wavelet processing, thus obtaining the filtered signal. ,like Figure 11 The image shown is a schematic diagram of the waveform after signal filtering. (Refer to...) Figure 12 The onboard computer demodulates the filtered signal and segments it into 100 sampling points / bits. Time-domain correlation operations are performed with 5kHz / 7.5kHz local carriers to calculate the energy difference. A sliding search is performed on the frame header 0x55 template in the decoded bitstream, recording the matching position. Valid data segments are extracted based on the length field after the frame header, and CRC checksums are separated. Data integrity is verified using CRC-8, and packets failing the verification are discarded. Finally, the original data is compared with the decoded bitstream, and the bit error rate is calculated. The verified binary stream is converted to text according to UTF-8 rules, and abnormal data is replaced with...<?> The final output is decoded text, which is then converted into speech by a speech conversion system.

[0099] One embodiment of this application also provides a free-beam communication method for automotive headlights based on flicker modulation, applied to the aforementioned free-beam communication system for automotive headlights based on flicker modulation. This method includes at least the following steps:

[0100] The vehicle collects road condition information about its environment using a camera and sends it to the onboard computer. The onboard computer receives the road condition information and converts it into a raw bit stream. An anti-interference encoder encapsulates the raw bit stream and performs redundant encoding to generate an intermediate signal with a checksum. A time-frequency interleaver interleaves the intermediate signal, and a drive circuit modulates the interleaved intermediate signal. The headlights receive the modulated signal and emit light signals in a strobe manner.

[0101] The photoelectric sensor captures the light signal emitted by the vehicle headlights. The noise reduction and filtering module performs wavelet noise reduction and filtering on the light signal. The decision decoding module performs frame positioning, demodulation and verification on the processed light signal to generate a restored bitstream. The on-board computer restores the restored bitstream to road condition information. The voice broadcast module broadcasts the road condition information via voice.

[0102] For relevant details, please refer to the above system implementation example.

[0103] Figure 13 This is a block diagram of an electronic device provided in one embodiment of this application. The device includes at least a processor 401 and a memory 402.

[0104] Processor 401 may include one or more processing cores, such as a quad-core processor or an octa-core processor. Processor 401 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 401 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 401 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 401 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0105] The memory 402 may include one or more computer-readable storage media, which may be non-transitory. The memory 402 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 402 is used to store at least one instruction, which is executed by the processor 401 to implement the strobe modulation-based free-beam communication method for automotive headlights provided in the method embodiments of this application.

[0106] In some embodiments, the electronic device may also optionally include: a peripheral device interface and at least one peripheral device. The processor 401, memory 402, and peripheral device interface can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface via a bus, signal line, or circuit board. Indicatively, peripheral devices include, but are not limited to: radio frequency circuits, touch displays, audio circuits, and power supplies.

[0107] Of course, electronic devices may also include fewer or more components, and this embodiment does not limit this.

[0108] Optionally, this application also provides a computer-readable storage medium storing a program that is loaded and executed by a processor to implement the strobe modulation-based free-beam communication method for automotive headlights described in the above method embodiments.

[0109] Optionally, this application also provides a computer product including a computer-readable storage medium storing a program, which is loaded and executed by a processor to implement the strobe modulation-based free-beam communication method for automotive headlights described in the above method embodiments.

[0110] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0111] The above embodiments merely illustrate several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A free-beam communication system for automotive headlights based on strobe modulation, comprising a transmitter and a receiver, characterized in that: The transmitting end includes a camera, an on-board computer, an anti-interference encoder, a time-frequency interleaver, a driving circuit, and vehicle lights. The camera is used to collect road condition information of the vehicle's environment. The on-board computer is used to receive the road condition information sent by the camera and convert it into a raw bit stream. The anti-interference encoder is used to encapsulate the raw bit stream and perform redundant encoding to generate an intermediate signal with a checksum. The time-frequency interleaver is used to interleave the intermediate signal. The driving circuit is used to modulate the interleaved intermediate signal. The vehicle lights are used to receive the modulated signal and emit light signals in a strobe manner. In the anti-interference encoder, the original bitstream is divided into optical signal frames in units of 64 bytes. In each optical signal frame, each bit is represented by the on / off state of an LED bead. Logic 1 is represented by a light that is continuously lit for 15ms, and logic 0 is represented by a light that is continuously off for 15ms. Each frame of data is marked with a start frame and an end frame. At the end of each frame, a special light pattern that flashes rapidly for three consecutive 30ms is introduced as a dynamic length declaration marker. After the start frame, three consecutive 30ms bright light pulses are inserted as a clock reference light signal for the receiving end to calibrate the sampling timing. A 10ms dark time slot is forcibly inserted after every 128 bits of data to break up sampling drift. The CRC-8 algorithm is used to calculate the CRC-8 check code for each frame of data, and the check code calculated for each frame of data is converted into an optical feature consisting of three consecutive 20-millisecond flash pulses, which is then appended to the end of each frame of data. Each frame of optical signal after processing is arranged in a uniform structure, including the start frame, sampling clock reference signal, effective data bit area, dynamic length declaration mark, CRC check code and termination frame in sequence. The receiving end includes a photoelectric sensor, a noise reduction and filtering module, a decision decoding module, an on-board computer, and a voice broadcasting module. The photoelectric sensor is used to capture the light signal emitted by the vehicle lights. The noise reduction and filtering module is used to perform wavelet noise reduction and filtering on the light signal. The decision decoding module is used to perform frame positioning, demodulation, and verification on the processed light signal to generate a restored bitstream. The on-board computer is used to restore the restored bitstream to road condition information. The voice broadcasting module is used to broadcast the road condition information via voice.

2. The automotive headlight free-beam communication system based on flicker modulation according to claim 1, characterized in that, In the noise reduction and filtering module, the Sym4 wavelet basis is used to decompose the optical signal into five layers. The adaptive soft threshold method is used to remove fluorescent lamp flicker and sunlight interference, while retaining the headlight flicker signal characteristics between 5kHz and 7.5kHz. An 80th-order FIR bandpass filter with a filtering range of 4.8kHz to 8kHz is used to filter out unwanted frequencies in the optical signal.

3. The automotive headlight free-beam communication system based on flicker modulation according to claim 1, characterized in that, In the decision decoding module, the start and end positions of each frame in the received preprocessed optical signal are determined, and a synchronization signal is searched in the decoded data stream of the optical signal. The synchronization signal consists of eight consecutive 50-millisecond high-frequency flashing pulses. The boundary of each frame of data is determined by the alternating flashing of the start frame and the end frame. A sliding window matching algorithm is introduced for real-time synchronization mark detection. A sliding window is set in the bit stream, and the window is gradually slid in chronological order. At each sliding position, the bit sequence in the window is matched with the synchronization symbol pattern. The matching similarity score is calculated by a preset pattern determination function. If the matching score is higher than the threshold, it is determined to be the start frame or end frame position.

4. The automotive headlight free-beam communication system based on strobe modulation according to claim 3, characterized in that, In the decision decoding module, after frame synchronization and positioning are completed, the FSK coherent demodulation stage is entered. The preprocessed optical signal is converted into a stable binary bit stream. The received signal is divided into 100ms units, and the continuous light intensity sampling data is divided into multiple independent time slots. In each time slot, temporal correlation analysis is performed with two locally preset types of optical pulse templates. In each 100ms time slot, two types of local reference templates are constructed: one is... One represents a 5kHz constant brightness waveform template, and the other is... This represents the 7.5kHz scintillation mode template, and the light intensity signal segment sampled within each time slot is denoted as... The correlation between these two types of templates and the templates mentioned above was calculated to obtain two sets of correlation scores: in, Indicates the first The received light intensity value at each sampling point; , These represent the 5kHz and 7.5kHz templates respectively at the [missing information - likely a typo, should be "at ... Standard light intensity value at each sampling point This represents the total number of light intensity sampling points within a 100ms time slot. , The similarity score between the received signal and the two templates in the current time slot; After calculating the two sets of correlation scores, compare the correlation values ​​between them. If If the value is positive, the bit represented by the time slot is determined to be logic 1; otherwise, it is determined to be logic 0. Through the decision process, the stroboscopic flash signal sequence is restored to a binary bit stream.

5. The automotive headlight free-beam communication system based on strobe modulation according to claim 4, characterized in that, In the decision decoding module, after the bit stream is restored, it enters the data structure parsing stage to reconstruct the structured data frame. Read two consecutive bytes after the start frame as the frame length field, and reconstruct the structured byte frame structure from the optical pulse bit stream that has completed frame synchronization; For each reconstructed data frame, a CRC code is calculated to obtain the local verification result at the receiving end. At the same time, the original CRC check code attached by the transmitting end is read from the end of the data frame and compared with the locally calculated value. If the two match, the frame is determined to have passed the verification; otherwise, it is marked as a transmission distortion frame and enters the error handling process. For frames that pass the CRC check, the decision decoding module further performs bit energy analysis, recording the coherent demodulation energy difference within each bit slot, i.e. The absolute value of the difference is considered to be in an energy ambiguity state when it is lower than the set threshold of 0.25 volts. It is marked as an ambiguous bit. In text reconstruction, the character mapped by this type of bit will be replaced with a special symbol to indicate that there is uncertainty in the receiver's information. The average bit error rate of the communication link is continuously counted. When more than three low-energy bits are detected in the same frame, the light intensity adaptive adjustment mechanism is automatically triggered to dynamically increase the light intensity of the LED transmitter or modify the modulation parameters. For frames that fail CRC check, a forward error correction algorithm is executed, using a Hamming code-based error correction mechanism to locate and repair single-bit errors in the data area. If the repair is successful, the frame can still be received and used. If the repair fails, a selective retransmission request will be sent to the frame, requesting only the retransmission of the current distorted frame.

6. The automotive headlight free-beam communication system based on strobe modulation according to claim 1, characterized in that, In the voice broadcast module, the vehicle computer inputs the parsed road condition information into the speech synthesis coprocessor, which converts Chinese characters into phoneme sequences with tones through a Chinese character to Pinyin engine. The phoneme sequences are then input into the speech synthesis chip, which uses an improved TD-PSOLA algorithm to generate audio signals. The generated audio stream undergoes pre-emphasis filtering, and the filtered audio signal is processed by the vehicle audio processor for dynamic range compression and multi-band spectral subtraction noise reduction. The processed audio signal is then broadcast directionally through the headrest directional speakers using the HRTF algorithm. When the ambient noise inside the vehicle exceeds a preset threshold, the seat vibration device is automatically activated to issue a synchronous warning.

7. A free-beam communication method for automotive headlights based on strobe modulation, applied to a free-beam communication system for automotive headlights based on strobe modulation as described in any one of claims 1 to 6, characterized in that, include: The vehicle collects road condition information of the environment in which the vehicle is located through a camera and sends it to the vehicle computer. The vehicle computer receives the road condition information and converts it into a raw bit stream. The anti-interference encoder performs structural encapsulation and redundant encoding processing on the raw bit stream to generate an intermediate signal with a check code. The time-frequency interleaver performs interleaving processing on the intermediate signal. The driving circuit modulates the interleaved intermediate signal. The vehicle lights receive the modulated signal and emit light signals in a strobe manner. The photoelectric sensor captures the light signal emitted by the vehicle headlights, the noise reduction and filtering module performs wavelet noise reduction and filtering on the light signal, the decision decoding module performs frame positioning, demodulation and verification on the processed light signal to generate a restored bitstream, the on-board computer restores the restored bitstream to road condition information, and the voice broadcast module broadcasts the road condition information via voice.

8. An electronic device, characterized in that, The device includes a processor and a memory; the memory stores a program, which is loaded and executed by the processor to implement a free-beam communication method for automotive headlights based on strobe modulation as described in claim 7.

9. A computer-readable storage medium, characterized in that, The storage medium stores a program, which, when executed by a processor, is used to implement the free-beam communication method for automotive headlights based on strobe modulation as described in claim 7.

Images