Vehicle-mounted station video backhaul and multi-mode audio interaction system

Abstract

The application relates to the technical field of wireless communication and vehicle-mounted multimedia interaction, in particular to a vehicle-mounted video backhaul and multi-mode audio interaction system, which comprises the following modules: a service shunting module, which is used for determining the priority label of audio stream data according to an identification result; a channel analysis module, which is used for evaluating the channel capacity and interference level of current time-frequency resources; a power domain allocation module, which is used for dynamically calculating a power allocation coefficient and dividing the total transmission power of a vehicle-mounted communication terminal into a first power level and a second power level; a superposition coding module, which is used for generating a multi-layer superposition signal; and a transmission control module, which is used for directly transmitting the multi-layer superposition signal to a base station side through a radio frequency unit in a scheduling-free transmission mode without sending a scheduling request signaling, so that the competition and conflict of high-definition video backhaul and key voice interaction for limited time-frequency resources in vehicle-mounted wireless communication are effectively solved, and efficient concurrency of heterogeneous services is realized.

Description

Technical Field

[0001] This invention relates to the field of wireless communication and in-vehicle multimedia interaction technology, specifically to an in-vehicle radio video transmission and multi-mode audio interaction system. Background Technology

[0002] In the vehicle-mounted wireless communication operation and maintenance environment, the vehicle-mounted radio needs to simultaneously undertake the task of high-volume monitoring video backhaul and critical multi-mode audio interaction. Since video streams have extremely high bandwidth requirements, while audio streams, especially emergency commands, have stringent requirements for real-time transmission and reliability, the service streams with different attributes exhibit significant heterogeneous characteristics in terms of physical layer resource consumption.

[0003] For concurrent transmission of such heterogeneous services, existing solutions generally adopt orthogonal resource partitioning mechanisms, that is, allocating video and audio services on non-overlapping time-frequency blocks through time-division or frequency-division multiplexing. This solution has a certain isolation capability in offline or low-load scenarios. However, in high-speed mobile and deep-fading channel environments, limited bandwidth resources, constrained by the spectral efficiency limit of orthogonal frequency division multiplexing, cannot simultaneously meet the throughput requirements of high-definition video and the real-time access needs of audio services. In addition, traditional transmission processes heavily rely on the scheduling request handshake mechanism of the media access control layer. When audio commands burst, a complete signaling interaction process must be performed, resulting in significant access latency and system overhead. When vehicles are in complex wireless environments, fixed resource allocation strategies are unable to cope with the fast fading characteristics of the channel, making critical audio commands susceptible to interference or unable to guarantee transmission continuity and stability due to time-frequency resource contention conflicts. Therefore, how to effectively solve the contention conflicts of time-frequency resources among heterogeneous services in vehicles, and improve the timeliness and reliability of critical audio service access while breaking through the spectral efficiency bottleneck of orthogonal multiplexing, has become an urgent technical problem to be solved. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a vehicle-mounted radio video transmission and multi-mode audio interaction system. Specifically, the technical solution of this invention includes:

[0005] The service routing module is used to receive video stream data and audio stream data to be transmitted, perform feature extraction and modality recognition on the audio stream data, and determine the priority label of the audio stream data based on the recognition results;

[0006] The channel analysis module is used to obtain the channel state information of the current wireless channel and to evaluate the channel capacity and interference level of the current time and frequency resources.

[0007] The power domain allocation module is used to dynamically calculate the power allocation coefficient according to the priority label and the channel state information, and divide the total transmit power of the vehicle communication terminal into a first power level and a second power level based on the power allocation coefficient, wherein the sum of the power values ​​of the first power level and the second power level is equal to the total transmit power.

[0008] The overlay encoding module is used to map the audio stream data to the first power level, map the video stream data to the second power level, and linearly overlay the mapped audio stream data and video stream data on the same time-frequency resource block to generate a multi-layer overlay signal;

[0009] The transmission control module is used to transmit the multi-layer superimposed signal directly to the base station via the radio frequency unit in a scheduling-free transmission mode without sending scheduling request signaling.

[0010] Preferably, the business routing module includes:

[0011] The instruction recognition unit is used to match the audio stream data with a preset key control instruction feature library. When the audio stream data is detected to contain the key control instruction feature, the priority tag is recorded as a high priority mode.

[0012] A conventional processing unit is used to record the priority tag as a normal priority mode and output the priority tag to the power domain allocation module when the key control command feature is not detected.

[0013] Preferably, the power domain allocation module includes:

[0014] A coefficient calculation unit is used to determine the value range of the power allocation coefficient based on the priority label;

[0015] When the priority label is in high priority mode, the coefficient calculation unit sets the power value of the first power level to be greater than the power value of the second power level, so that the audio stream data is overlaid on the video stream data as a strong signal.

[0016] When the priority label is in normal priority mode, the coefficient calculation unit sets the power value of the first power level to be less than the power value of the second power level, so that the audio stream data is embedded as a weak signal under the video stream data.

[0017] Preferably, the overlay encoding module includes:

[0018] The constellation mapping unit is used to modulate and map the data according to the power allocation coefficient, and to process the data stream allocated to the high power level using a low-order modulation method to generate a constellation point cloud with a large Euclidean distance, and to process the data stream allocated to the low power level using a high-order modulation method to generate fine constellation points around the constellation point cloud.

[0019] The signal synthesis unit is used to vector-add the two signals processed by the constellation mapping unit on the complex plane to generate the multi-layer superimposed signal.

[0020] Preferably, the transmission control module includes:

[0021] A blind detection identification unit is used to generate a preset feature sequence and embed the feature sequence into the pilot position of the multi-layer superimposed signal. The feature sequence is used to instruct the receiver to start the serial interference cancellation decoding process. The physical index number of the feature sequence is mapped to the priority label to assist the receiver in determining the iteration depth of serial interference cancellation.

[0022] The scheduling-free transmission unit is used to skip the scheduling request handshake process of the media access control layer and, in response to the arrival of the audio stream data, immediately occupy the physical resource block currently transmitting the video stream data to send the multi-layer overlay signal.

[0023] The preferred vehicle-mounted radio video transmission and multi-mode audio interaction system also includes:

[0024] The link adaptive module is used to process feedback information from the base station and perform closed-loop correction of transmission parameters;

[0025] The link adaptive module monitors the feedback of the hybrid automatic repeat request of the video stream data. When a negative response is received and the cause is determined to be superimposed interference, the modulation and coding strategy order of the video stream data is reduced to improve the demodulation robustness of the second power level data while keeping the first power level unchanged.

[0026] Preferably, the power domain allocation module further includes:

[0027] The threshold determination and adjustment unit is used to compare the signal-to-noise ratio parameter in the channel state information with a preset demodulation threshold.

[0028] When the signal-to-noise ratio parameter is lower than the preset demodulation threshold, a power compensation mechanism is triggered to increase the power ratio of the first power level until the estimated bit error rate of the audio stream data is lower than the preset bit error rate threshold.

[0029] When the signal-to-noise ratio parameter is greater than or equal to the preset demodulation threshold, the power ratio of the first power level is reduced to free up power margin for the second power level.

[0030] In the preferred vehicle-mounted radio video backhaul and multi-mode audio interaction system, the linear superposition performed by the superposition encoding module satisfies the power normalization constraint, and the generation of the multi-layer superimposed signal follows the following logic:

[0031] The first component signal is obtained by multiplying the modulation symbol of the audio stream data by the square root of the power allocation coefficient;

[0032] The second component signal is obtained by multiplying the modulation symbols of the video stream data by the square root of the difference after subtracting the power allocation coefficient;

[0033] The first component signal and the second component signal are directly added together to form the baseband complex signal to be transmitted.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] 1. This invention significantly improves spectrum efficiency and concurrency capabilities: By introducing a power domain multiplexing mechanism, it breaks through the spectrum efficiency bottleneck caused by traditional orthogonal resource partitioning; the system uses high-volume video streams as the carrier background for audio data to achieve non-orthogonal superposition transmission on the same time-frequency resource block, greatly improving the information carrying capacity per unit bandwidth; it effectively solves the competition conflict between high-definition video backhaul and key voice interaction on limited time-frequency resources in vehicle wireless communication, and realizes efficient concurrency of heterogeneous services;

[0036] 2. This invention significantly reduces the transmission latency of critical services: By adopting a scheduling-free transmission method combined with overlay coding technology, the complex scheduling request and authorization handshake process of the traditional media access control layer is completely eliminated; in response to sudden demands for audio commands, millisecond-level instant access to audio services can be achieved without waiting for base station resource allocation; this zero-wait transmission mechanism significantly reduces end-to-end latency, providing extremely high response time for vehicle-mounted emergency command and real-time dispatch.

[0037] 3. This invention enhances transmission reliability in complex environments: Through intelligent feature extraction and dynamic power domain allocation in the service offloading module, accurate identification and differentiated protection of service attributes are achieved; by real-time monitoring of channel status and execution of closed-loop power compensation, the system can ensure that critical audio commands are always at the optimal power level under harsh conditions such as high-speed movement or deep fading; this greatly enhances the demodulation robustness and anti-interference capability of the signal, ensuring the reachability of critical control information in complex wireless environments;

[0038] 4. This invention achieves dynamic optimization and stabilization of system performance: It establishes a complete link adaptive and closed-loop adjustment mechanism, realizing a dynamic balance between system throughput and transmission reliability; through linkage retransmission request feedback, it can flexibly adjust the modulation and coding strategy of the video stream according to the interference level while ensuring the continuity of audio interaction; this flexible trade-off scheme of exchanging video quality for audio timeliness effectively alleviates the fluctuations caused by superimposed interference and improves the overall operational stability of the vehicle communication system in dynamic channel environments. Attached Figure Description

[0039] The present invention will be further explained below with reference to the accompanying drawings and embodiments:

[0040] Figure 1 This is a structural diagram of the system of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0042] Example 1:

[0043] Please see Figure 1 A vehicle-mounted radio video transmission and multi-mode audio interaction system, comprising:

[0044] The service routing module is used to receive video stream data and audio stream data to be transmitted, perform feature extraction and modality recognition on the audio stream data, and determine the priority label of the audio stream data based on the recognition results;

[0045] The channel analysis module is used to obtain the channel state information of the current wireless channel and to evaluate the channel capacity and interference level of the current time and frequency resources.

[0046] The power domain allocation module is used to dynamically calculate the power allocation coefficient according to the priority label and the channel state information, and divide the total transmit power of the vehicle communication terminal into a first power level and a second power level based on the power allocation coefficient, wherein the sum of the power values ​​of the first power level and the second power level is equal to the total transmit power.

[0047] The overlay encoding module is used to map the audio stream data to the first power level, map the video stream data to the second power level, and linearly overlay the mapped audio stream data and video stream data on the same time-frequency resource block to generate a multi-layer overlay signal;

[0048] The transmission control module is used to transmit the multi-layer superimposed signal directly to the base station via the radio frequency unit in a scheduling-free transmission mode without sending scheduling request signaling.

[0049] In this embodiment, the vehicle-mounted radio and the vehicle-mounted communication terminal are different descriptions of the same device. This embodiment elaborates on the overall architecture and operation process of the vehicle-mounted radio video backhaul and multi-mode audio interaction system, aiming to solve the problem of competition for time and frequency resources by heterogeneous services in vehicle wireless communication scenarios. The service diversion module, as the input front end of the system, receives video stream data from the vehicle monitoring camera and audio stream data from the driver's microphone in real time. This module has a built-in feature extraction algorithm to perform modal recognition on the audio stream data to generate priority labels, thereby completing the digital definition of service attributes at the source end.

[0050] The channel analysis module obtains the current physical channel's channel state information (CSI) by periodically monitoring the channel state information reference signal sent by the base station, or by obtaining the CSI report measured and fed back by the base station based on the configured uplink sounding reference signal (SRS). The CSI includes at least the channel power gain of each subcarrier. and the estimated signal-to-interference-plus-noise ratio This module assesses the channel capacity and interference level of the current time-frequency resources based on the acquired CSI and by comprehensively considering indicators such as Reference Received Power (RSRP) and Interference Received Power (ISRP). The interference level... The equivalent signal-to-interference ratio can be calculated. To quantify, among which For reference signal power, The additive white Gaussian noise power is used to quantify and assess the channel capacity and interference level of current time-frequency resources used for video transmission, providing environmental parameters for subsequent power allocation; among which, It characterizes the channel power gain and is a dimensionless quantity. The signal-to-noise ratio is estimated based on the reference signal, and the unit is dB.

[0051] The power domain allocation module dynamically calculates the power allocation coefficient based on the priority label and the channel state information, dividing the total transmit power of the vehicle communication terminal into two complementary power levels in a mathematical dimension, namely the first power level and the second power level, ensuring that the sum of the two strictly conforms to the hardware radio frequency specifications. On this basis, the overlay coding module performs multi-service fusion at the physical layer, mapping the audio stream data to the first power level and the video stream data to the second power level, and linearly overlaying the two signals on the same time-frequency resource block to generate a multi-layer overlay signal. The transmission control module adopts a scheduling-free transmission method, and in response to the arrival of the audio stream data, skips the traditional scheduling request handshake process and directly drives the radio frequency unit to transmit the multi-layer overlay signal to the base station side.

[0052] This embodiment constructs a heterogeneous service concurrent transmission mechanism based on power domain multiplexing. In high-speed mobile scenarios, it uses high-volume video as the carrier background for low-data audio, achieving millisecond-level zero-wait access for audio services. At the same time, it breaks through the upper limit of spectrum efficiency of orthogonal segmentation, maintaining the coexistence of high-definition video backhaul and key voice interaction under the same bandwidth.

[0053] Example 2:

[0054] The business routing module includes:

[0055] The instruction recognition unit is used to match the audio stream data with a preset key control instruction feature library. When the audio stream data is detected to contain the key control instruction feature, the priority tag is recorded as a high priority mode.

[0056] A conventional processing unit is used to record the priority tag as a normal priority mode and output the priority tag to the power domain allocation module when the key control command feature is not detected.

[0057] This embodiment is a further specification of the internal logic of the service diversion module in Embodiment 1, aiming to achieve differentiated labeling of audio data with different levels of urgency; the instruction recognition unit calls the built-in key control instruction feature library, which stores preset voice acoustic fingerprints or specific audio instruction keyword features to define high-value information; the unit performs matching operations between the real-time input audio stream data and the feature library;

[0058] Specifically, the matching operation includes: dividing the original audio pulse code modulation (PCM) sequence into frames using a preset sliding window, and applying a Hamming window to reduce spectral leakage; extracting the Mel-frequency cepstral coefficients (MFCCs) of each frame of audio to construct a feature vector. The preset key control instruction feature library is constructed by collecting multiple key instruction audio samples with standard pronunciation in a clean environment during system initialization or training, and calculating the mean and variance of their MFCC feature vectors. It supports updates via OTA to expand the instruction set; and calculates real-time audio feature vectors. Cosine similarity between the vector and the baseline instruction vector stored in the feature library ;

[0059] To handle multi-command, continuous speech, and noisy environments, the system employs a dynamic threshold and state machine mechanism: basic matching threshold. The value is set to 0.85, which is based on offline simulation optimization with a target error rate of <1%. This value is also based on Monte Carlo simulations of the training set in an additive white Gaussian noise environment with a signal-to-noise ratio of 15 dB, optimized with a target instruction recall rate of 95%. In the continuous speech stream, the system maintains a length of... The sliding decision window, such as Frame, if inside the window If the proportion of frames exceeds a preset proportion, such as 60%, it is determined that a key instruction has been detected; for background noise, a Wiener filter can be optionally applied for noise reduction preprocessing before feature extraction.

[0060] In response to the detection that the audio stream data contains the key control command features, the system determines that the current service is an emergency command and marks the audio stream data as a high-priority mode; the regular processing unit processes the remaining background speech, and in response to the command recognition unit not detecting the key control command features, the system determines that the current service is a daily conversation or background noise and marks the audio stream data as a normal priority mode; the generated priority label is output to the power domain allocation module as a decision input for subsequent power level division;

[0061] This embodiment introduces an instruction feature matching mechanism to intelligently distinguish between commands and noise, avoiding indiscriminate preemption of high-power resources by non-critical audio data, thereby preventing unnecessary and continuous strong interference to the video stream and achieving intelligent balance of service experience in different scenarios. Through the above matching mechanism, in simulation tests, i.e., under an AWGN channel with a signal-to-noise ratio of 15dB, the system can achieve a recall rate of over 95% for the recognition of pre-stored critical instructions and a false trigger rate of less than 1%, providing a reliable basis for subsequent priority determination.

[0062] Example 3:

[0063] The power domain allocation module includes:

[0064] A coefficient calculation unit is used to determine the value range of the power allocation coefficient based on the priority label;

[0065] When the priority label is in high priority mode, the coefficient calculation unit sets the power value of the first power level to be greater than the power value of the second power level, so that the audio stream data is overlaid on the video stream data as a strong signal.

[0066] When the priority label is in normal priority mode, the coefficient calculation unit sets the power value of the first power level to be less than the power value of the second power level, so that the audio stream data is embedded as a weak signal under the video stream data;

[0067] The threshold determination and adjustment unit is used to compare the signal-to-noise ratio parameter in the channel state information with a preset demodulation threshold.

[0068] When the signal-to-noise ratio parameter is lower than the preset demodulation threshold, a power compensation mechanism is triggered to increase the power ratio of the first power level until the estimated bit error rate of the audio stream data is lower than the preset bit error rate threshold.

[0069] When the signal-to-noise ratio parameter is greater than or equal to the preset demodulation threshold, the power ratio of the first power level is reduced to free up power margin for the second power level.

[0070] This embodiment details the core algorithm and adaptive adjustment strategy of the power domain allocation module, aiming to cope with the fast fading characteristics of mobile channels; during system initialization, the power allocation coefficients... This represents the proportion of power allocated to the first power level, i.e., the audio stream, out of the total transmit power; its value range is... Its initial value Based on priority label settings: high priority mode is set to 0.7, and normal priority mode is set to 0.3; the power adjustment cycle is synchronized with the uplink transmission time interval (TTI), i.e., evaluation and potential adjustment are performed once every TTI, for example, 1ms; to avoid power oscillations or frequent switching, the system introduces a hysteresis mechanism: only when continuous... Each TTI, such as Inside All consistently below or higher ;in, To allow for hysteresis, for example, 1 dB, power compensation or release mechanisms are only triggered then; adjust the step size. It can also be based on and The difference is adaptively scaled; the larger the difference, the larger the step size can be to accelerate convergence.

[0071] The source is the real-time output of the channel analysis module, and its physical meaning is the estimated signal-to-interference-plus-noise ratio of the audio signal at the receiver, with the unit being dB;

[0072] The source is the preset system parameters, and the physical meaning is the minimum signal-to-noise ratio requirement under the target bit error rate, with the unit being dB;

[0073] The source is a mapping operation, and its physical meaning is the theoretical bit error rate obtained by looking up a table based on the current effective signal-to-noise ratio. It is dimensionless.

[0074] In response to Below The system triggers a power compensation mechanism, gradually increasing the power proportion of the first power level until the estimated bit error rate of the audio stream data is reached. The error rate is lower than the preset bit error rate threshold; specifically, the power compensation mechanism executes the following discrete iterative algorithm, and the system pre-stores an SNR-BER performance mapping table for the current modulation scheme:

[0075] Set iteration step size For example, if the value is 0.02, in the first... In the next adjustment cycle, if Less than Updated coefficients The calculation formula is:

[0076]

[0077] in, This represents the power allocation coefficient for the current k-th iteration period. The maximum power ratio is limited by hardware linearity, for example, 0.95; this value It is the linear operating range limit value measured based on the 1dB compression point P1dB of the RF power amplifier PA. This value represents the maximum power ratio that the audio layer signal can be allocated while ensuring no signal distortion. It is determined through calibration before leaving the factory and pre-stored in the device memory to prevent excessive peak-to-average power ratio of the superimposed signal from causing baseband signal distortion. The system updates the value based on the value in each iteration. Recalculate the effective signal-to-noise ratio and query the mapping table to obtain The iteration stops once the condition is met.

[0078] The calculation of the effective signal-to-noise ratio in the power domain needs to consider non-orthogonal superposition interference, and its calculation formula is as follows:

[0079]

[0080] in, This represents the total transmit power of the vehicle-mounted communication terminal. The current subcarrier channel power gain is obtained by the channel analysis module. For additive white Gaussian noise power, The term represents the co-channel interference caused by the video layer signal to the audio layer; through this formula, the system can quantify the impact of the video layer signal, i.e., the power domain part, as background interference on the audio demodulation performance.

[0081] Conversely, responding to Greater than or equal to The system reduces the power proportion of the first power level to free up power margin for the second power level. The adjustment logic follows:

[0082]

[0083] in, This is the baseline lower limit value under the current priority mode; it is 0.51 under high priority mode and 0.1 under normal mode.

[0084] In this embodiment, the signal-to-interference ratio of critical audio is forcibly increased by sacrificing video throughput in the deep fading region, ensuring the reachability of life-saving commands. Under high-quality channels, power is automatically released to video services, realizing dynamic optimization of power resources between reliability and throughput.

[0085] Example 4:

[0086] The overlay encoding module includes: a constellation mapping unit, used to modulate and map the data according to the power allocation coefficient, using a low-order modulation method to process the data stream allocated to the high power level to generate a constellation point cloud with a large Euclidean distance, and using a high-order modulation method to process the data stream allocated to the low power level to generate fine constellation points around the constellation point cloud;

[0087] The signal synthesis unit is used to vector-add the two signals processed by the constellation mapping unit on the complex plane to generate the multi-layer superimposed signal.

[0088] The linear superposition performed by the superposition coding module satisfies the power normalization constraint. The generation of the multi-layer superposition signal follows the following logic: the first component signal is obtained by multiplying the modulation symbol of the audio stream data by the square root of the power allocation coefficient.

[0089] The second component signal is obtained by multiplying the modulation symbol of the video stream data by the square root of the difference after subtracting the power allocation coefficient; the first component signal and the second component signal are directly added to form the baseband complex signal to be transmitted.

[0090] This embodiment specifically describes the mathematical process and mapping method of the superposition coding module in the physical layer symbol generation stage; the constellation mapping unit performs layered modulation, using a low-order modulation method for data streams allocated to high-power levels to generate base points with large Euclidean distances, and using a high-order modulation method for data streams allocated to low-power levels to generate fine constellation points distributed around the base points; the signal synthesis unit performs linear superposition based on power normalization constraints, and multi-layer superimposed signals. The generation follows the following formula:

[0091]

[0092] in, The source is the output of the signal synthesis unit, and its physical meaning is the baseband complex signal to be transmitted, which is carried on the physical resource block; The source is the modulation result of the audio stream data, and its physical meaning is the complex value modulation symbol of the audio data, with the power normalized to 1; The source is the power domain allocation module, and its physical meaning is the power allocation coefficient. The value range is 0, 1, which is consistent with the symbol definition in the previous embodiment. The source is the modulation result of the video stream data, and its physical meaning is the complex numerical modulation symbol of the video data, with the power normalized to 1;

[0093] The system will Multiply The square root of the first component signal is obtained. Multiply The square root of the signal yields the second component signal, and the two components are then added together to form the signal to be transmitted.

[0094] This embodiment constructs a hierarchical composite constellation diagram on the complex plane through vector addition. This structure naturally suits the serial interference cancellation (SIC) receiver at the receiving end, allowing the two signals to be completely overlapped in time and frequency but spatially separable, thus supporting non-orthogonal multiple access; for example, when At the same time, the audio layer uses QPSK modulation and the video layer uses 16QAM modulation. The receiver can first demodulate the high-power QPSK symbols through SIC, eliminate their interference, and then demodulate the remaining 16QAM symbols. Simulation shows that compared with traditional time division multiplexing, this scheme can improve the spectral efficiency by about 25% under the same bandwidth.

[0095] Example 5:

[0096] The transmission control module includes:

[0097] A blind detection identification unit is used to generate a preset feature sequence and embed the feature sequence into the pilot position of the multi-layer superimposed signal. The feature sequence is used to instruct the receiving end to start the serial interference cancellation decoding process.

[0098] Furthermore, a mapping relationship is established between the physical index number of the feature sequence and the priority label to assist the receiver in determining the iteration depth of serial interference cancellation;

[0099] The scheduling-free transmission unit is used to skip the scheduling request handshake process of the media access control layer and, in response to the arrival of the audio stream data, immediately occupy the physical resource block currently transmitting the video stream data to send the multi-layer overlay signal.

[0100] This embodiment provides a specific signaling support scheme for the scheduling-free mechanism in the transmission control module; the blind detection identification unit generates a preset feature sequence, which is selected from Zadoff-Chu sequences or pseudo-random noise sequences with good autocorrelation characteristics, and embeds the feature sequence into the pilot position of the multi-layer superimposed signal; the feature sequence serves as an implicit flag of the physical layer, used to instruct the receiver to continuously monitor and initiate the serial interference cancellation decoding process; the detection process of the feature sequence of the scheduling-free transmission unit at the receiver is as follows: the base station performs blind detection at the pilot position of each physical resource block, and calculates the cross-correlation value between the received signal and all possible feature sequences stored locally, including their different cyclic shift versions, corresponding to different priority tags and terminal identifiers; if the cross-correlation peak exceeds the preset detection threshold... If a valid superimposed signal is detected, the physical index number and collision avoidance flag are extracted, and the base station initiates the SIC decoding process.

[0101] Base station-side SIC decoding and link adaptive feedback: The base station determines the priority mode of the current signal based on the extracted physical index number, thereby determining the iteration depth of SIC decoding; for high-priority modes, the base station will receive the signal... Multiply by the inverse of the channel estimation Perform equalization and attempt to demodulate audio symbols using a low-end modem. After demodulation, the audio components are reconstructed and eliminated from the original received signal. For the remaining signal Perform equalization and demodulation of video symbols In normal priority mode, the order is reversed: video is demodulated first, then audio.

[0102] During the decoding process, if the Cyclic Redundancy Check (CRC) of any data stream fails within the preset maximum number of iterations, such as 2, the SIC decoding is determined to have failed, and a SIC decoding failure flag is fed back to the vehicle-mounted radio via the downlink. This flag can be carried in the HARQ feedback or in the dedicated Media Access Control (MACCE) element. Specifically, when the physical index number belongs to the first preset set, such as indexes 0-3, it is mapped to a high-priority mode, instructing the receiver to perform two serial interference cancellation iterations; when it belongs to the second preset set, such as indexes 4-7, it is mapped to a normal-priority mode, instructing the receiver to perform one serial interference cancellation iteration.

[0103] This embodiment eliminates the signaling interaction latency caused by the traditional handshake process by superimposing transmission on the existing video channel and cooperating with physical layer blind detection flags, thereby realizing real-time communication capabilities for audio services.

[0104] Example 6:

[0105] The system also includes a link adaptive module, which processes feedback information from the base station and performs closed-loop correction on the transmission parameters. The link adaptive module monitors the hybrid automatic repeat request feedback of the video stream data. When a negative response is received and the cause is determined to be superimposed interference, the modulation and coding strategy order of the video stream data is reduced to improve the demodulation robustness of the second power level data while maintaining the first power level unchanged.

[0106] This embodiment introduces a link adaptive module to address the video service quality fluctuations that may be caused by superimposed transmission. The link adaptive module continuously processes feedback information from the base station, particularly monitoring the Hybrid Automatic Repeat Request (HARQ) feedback for the video stream data. This module executes interference cause determination logic, comparing the audio transmission time window with the error time window. Specifically, the logic for determining the cause as superimposed interference is implemented through the following time intersection operation:

[0107] Let the transmission time interval of the video data block that received NACK be . The active transmission range of high-priority audio streams within the same time period is If the aforementioned time-domain overlap condition is satisfied... And the power allocation coefficient in the current cycle Exceeding the preset interference sensitivity threshold ,For example This threshold Through system-level simulation, under a typical vehicular channel model, the inflection point corresponding to the video stream block error rate (BLER) sharply increasing from 10% to 30% is determined. The value is used as a reference setting; at the same time, the serial interference feedback from the receiver is combined to eliminate the SIC decoding status. When the SIC decoding failure flag is true, the system determines that the negative response is likely caused by power domain superimposed interference, thereby avoiding the accidental triggering of video degradation operation in a low interference background.

[0108] In response to receiving a negative acknowledgment (NACK) and determining that the cause is superimposed interference, the system executes an adaptive degradation strategy to reduce the MCS order of the video stream data, for example, from 64QAM to 16QAM or reduce the bit rate. This operation aims to improve the demodulation robustness of the second power level data by reducing the spectral efficiency requirement of the video data while maintaining the first power level, i.e., the audio power, unchanged.

[0109] This embodiment, through its linkage with the HARQ mechanism, achieves a flexible trade-off between video quality and audio timeliness. When audio bursts, the video service automatically reduces its resolution to accommodate higher interference levels, thereby ensuring the continuity and stability of the overall system services. For example, when the system detects a sudden audio interruption... When high-priority audio transmission causes the video stream HARQNACK rate to exceed 20%, the video MCS is automatically reduced from 64QAM to 16QAM. After this adjustment, the video throughput decreases by about 50%, but its BLER under superimposed interference can be recovered to below 10%, while the audio command transmission success rate remains above 99.9%, achieving a dynamic resource trade-off with reliability as the priority.

[0110] This embodiment details the data flow, triggering sequence, and resource conflict coordination mechanism between key modules of the system.

[0111] Data flow and triggering sequence between modules: The service splitting module immediately outputs the updated priority tag to the power domain allocation module after processing each audio frame; at the beginning of each TTI, the power domain allocation module synchronously reads the latest priority tag from the service splitting module and the latest CSI from the channel analysis module, and executes the power allocation coefficients. The calculation or adjustment; the overlay encoding module is based on the updated The system superimposes and encodes the audio and video symbols to be transmitted within the current TTI; power allocation recalculation is triggered by the periodic clock of each TTI, rather than just by priority label changes, to ensure rapid response to channel changes; the entire data processing and transmission process can be completed within a single TTI; through system-level simulation, under the 3GPP V2X channel model, this scheme, compared to the traditional method requiring scheduling requests, can reduce the end-to-end transmission latency of audio commands from an average of over 10ms to less than 3ms, and ensure that the reliability of audio transmission in high-mobility scenarios is not less than 99.5%;

[0112] Dispatch-free transmission collision avoidance: When multiple vehicle-mounted stations attempt to preempt the same physical resource block for dispatch-free audio transmission at the same time, collisions may occur. This system coordinates through the following mechanism: Each vehicle-mounted station is assigned a unique pseudo-random sequence offset as its collision avoidance identifier upon power-up; when the audio service arrives and triggers dispatch-free transmission, the transmitter superimposes this unique identifier into the embedded blind detection feature sequence; after detecting the feature sequence, the base station receiver identifies multiple different collision avoidance identifiers and determines that a collision has occurred; at this time, the base station can send a simplified temporary authorization confirmation to one of the detected vehicle-mounted stations through the downlink control channel, if the identifier sequence has the strongest one, while not confirming other conflicting terminals. Vehicle-mounted stations that do not receive confirmation will back off for a random delay, the duration of which is calculated based on their unique identifier, and then retry dispatch-free access, thereby dispersing the collisions.

[0113] To verify the technical effectiveness of this system, tests were conducted in a vehicle-to-everything (V2X) simulation scenario with a bandwidth of 20MHz and a carrier frequency of 5.9GHz. The vehicle speed was set to 60km / h, and the EVA channel model was used. Experimental data showed that the signal-to-noise ratio... Below the demodulation threshold In weak coverage scenarios, such as When set to 6dB, the system automatically triggers the power compensation mechanism, adjusting the power distribution coefficient. Adjusted to 0.85; at this point, the bit error rate of the audio service... Maintain at This ensures reliable transmission of critical commands, while traditional orthogonal transmission schemes at this signal-to-noise ratio have an audio packet loss rate exceeding 5%; in a good channel environment with a signal-to-noise ratio increased to 15dB, the system will... When adjusted to 0.55, the total throughput of the superimposed coding transmission reached 45Mbps, which is about 40% higher than the 32Mbps of the traditional time-division multiplexing scheme. In the burst audio command access test, the scheduling-free transmission mechanism reduced the end-to-end average latency from 15ms of the traditional scheduling request mechanism to 2ms, which meets the timeliness requirements of vehicle emergency communication.

[0114] The above data verifies that the present invention effectively improves the overall spectrum efficiency and response speed of the system while ensuring the reliability of critical services.

[0115] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A vehicle-mounted radio video transmission and multi-mode audio interaction system, characterized in that: include: The service routing module is used to receive video stream data and audio stream data to be transmitted, perform feature extraction and modality recognition on the audio stream data, and determine the priority label of the audio stream data based on the recognition results; The channel analysis module is used to obtain the channel state information of the current wireless channel and to evaluate the channel capacity and interference level of the current time and frequency resources. The power domain allocation module is used to dynamically calculate the power allocation coefficient according to the priority label and the channel state information, and divide the total transmit power of the vehicle communication terminal into a first power level and a second power level based on the power allocation coefficient, wherein the sum of the power values ​​of the first power level and the second power level is equal to the total transmit power. The overlay encoding module is used to map the audio stream data to the first power level, map the video stream data to the second power level, and linearly overlay the mapped audio stream data and video stream data on the same time-frequency resource block to generate a multi-layer overlay signal; The transmission control module is used to transmit the multi-layer superimposed signal directly to the base station via the radio frequency unit in a scheduling-free transmission mode without sending scheduling request signaling. The power domain allocation module includes: A coefficient calculation unit is used to determine the value range of the power allocation coefficient based on the priority label; When the priority label is in high priority mode, the coefficient calculation unit sets the power value of the first power level to be greater than the power value of the second power level, so that the audio stream data is overlaid on the video stream data as a strong signal. When the priority label is in normal priority mode, the coefficient calculation unit sets the power value of the first power level to be less than the power value of the second power level, so that the audio stream data is embedded as a weak signal under the video stream data; The transmission control module includes: A blind detection identification unit is used to generate a preset feature sequence and embed the feature sequence into the pilot position of the multi-layer superimposed signal. The feature sequence is used to instruct the receiving end to start the serial interference cancellation decoding process. The scheduling-free transmission unit is used to skip the scheduling request handshake process of the media access control layer and, in response to the arrival of the audio stream data, immediately occupy the physical resource block currently transmitting the video stream data to send the multi-layer superimposed signal. The power domain allocation module further includes: The threshold determination and adjustment unit is used to compare the signal-to-noise ratio parameter in the channel state information with a preset demodulation threshold. When the signal-to-noise ratio parameter is lower than the preset demodulation threshold, a power compensation mechanism is triggered to increase the power ratio of the first power level until the estimated bit error rate of the audio stream data is lower than the preset bit error rate threshold. When the signal-to-noise ratio parameter is greater than or equal to the preset demodulation threshold, the power ratio of the first power level is reduced to free up power margin for the second power level.

2. The vehicle-mounted radio video transmission and multi-mode audio interaction system according to claim 1, characterized in that: The service diversion module includes: The instruction recognition unit is used to match the audio stream data with a preset key control instruction feature library. When the audio stream data is detected to contain the key control instruction feature, the audio stream data is marked as a high-priority mode. A conventional processing unit is used to mark the audio stream data as a normal priority mode and output the priority label to the power domain allocation module when the key control command feature is not detected.

3. The vehicle-mounted radio video transmission and multi-mode audio interaction system according to claim 1, characterized in that: The overlay encoding module includes: The constellation mapping unit is used to modulate and map the data according to the power allocation coefficient, and to process the data stream allocated to the high power level using a low-order modulation method to generate a constellation point cloud with a large Euclidean distance, and to process the data stream allocated to the low power level using a high-order modulation method to generate fine constellation points around the constellation point cloud. The signal synthesis unit is used to vector-add the two signals processed by the constellation mapping unit on the complex plane to generate the multi-layer superimposed signal.

4. The vehicle-mounted radio video transmission and multi-mode audio interaction system according to claim 1, characterized in that: Also includes: The link adaptive module is used to process feedback information from the base station and perform closed-loop correction of transmission parameters; The link adaptive module monitors the feedback of the hybrid automatic repeat request of the video stream data. When a negative response is received and the cause is determined to be superimposed interference, the modulation and coding strategy order of the video stream data is reduced to improve the demodulation robustness of the second power level data while keeping the first power level unchanged.

5. The vehicle-mounted radio video transmission and multi-mode audio interaction system according to claim 1, characterized in that: The linear superposition performed by the superposition encoding module satisfies the power normalization constraint, and the generation of the multi-layer superimposed signal follows the following logic: The first component signal is obtained by multiplying the modulation symbol of the audio stream data by the square root of the power allocation coefficient; The second component signal is obtained by multiplying the modulation symbols of the video stream data by the square root of the difference after subtracting the power allocation coefficient; The first component signal and the second component signal are directly added together to form the baseband complex signal to be transmitted.

Images