CN-121814930-B - Adaptive optimization method and system for optical transmission quality of video stream

CN121814930BCN 121814930 BCN121814930 BCN 121814930BCN-121814930-B

Abstract

The invention relates to the technical field of optical communication link regulation and control, in particular to a method and a system for adaptively optimizing optical transmission quality of video streams. According to the invention, a cross-layer mapping mechanism between nonlinear characteristics of an optical physical layer and coding parameters of a video application layer is constructed, so that the cross-phase modulation interference degree caused by multi-wavelength channel interaction can be quantized in real time, directional decoupling adjustment of a chromaticity quantization step length is triggered when channel degradation risks are perceived, noise interference of non-sensitive chromaticity data is restrained while brightness details are reserved, meanwhile, key reference nodes are accurately identified according to the referenced frequency of video frames in a time domain prediction sequence, probability shaping modulation protection based on information entropy is implemented for high-value data, limited optical transmission energy is concentrated to ensure the integrity of core prediction frames, global error code diffusion caused by single frame damage is effectively restrained, and threshold dynamic closed-loop correction logic based on real-time error symbol rate is matched.

Inventors

WANG LIN
FENG XINXIANG
WANG YI
YU LI
CHENG JUNFENG
TANG WANSONG
Lai Xianghong

Assignees

重庆君歌电子科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260310

Claims (8)

1. The adaptive optimization method for the optical transmission quality of the video stream is characterized by comprising the following steps of: S1, collecting instantaneous fiber-entering optical power of a current video stream wavelength channel and an adjacent video stream wavelength channel, calculating power fluctuation statistical variance, and calculating a cross phase modulation interference index by combining a set fiber nonlinear coefficient; s2, comparing the cross phase modulation interference index with a preset light transmission interference warning threshold value, and setting a chroma quantization step decoupling configuration according to a chroma quantization parameter of head information configuration of the current video slice to be encoded; s3, analyzing a video compression code stream generated based on the chroma quantization step decoupling configuration code, identifying a unique frame number identification of a current video stream prediction frame in a video sequence, and counting the total number of times the video stream prediction frame is referenced; S4, comparing the total quoted times of the video stream predicted frames with a preset video frame importance threshold value, extracting target frame compression data of the current video stream predicted frames according to the comparison result, and constructing an optical symbol probability shaping mapping table by combining a preset standard optical communication constellation diagram; S5, performing symbol mapping and light modulation transmission on the target frame compressed data based on the optical symbol probability shaping mapping table, and correcting an optical transmission interference warning threshold and a video frame importance threshold according to a transmission result to obtain a video streaming light transmission quality self-adaptive optimization result; The step of obtaining the cross phase modulation interference index specifically comprises the following steps: S111, setting a sliding time window based on a video stream symbol rate period, acquiring a material refractive index parameter and an effective mode field area of an optical fiber transmission medium carrying a current video stream wavelength channel, setting an optical fiber nonlinear coefficient, acquiring instantaneous fiber-in optical power of the current video stream wavelength channel and instantaneous fiber-in optical power of an adjacent video stream wavelength channel according to a preset sampling period, and constructing an optical fiber channel basic physical parameter set; S112, mapping the instantaneous fiber-entering optical power of a plurality of adjacent video stream wavelength channels in the fiber channel basic physical parameter set into a sliding time window, and counting the adjacent channel power time domain fluctuation variance of the instantaneous fiber-entering optical power of each sampling point in the sampling period and the average instantaneous fiber-entering optical power in the sliding time window; S113, calculating the product of the adjacent channel power time domain fluctuation variance, the instantaneous fiber-in optical power of the current video stream wavelength channel and the fiber nonlinear coefficient as a cross phase modulation interference index; the step of obtaining the chroma quantization step length decoupling configuration specifically comprises the following steps: S211, when the cross phase modulation interference index exceeds a preset light transmission interference warning threshold value, configuring a chromaticity quantization parameter of head information for a current video slice to be coded, reading and keeping a brightness component quantization parameter unchanged, and identifying an original blue color difference component quantization parameter and an original red color difference component quantization parameter in a video slice head information area to generate a to-be-corrected chromaticity component parameter set; s212, introducing a preset positive integer offset, calculating the superposition sum of the original blue color difference component quantization parameter and the positive integer offset in the to-be-corrected chroma component parameter set, and simultaneously calculating the superposition sum of the original red color difference component quantization parameter and the positive integer offset to obtain a chroma component bias correction value; and S213, summarizing the chroma component bias correction value and the brightness component quantization parameter, and repackaging according to a format of a video coding standard to be used as a chroma quantization step length decoupling configuration.
2. The method according to claim 1, wherein the cross-phase modulation interference index is obtained by calculating a continuous product of a power fluctuation statistical variance, an instantaneous fiber-in optical power of a current video stream wavelength channel, and an optical fiber nonlinear coefficient, the chroma quantization step decoupling configuration includes a sum of an original blue color component quantization parameter and a preset positive integer offset, a sum of an original red color component quantization parameter and a preset positive integer offset, the video stream prediction frame is calculated by counting a total number of references of a subsequent video stream frame, the reference total number being identified by a unique frame number directly pointing to the current video stream prediction frame through a motion vector, the optical symbol probability shaping mapping table includes a coordinate mapping relation between a symbol to be transmitted and each optical communication constellation point in a standard optical communication constellation, and an occurrence probability of each optical communication constellation point, and the video stream optical transmission quality self-adaptive optimization result includes an optimized optical transmission interference warning threshold and an optimized video frame importance threshold.
3. The method for adaptively optimizing the optical transmission quality of a video stream according to claim 1, wherein the step of acquiring the total number of times the predicted frame of the video stream is referenced is specifically: S311, performing compression coding on the current video stream by utilizing the chroma quantization step decoupling configuration, generating a video compression code stream, analyzing a coding strip head grammar structure of the video compression code stream, positioning a network abstraction layer for bearing predicted frame data of the current video stream, extracting a unique frame number identification of the predicted frame of the current video stream in a video sequence from the network abstraction layer, and generating a predicted frame unique sequence identification code; S312, retrieving slice header information of each subsequent video stream frame positioned behind the predicted frame of the current video stream in the video compression code stream, extracting a reference image list defined in the slice header information, matching and verifying the unique sequence identification code of the predicted frame with each reference image in the reference image list, and screening a reference index data set of the subsequent frames; s313, matching the entries in the reference index dataset of the subsequent frames, which directly point to the unique sequence identification code of the predicted frame through the motion vector, accumulating the quantity of all matched entries, counting the frequency of the reference of the unique sequence identification code of the predicted frame by the subsequent video stream frames, and generating the total reference times of the predicted frame of the video stream.
4. The method for adaptively optimizing optical transmission quality of a video stream according to claim 3, wherein the step of obtaining the optical symbol probability shaping mapping table comprises the steps of: S411, when the total number of cited video stream predicted frames is larger than a preset video frame importance threshold, extracting target frame compressed data corresponding to the current video stream predicted frames from a video compressed code stream, determining a modulation order of a standard optical communication constellation diagram adapting to the target frame compressed data, calculating a theoretical maximum information entropy value based on the modulation order, and simultaneously calculating a difference value between the theoretical maximum information entropy value and a preset probability plastic redundancy adjustment amount to be used as a target information entropy dispersion value; S412, reversely deducing a probability shaping scaling factor based on the target information entropy dispersion value, establishing a coordinate mapping relation between a symbol to be transmitted and each optical communication constellation point in a standard optical communication constellation, determining the position of each optical communication constellation point in a complex plane, calculating the geometric Euclidean distance of each optical communication constellation point relative to the origin of the standard optical communication constellation, and generating a constellation point geometric distribution characteristic parameter; s413, calling the geometric distribution characteristic parameters of the constellation points, substituting the probability shaping scaling coefficient and the geometric Euclidean distance into a Maxwell-Boltzmann distribution function, calculating the occurrence probability of each optical communication constellation point under the non-uniform probability distribution, and reconstructing the mapping rule of the constellation points according to the occurrence probability to construct an optical symbol probability shaping mapping table.
5. The method for adaptively optimizing optical transmission quality of a video stream according to claim 4, wherein the step of obtaining the adaptive optimization result of optical transmission quality of the video stream is specifically: S511, performing symbol mapping and optical modulation transmission on the target frame compressed data based on the optical symbol probability shaping mapping table, acquiring a real-time optical transmission symbol error rate returned by an optical receiving end, and calculating a difference value between the real-time optical transmission symbol error rate and an optical link safety base line when the real-time optical transmission symbol error rate is larger than a preset optical link safety base line, so as to generate an optical transmission symbol error rate overrun amplitude; S512, calculating the product of the optical transmission interference warning threshold and the optical transmission symbol error rate overrun, determining the down-regulation amplitude of the interference threshold, simultaneously calculating the product of the video frame importance threshold and the optical transmission symbol error rate overrun, and determining the down-regulation amplitude of the importance threshold as a dual threshold dynamic down-regulation correction quantity; S513, based on the dual-threshold dynamic adjustment correction quantity, calculating the difference value between the light transmission interference warning threshold and the corresponding interference warning threshold adjustment correction quantity, calculating the difference value between the video frame importance threshold and the corresponding video frame importance threshold adjustment correction quantity, respectively obtaining the optimized light transmission interference warning threshold and the optimized video frame importance threshold, and generating a video stream light transmission quality self-adaptive optimization result.
6. A method of adaptively optimizing optical transmission quality of a video stream according to claim 3, wherein the reference picture list includes relative index addressing data based on picture play order count and absolute position addressing data based on picture play order count; The relative index addressing data based on the image playing sequence count comprises a difference value calculation parameter between the sequence number of the current follow-up video stream frame and the sequence number of the reference frame stored in the decoding image buffer; The absolute position addressing data based on the image play order count includes a sequence number least significant bit or sequence number corresponding to the predicted frame unique sequence identification code.
7. The method for adaptively optimizing optical transmission quality of a video stream according to claim 4, wherein for the target information entropy dispersion value, the formula is adopted: ; In the formula, Represents the entropy dispersion value of the target information, Representing the modulation order of a standard optical communication constellation, Representing the basic redundancy scaling factor of the system, The importance weighting coefficient is represented by a set of importance weighting coefficients, Indicating the total number of times a predicted frame of the video stream is referenced, Representing a preset video frame importance threshold, Is a mathematical correction constant.
8. An optical transmission quality adaptive optimization system of a video stream, characterized in that the optical transmission quality adaptive optimization method of a video stream according to any one of claims 1-7, the system comprising: The optical interference assessment module is used for collecting instantaneous fiber-entering optical power of a current video stream wavelength channel and an adjacent video stream wavelength channel, calculating power fluctuation statistical variance, and calculating a cross phase modulation interference index by combining a set optical fiber nonlinear coefficient; The quantization decoupling configuration module is used for comparing the cross phase modulation interference index with a preset light transmission interference warning threshold value, and setting a chroma quantization step length decoupling configuration for configuring a chroma quantization parameter of head information for a current video slice to be encoded according to a comparison result; The predicted frame statistics module analyzes the video compression code stream generated based on the chroma quantization step decoupling configuration code, identifies a unique frame number identification of a current video stream predicted frame in a video sequence, and counts the total number of times the video stream predicted frame is referenced; The important frame mapping module is used for comparing the total quoted times of the video stream predicted frames with a preset video frame importance threshold value, extracting target frame compression data of the current video stream predicted frames according to the comparison result, and constructing an optical symbol probability shaping mapping table by combining a preset standard optical communication constellation diagram; And the adaptive transmission optimization module performs symbol mapping and optical modulation transmission on the target frame compressed data based on the optical symbol probability shaping mapping table, corrects an optical transmission interference warning threshold and a video frame importance threshold according to a transmission result, and obtains a video stream optical transmission quality adaptive optimization result.

Description

Adaptive optimization method and system for optical transmission quality of video stream Technical Field The invention relates to the technical field of optical communication link regulation, in particular to a method and a system for adaptively optimizing optical transmission quality of video streams. Background The technical field of optical communication link regulation mainly relates to a related technology for managing and regulating an optical signal transmission process in an optical fiber transmission system, and generally expands around factors such as dispersion attenuation noise, bandwidth limitation and the like of an optical signal in the transmission process, and ensures stable transmission and controllable operation of information in an optical communication network by carrying out coordinated control on parameters of links of the link. The adaptive optimization method for the optical transmission quality of the video stream refers to a processing mode for adjusting the data transmission quality of the video stream in an optical communication link. The video stream is typically layered encoded or multi-rate encoded before video transmission, and the corresponding video stream level is selected according to preset optical link parameters (e.g., fixed optical power level, fixed modulation format, and fixed channel bandwidth). In the transmission process, the video code rate is switched or reselected according to the error rate, packet loss statistics or signal-to-noise ratio threshold information returned by the receiving end, so that the matching adjustment between the video stream and the optical transmission condition is realized. In the prior art, the video stream level is regulated by depending on preset static link budget or a passive feedback mechanism based on average error statistics, the operation mode depending on fixed parameters and hysteresis feedback is difficult to capture nonlinear physical layer damage caused by transient state in optical fiber transmission, especially, rapid optical power fluctuation caused by cross phase modulation among multiple channels often breaks signal integrity before the compensation mechanism is triggered, and the time correlation and level importance difference among video frames are ignored by adopting a coarse granularity regulation strategy of the whole switching code stream level, so that differential protection cannot be provided for key reference frames when burst interference is faced, the diffusion effect of error codes in the time domain is extremely easy to be caused, the mosaic or the katon phenomenon which is difficult to recover is caused to appear on video pictures at a receiving end, the fine utilization efficiency of optical spectrum resources is reduced, and stable visual experience cannot be maintained under a dynamically-changed channel environment. Disclosure of Invention The invention aims to solve the defects in the prior art, and provides a self-adaptive optimization method and a self-adaptive optimization system for the optical transmission quality of a video stream. In order to achieve the above object, the present invention adopts the following technical scheme, and an adaptive optimization method for optical transmission quality of a video stream, which includes the following steps: S1, collecting instantaneous fiber-entering optical power of a current video stream wavelength channel and an adjacent video stream wavelength channel, calculating power fluctuation statistical variance, and calculating a cross phase modulation interference index by combining a set fiber nonlinear coefficient; s2, comparing the cross phase modulation interference index with a preset light transmission interference warning threshold value, and setting a chroma quantization step decoupling configuration according to a chroma quantization parameter of head information configuration of the current video slice to be encoded; s3, analyzing a video compression code stream generated based on the chroma quantization step decoupling configuration code, identifying a unique frame number identification of a current video stream prediction frame in a video sequence, and counting the total number of times the video stream prediction frame is referenced; S4, comparing the total quoted times of the video stream predicted frames with a preset video frame importance threshold value, extracting target frame compression data of the current video stream predicted frames according to the comparison result, and constructing an optical symbol probability shaping mapping table by combining a preset standard optical communication constellation diagram; and S5, performing symbol mapping and light modulation transmission on the target frame compressed data based on the optical symbol probability shaping mapping table, and correcting an optical transmission interference warning threshold and a video frame importance threshold according to a transmission result to obtain a