A method for transmitting video data over an IP network

Abstract

This application relates to the field of network transmission technology and proposes a method for IP network transmission of video data, including: encapsulating video data encoded in video packets into IP datagrams; receiving IP datagrams simultaneously at all receiving ends; marking the target IP datagram and its preceding adjacent IP datagrams; collecting packet loss rate; determining the smoothing estimate and trend index of the target IP datagram; and calculating the path congestion characteristic value when the receiving end receives the target IP datagram based on the packet loss rate; setting the priority range of IP datagrams; generating an initial population; calculating the fitness of the priority vector; adjusting the priority of IP datagrams based on the fitness of the priority vector and the initial population; and transmitting IP datagrams to all receiving ends. This application can improve NDI transmission performance.

Description

Technical Field

[0001] This application relates to the field of network transmission technology, specifically to a method for transmitting video data over an IP network. Background Technology

[0002] Ultra-high-definition (UHD) OB vans can provide high-quality real-time video production and transmission services for large-scale events. The core technology supporting the operation of UHD OB vans is the UHD broadcasting system, which covers the entire process from raw video data acquisition and production to video data transmission, providing users with high-quality UHD video broadcasting services. UHD video broadcasting demands high image quality and involves significant dynamic changes in the visuals. Therefore, a variable bitrate encoding mode is typically used, dynamically adjusting the bitrate according to the complexity of the image to optimize bandwidth usage while maintaining image quality.

[0003] Live broadcasts of ultra-high-definition television are affected by factors such as wireless interference, network load, and network conditions. These unstable factors can significantly impact the real-time performance and quality of video transmission. Real-time transmission of ultra-high-definition video generally uses the NDI (Network Device Interface) video transmission protocol. NDI, or Network Device Interface, has a certain degree of adaptability to variable bitrate characteristics and the time-varying nature of multicast network conditions. However, under conditions of high dynamic bitrates and significant network fluctuations, its packet loss compensation mechanism may not fully meet real-time requirements, leading to increased transmission latency and affecting overall transmission performance. Summary of the Invention

[0004] This application provides a method for transmitting video data over an IP network to address the problems of increased NDI transmission latency and poor transmission performance caused by high dynamic bitrates and large network fluctuations. The specific technical solution adopted is as follows:

[0005] One embodiment of this application provides a method for transmitting video data over an IP network, the method comprising the following steps:

[0006] The video data after video encoding is encapsulated in IP datagrams. Through IP network multicast technology, all receiving ends can receive IP datagrams simultaneously. Any IP datagram is recorded as the target IP datagram, and the preceding adjacent IP datagrams of the target IP datagram are marked. The packet loss rate of each IP datagram received by each receiving end is collected.

[0007] Based on the difference in size between the target IP datagram and its preceding neighboring IP datagrams, a smoothed estimate of the target IP datagram is determined. Based on the smoothed estimate of the target IP datagram, an estimate of the size of the next neighboring IP datagrams is obtained. Based on all the estimates, a trend index of the target IP datagram is determined. Based on the packet loss rate and trend index of the target IP datagram received by the receiver, a path congestion characteristic value is calculated when the receiver receives the target IP datagram.

[0008] Set the priority range for each receiver's IP datagrams at the router, randomly generate an initial population based on the priority range, calculate the fitness of the priority vector based on the priority vector, the priority range of IP datagrams, and the path congestion characteristics of all receivers when receiving IP datagrams, adjust the priority of IP datagrams based on the fitness of the priority vector and the initial population, and transmit IP datagrams to all receivers.

[0009] Furthermore, the specific method for marking the preceding adjacent IP datagram of the target IP datagram includes:

[0010] The first preset number of consecutive IP datagrams preceding the target IP datagram are denoted as the preceding adjacent IP datagrams of the target IP datagram.

[0011] Furthermore, the method for determining the smoothed estimate of the target IP datagram is as follows:

[0012] The mean square error of the size of the target IP datagram and the size of the target IP datagram's preceding adjacent IP datagram is denoted as the first fluctuation of the target IP datagram;

[0013] The coefficient of variation of the size of the target IP datagram and its preceding adjacent IP datagram is denoted as the second fluctuation of the target IP datagram;

[0014] The negative correlation result of the product of the first fluctuation and the second fluctuation of the target IP datagram is denoted as the smoothed estimate of the target IP datagram.

[0015] Furthermore, the specific method for obtaining the estimated size of the IP datagram adjacent to the target IP datagram based on the smoothed estimate of the target IP datagram includes:

[0016] The smoothed estimate of the target IP datagram is used as the smoothing factor value of the EWMA exponentially weighted moving average algorithm. Based on the size of the target IP datagram and the size of the IP datagram preceding the target IP datagram, the estimated size of the second preset number of IP datagrams following the target IP datagram is obtained.

[0017] Furthermore, the method for determining the trend index of the target IP datagram is as follows:

[0018] A straight line is fitted to the estimated size of all adjacent IP datagrams following the target IP datagram, and the slope of the fitted line is denoted as the trend index of the target IP datagram.

[0019] Furthermore, the method for calculating the path congestion characteristic value when the receiving end receives the target IP datagram is as follows:

[0020] The normalized value of the trend index of the target IP datagrams received by the receiving end is used as the exponent, and the packet loss rate of the target IP datagrams received by the receiving end is used as the base. The value of the exponent is calculated and recorded as the path congestion characteristic value.

[0021] Furthermore, the specific range of priority values ​​for the IP datagram includes:

[0022] All integers greater than or equal to 0 and less than or equal to 7.

[0023] Furthermore, the specific method for generating the initial population is as follows:

[0024] Randomly select values ​​within the priority range and repeat. Next, arrange all the randomly selected values ​​in the order they were selected to obtain the adjustment priority vector, where... This represents the total number of receivers.

[0025] Repeatedly obtain a third preset number of priority adjustment vectors and form an initial population.

[0026] Furthermore, the fitness of the priority vector is calculated as follows:

[0027]

[0028] in, Represents the fitness of the priority vector; This represents the maximum value within the range of priority values ​​for IP datagrams. This represents the mean of all values ​​contained within the priority vector; This represents the first preset constant; Indicates the total number of receivers; This indicates the order in which the receiving end receives IP datagrams; Indicates the receiver's first The path congestion characteristic value of the receiving end when receiving IP datagrams; This represents an exponential function with the natural constant as its base.

[0029] Furthermore, the specific method for adjusting the priority of IP datagrams based on the fitness of the priority vector and the initial population, and transmitting IP datagrams to all receivers, includes:

[0030] The objective function is to maximize the fitness of the priority vector. The particle swarm optimization algorithm is then used to process the initial population to obtain the optimal priority vector.

[0031] The values ​​within the optimal priority vector are used as the priority values ​​of the corresponding receivers to adjust the priority of IP datagrams. IP datagrams are then transmitted to all receivers with adjusted priorities.

[0032] The beneficial effects of this application are:

[0033] This application estimates the bitrate of IP datagrams following the target IP datagram based on the fluctuations in the video data bitrate of the target IP datagram and its preceding adjacent IP datagrams. It then obtains estimated sizes of the adjacent IP datagrams following the target IP datagram and evaluates the changing trends of these estimated sizes based on all estimates, determining a trend index for the target IP datagram. Finally, it combines this with the packet loss rate to evaluate the degree of path congestion at the receiving end when receiving the target IP datagram, obtaining path congestion characteristic values. To improve the overall performance of multicast group video transmission, the application adjusts the priority of IP datagram replication and forwarding by the router to each receiving end to adapt to the time-varying nature of the IP network state. Specifically, it first sets the priority of each... The priority range of IP datagrams at the receiving end at the router is determined, and an initial population is generated. To ensure the overall video data transmission performance of the multicast group, the IP datagrams received earlier in the receiving order should have a higher priority. At the same time, the path congestion of the receiving end should be less. The fitness of the priority vector is calculated. When the fitness of the priority vector is higher, the overall transmission performance of ultra-high-definition video multicast is better when determining the priority of IP datagram copies of each receiving end according to the priority vector. Furthermore, based on the fitness of the priority vector and the initial population, the priority of IP datagrams is adjusted, and IP datagrams are transmitted to all receiving ends. This solves the problem of increased NDI transmission delay and poor transmission performance caused by high dynamic bitrate and large network fluctuations, and improves the IP network transmission quality of video data. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 A schematic flowchart illustrating a video data IP network transmission method according to an embodiment of this application;

[0036] Figure 2 This is a flowchart illustrating the process of obtaining a smoothed estimate as provided in one embodiment of this application. Detailed Implementation

[0037] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0038] Please see Figure 1 The diagram illustrates a flowchart of an IP network transmission method for video data according to an embodiment of this application. The method includes the following steps:

[0039] Step S001: Encapsulate the encoded video data in IP datagrams. Using IP network multicast technology, enable all receiving ends to receive IP datagrams simultaneously. Record any IP datagram as the target IP datagram and mark the preceding adjacent IP datagrams of the target IP datagram. Collect the packet loss rate of each IP datagram received by each receiving end.

[0040] The ultra-high-definition (UHD) OB van utilizes 5G technology, significantly improving the efficiency and stability of video data transmission. Leveraging the high speed, low latency, and massive connectivity of 5G networks, the OB van can achieve real-time streaming and multi-channel concurrent transmission of UHD video, making it particularly suitable for outdoor mobile scenarios. The UHD broadcasting system is the core technology supporting the operation of the UHD OB van, encompassing the entire process from raw video data acquisition and production to video data transmission, providing users with high-quality UHD video broadcasting services. Combined with 5G technology, the UHD broadcasting system achieves a more efficient video data transmission and processing workflow. Through the high bandwidth of 5G networks, the system can support higher resolution and higher frame rate video encoding and transmission.

[0041] This application further integrates 5G communication technology, leveraging its high bandwidth, low latency, and massive connectivity capabilities to significantly improve the efficiency and stability of multicast transmission. In a 5G network environment, video data can be transmitted wirelessly to IP networks at high speed, making it particularly suitable for highly mobile and flexibly deployed ultra-high-definition broadcast van systems.

[0042] This application employs the NDI and IGMP protocols to achieve real-time broadcasting of ultra-high-definition video to multiple receiving ends. The video is encoded using H.265, and the encoded video data is encapsulated in IP datagrams. All receiving ends are configured into a multicast group, and the same multicast address is distributed via the IGMP protocol. The ultra-high-definition video broadcasting system uses the multicast address as the destination address of the IP datagram and sends it to the IP network. Routers forward and replicate the datagrams according to the IGMP protocol, transmitting copies of the replicated IP datagrams to each receiving end. Based on the aforementioned IP network multicast technology, the video data from the ultra-high-definition broadcasting vehicle can be simultaneously received by multiple receiving ends without establishing an independent connection for each receiving end.

[0043] IGMP stands for Internet Group Management Protocol; H.265 stands for High Efficiency Video Coding; the total number of receivers is... This embodiment is for The value is 100; the NDI and IGMP protocols are used to achieve real-time rebroadcasting from multiple receivers, and H.265 is used to encode video data. These are well-known technologies and will not be described in detail here.

[0044] According to video coding principles, dynamic scenes typically require higher bitrates to maintain image quality, while static or slowly changing scenes can use lower bitrates. During the broadcast of ultra-high-definition video, the degree of dynamic change in the image varies significantly across different time periods, leading to substantial differences in the bitrate of the encoded video data at different times. Furthermore, considering the continuity of image changes, the bitrate trend of video data exhibits a certain degree of continuity within a short time window, allowing for the estimation of subsequent bitrates using the bitrates of adjacent, previous video data.

[0045] The NDI protocol employs a time-based fragmentation strategy to ensure the temporal consistency of the video stream, facilitating synchronization processing at the receiving end. Therefore, each IP datagram typically corresponds to a fixed length of video data. This embodiment sets each IP datagram to 20ms of video data. In practical applications, as other implementation methods, implementers can determine the duration of the video data corresponding to each IP datagram according to their actual needs; this application does not impose any special restrictions. It is understood that the size of the IP datagram can reflect the video bitrate.

[0046] Let any one of the IP datagrams be designated as the destination IP datagram, and then denote the consecutive IP datagrams preceding the destination IP datagram as... Each IP datagram is denoted as the preceding adjacent IP datagram of the destination IP datagram.

[0047] in, This represents the first preset quantity, which is 500 in this embodiment.

[0048] Use the Wireshark network protocol analysis tool to obtain the packet loss rate of the target IP datagram and the preceding adjacent IP datagram received by the receiving end. Record the packet loss rate of the target IP datagram and the preceding adjacent IP datagram as the packet loss rate of the target IP datagram.

[0049] The packet loss rate for each IP datagram can be obtained using the same method.

[0050] At this point, the packet loss rate of all IP datagrams received by the receiving end is obtained, and the preceding adjacent IP datagrams of the target IP datagram are determined.

[0051] Step S002: Based on the difference in size between the target IP datagram and its preceding adjacent IP datagrams, determine the smoothed estimate of the target IP datagram. Based on the smoothed estimate of the target IP datagram, obtain the estimated size of the IP datagrams following the target IP datagram. Based on all the estimated values, determine the trend index of the target IP datagram. Based on the packet loss rate and trend index of the target IP datagram received by the receiver, calculate the path congestion characteristic value when the receiver receives the target IP datagram.

[0052] This application uses the EWMA (Exponentially Weighted Moving Average) algorithm to estimate the bitrate of IP datagrams following the target IP datagram based on the video data bitrate of the target IP datagram and the IP datagram preceding the target IP datagram. When the fluctuation of the video data bitrate of the target IP datagram and the IP datagram preceding the target IP datagram is greater, the reference value of the target IP datagram and the IP datagram preceding the target IP datagram to the bitrate of the IP datagram following the target IP datagram is smaller, and a smaller smoothing factor needs to be selected for estimation.

[0053] A smoothed estimate of the target IP datagram is determined based on the difference in size between the target IP datagram and its preceding neighboring IP datagrams.

[0054] The mean square error of the size of the target IP datagram and its preceding neighboring IP datagrams is denoted as the first fluctuation of the target IP datagram. The coefficient of variation of the size of the target IP datagram and its preceding neighboring IP datagrams is denoted as the second fluctuation of the target IP datagram. The negative correlation result of the product of the first fluctuation and the second fluctuation of the target IP datagram is denoted as the smoothed estimate of the target IP datagram.

[0055] It is understood that negative correlation processing is applied to the first and second fluctuations of the target IP datagram, ensuring that the product of the first and second fluctuations of the target IP datagram is negatively correlated with the smoothed estimate. It is understood that the negative correlation in this application refers to the relationship between the independent and dependent variables, where the independent variable is the product of the first and second fluctuations, and the dependent variable is the smoothed estimate. The negative correlation means that the dependent variable decreases (increases) as the independent variable increases (decreases), and can be an inverse relationship, a subtraction relationship, etc.

[0056] Preferably, as an embodiment of this application, the normalized value of the reciprocal of the product of the first fluctuation and the second fluctuation of the target IP datagram is denoted as the smoothed estimate of the target IP datagram.

[0057] The flowchart for obtaining the smoothed estimate is as follows: Figure 2 As shown.

[0058] The coefficient of variation is a well-known statistic that measures the relative dispersion of data. It represents the ratio of the standard deviation to the mean and is usually expressed as a percentage. The calculation method of the coefficient of variation will not be elaborated here. This embodiment uses the Z-Score standard normalization method to calculate the normalized value. In practical applications, implementers may use other existing methods such as the maximum-minimum normalization method or the sigmoid function to calculate the normalized value, which is not limited here.

[0059] Based on the smoothed estimate of the target IP datagram, obtain the estimated size of the IP datagrams adjacent to the target IP datagram, and determine the trend index of the target IP datagram based on all estimates.

[0060] Using the smoothed estimate of the target IP datagram as the smoothing factor, the EWMA exponentially weighted moving average algorithm is employed. Based on the size of the target IP datagram and its preceding neighboring IP datagrams, the values ​​of the next adjacent IP datagrams after the target IP datagram are obtained. An estimate of the size of each IP datagram. The adjacent datagrams following the destination IP datagram... The estimated size of each IP datagram is used as the dependent variable, and the order of the IP datagram after the target IP datagram is used as the independent variable. A straight line is fitted, and the slope of the fitted line is obtained. The slope of the fitted line is recorded as the trend index of the target IP datagram.

[0061] in, The second preset quantity is indicated, and in this embodiment, the value of the second preset quantity is 500. This embodiment uses the least squares method for line fitting. In practical applications, as other implementation methods, while achieving the purpose of line fitting, the implementer may use other existing methods such as polynomial fitting techniques to fit the line. This application does not impose any special restrictions. The trend index of the target IP datagram is used to evaluate the trend of the estimated size of each IP datagram adjacent to the target IP datagram.

[0062] During multicast video data transmission, the sending end sends only one IP datagram. Routers in the IP network are responsible for copying and forwarding this IP datagram, ensuring that each receiving end receives the same copy of the IP datagram and guaranteeing consistency of the video data received. However, the network conditions of the routing paths for different receiving ends in a multicast group can vary significantly. Therefore, it is necessary to evaluate the routing path status of each receiving end individually in order to adjust the transmission strategy for different receiving ends.

[0063] During ultra-high-definition video broadcasting, video encoding and decoding characteristics are highly sensitive to packet loss. Specifically, packet loss affects the continuity of the video and reduces the real-time performance of the broadcast. Therefore, the packet loss rate needs to be controlled within an extremely low range. The network status of the routing paths at different receiving ends can be assessed based on the packet loss rate. Specifically, the higher the packet loss rate at the receiving end, the greater the probability of path congestion, the worse the transmission performance of the IP network, and the greater the tendency for the size of IP data packets transmitted by the receiving end that has already experienced packet loss to increase, further exacerbating the deterioration of the communication status.

[0064] Based on the packet loss rate and trend index of the target IP datagrams received by the receiving end, the path congestion characteristic value when the receiving end receives the target IP datagram is calculated.

[0065] Preferably, in one embodiment of this application, the normalized value of the trend index of the target IP datagram received by the receiving end is used as the index, and the packet loss rate of the target IP datagram received by the receiving end is used as the base. The value of the exponent is calculated and recorded as the path congestion characteristic value.

[0066] The path congestion characteristic value when the receiving end receives the target IP datagram is used to evaluate the degree of path congestion when the receiving end receives the target IP datagram. The larger the path congestion characteristic value, the greater the degree of path congestion when the receiving end receives the target IP datagram, and the greater the possibility that the transmission performance of the receiving end will continue to deteriorate after receiving the target IP datagram.

[0067] The same method can be used to obtain the path congestion characteristic value of each receiving end when receiving each IP datagram.

[0068] At this point, the path congestion characteristic values ​​of all receiving ends when receiving all IP datagrams are obtained.

[0069] Step S003: Set the priority range of IP datagrams at the router for each receiver, randomly generate an initial population based on the priority range, calculate the fitness of the priority vector based on the priority vector, the priority range of IP datagrams, and the path congestion characteristic values ​​of all receivers when receiving IP datagrams, adjust the priority of IP datagrams based on the fitness of the priority vector and the initial population, and transmit IP datagrams to all receivers.

[0070] To improve the overall performance of multicast video transmission, the priority of IP datagram copies copied and forwarded by the router to each receiving end is adjusted to adapt to the time-varying nature of IP network states.

[0071] Specifically, the priority of each IP datagram at the router is set to an integer value greater than or equal to 0 and less than or equal to 7. Priority is a well-known field in IP datagrams, used to indicate the transmission priority and processing method of IP datagrams.

[0072] The key point to note is that during the process of the router forwarding and copying IP datagrams according to the IGMP protocol and transmitting the copied copies to each receiving end, only the first... Each receiving end performs the copying, forwarding, and transmission of IP datagram copies. Each receiver obtains its routing information from the router's routing table. The last receiver does not need to copy the information again; therefore, no priority adjustment is required for the last receiver. This represents the total number of receivers.

[0073] First, when establishing routing connections between the router and each receiving end, the initial priority of the IP datagrams at the router for each receiving end is set to 3. The initial values ​​of the IP datagram priorities at each receiving end at the router are arranged according to the order in which the router transmits the datagrams to each receiving end, and a priority vector is obtained.

[0074] Randomly select values ​​within the priority range and repeat. Next, arrange all randomly selected values ​​in the order of selection to obtain the priority vector. Repeat this process for a total of... There are 10 priority vectors, and all priority vectors are used to form the initial population.

[0075] It is important to understand that each adjustment priority vector in the initial population contains... The values ​​within the priority range, this The number includes the first The values ​​corresponding to each receiver and the values ​​corresponding to each router that sends data; This indicates the third preset quantity, and in this embodiment, the value of the third preset quantity is 30.

[0076] To ensure the overall video data transmission performance of a multicast group, the IP datagrams received by earlier receivers should have higher priority, and simultaneously, the path congestion of earlier receivers should be lower. Furthermore, to avoid excessively high priority IP datagrams from different receivers affecting other transmission services in the IP network, the priority of datagrams received by earlier receivers should be... If the sum of the priorities of each receiver is too large, the priority of the first receiver should be appropriately reduced. The priority value of each receiver.

[0077] Based on the priority vector, the range of IP datagram priority values, and the path congestion characteristics of all receivers when receiving IP datagrams, the fitness of the priority vector is calculated. The formula for calculating the fitness of the priority vector is:

[0078]

[0079] in, Represents the fitness of the priority vector; This represents the maximum value within the range of IP datagram priority values. In this embodiment, the priority value range is set to an integer greater than or equal to 0 and less than or equal to 7. The value of is 7; This represents the mean of all values ​​contained within the priority vector; This represents the first preset constant. The purpose of the first preset constant is to avoid the distribution being 0, which would render the fraction meaningless. In this embodiment, the value of the first preset constant is 1. Indicates the total number of receivers; This indicates the order in which the receiving end receives IP datagrams; Indicates the receiver's first The path congestion characteristic value of the receiving end when receiving IP datagrams; This represents an exponential function with the natural constant as its base.

[0080] The greater the fitness of the priority vector, the better the overall transmission performance of ultra-high-definition video multicast when determining the priority of each receiver's IP datagram copy according to the priority vector.

[0081] Using the maximum fitness of the priority vector as the objective function, the particle swarm optimization algorithm is used to process the initial population to obtain the optimal priority vector.

[0082] In this embodiment, the maximum number of iterations of the particle swarm optimization algorithm is set to 30, and both the individual learning factor and the swarm learning factor are set to 2. Iterating using the particle swarm optimization algorithm to obtain the optimal optimization result is a well-known technique and will not be described in detail here.

[0083] The values ​​within the optimal priority vector are used as the priority values ​​of the corresponding receivers to adjust the priority of IP datagrams. After the priority is adjusted, IP datagrams are transmitted to all receivers.

[0084] This completes the IP network transmission of video data.

[0085] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A method for IP network transmission of video data, characterized in that, The method comprises the following steps: The video data after video coding is encapsulated in an IP datagram, and all receiving ends simultaneously receive the IP datagram through IP network multicasting technology; any IP datagram is recorded as a target IP datagram, and a front adjacent IP datagram of the target IP datagram is marked; a packet loss rate of each IP datagram received by each receiving end is collected; A smooth estimation value of the target IP datagram is determined according to the difference between the size of the target IP datagram and the size of the front adjacent IP datagram of the target IP datagram; an estimation value of the size of an adjacent IP datagram after the target IP datagram is obtained according to the smooth estimation value of the target IP datagram; a change trend index of the target IP datagram is determined according to all estimation values; a path congestion characteristic value when the receiving end receives the target IP datagram is calculated according to the packet loss rate of the target IP datagram received by the receiving end and the change trend index; A value range of the priority of the IP datagram of each receiving end at the router is set; an initial population is randomly generated according to the value range of the priority; the fitness of the priority vector is calculated according to the priority vector, the value range of the priority of the IP datagram and the path congestion characteristic value of all receiving ends when receiving the IP datagram; the priority of the IP datagram is adjusted according to the fitness of the priority vector and the initial population, and the transmission of the IP datagram of all receiving ends is performed.

2. The method for transmitting video data over an IP network according to claim 1, wherein, The specific method for marking the front adjacent IP datagram of the target IP datagram comprises the following steps: A first preset number of IP datagrams before the target IP datagram are recorded as the front adjacent IP datagram of the target IP datagram.

3. The method for transmitting video data over an IP network according to claim 1, wherein, The specific method for determining the smooth estimation value of the target IP datagram comprises the following steps: The mean square error of the size of the target IP datagram and the size of the front adjacent IP datagram of the target IP datagram is recorded as the first fluctuation of the target IP datagram; The coefficient of variation of the size of the target IP datagram and the size of the front adjacent IP datagram of the target IP datagram is recorded as the second fluctuation of the target IP datagram; The negative correlation processing result of the product of the first fluctuation and the second fluctuation of the target IP datagram is recorded as the smooth estimation value of the target IP datagram.

4. The method for transmitting video data over an IP network according to claim 1, wherein, The specific method for obtaining the estimation value of the size of the adjacent IP datagram after the target IP datagram according to the smooth estimation value of the target IP datagram comprises the following steps: The smooth estimation value of the target IP datagram is taken as the value of the smoothing factor of the EWMA exponential weighted moving average algorithm; the estimation values of the sizes of the second preset number of IP datagrams adjacent to the target IP datagram are obtained according to the size of the target IP datagram and the size of the front adjacent IP datagram of the target IP datagram.

5. The method for transmitting video data over an IP network according to claim 1, wherein, The specific method for determining the change trend index of the target IP datagram comprises the following steps: The estimation values of the sizes of all IP datagrams adjacent to the target IP datagram are linearly fitted; the slope of the fitted straight line is recorded as the change trend index of the target IP datagram.

6. The method for transmitting video data over an IP network according to claim 1, wherein, The specific method for calculating the path congestion characteristic value when the receiving end receives the target IP datagram comprises the following steps: The normalized value of the change trend index of the target IP datagram received by the receiving end is taken as the index, the packet loss rate of the target IP datagram received by the receiving end is taken as the base, the value of the index power is calculated and recorded as the path congestion characteristic value.

7. The method for transmitting video data over an IP network according to claim 1, wherein, The value range of the priority of the IP datagram specifically includes: All integers greater than or equal to 0 and less than or equal to 7.

8. The method for transmitting video data over an IP network according to claim 1, wherein, The specific generation method of the initial population is: randomly selecting values in the range of the priority level values and repeating Next, all the randomly selected values are arranged in the order of selection to obtain an adjusted priority vector, wherein, is the total number of receiving ends. The third preset number of adjusted priority vectors are repeatedly obtained and combined to form the initial population.

9. The method for transmitting video data over an IP network according to claim 1, wherein, The fitness of the priority vector is calculated as: in, Represents the fitness of the priority vector; This represents the maximum value within the range of priority values ​​for IP datagrams. This represents the mean of all values ​​contained within the priority vector; This represents the first preset constant; Indicates the total number of receivers; This indicates the order in which the receiving end receives IP datagrams; Indicates the receiver's first The path congestion characteristic value of the receiving end when receiving IP datagrams; This represents an exponential function with the natural constant as its base.

10. The method for transmitting video data over an IP network according to claim 1, wherein, The priority of the IP datagram is adjusted according to the fitness of the priority vector and the initial population, and the transmission of the IP datagram to all receiving ends includes the specific method that: The maximum fitness of the priority vector is taken as the objective function, the initial population is processed using the particle swarm optimization algorithm to obtain the optimal priority vector; The values in the optimal priority vector are taken as the values of the priority of the receiving end corresponding to the values, the adjustment of the priority of the IP datagram is completed, and the transmission of the IP datagram to all receiving ends after the priority adjustment is performed.

Images