Congestion prediction system for holographic media transmission based on programmable network

Abstract

The present invention discloses a holographic media transmission congestion prediction system based on a programmable network, and specifically relates to the technical field of transmission congestion prediction. By dividing the task path of holographic media data transmission into sub-paths, acquiring the network performance information of the switch and the quality information of the holographic media data, constructing a path congestion model, generating a path congestion impact index, evaluating the degree of congestion impact of the network performance information of the switch and the quality information of the holographic media data on the sub-path of the holographic media data transmission, timely responding to potential congestion risks in the task path of the holographic media data transmission according to the congestion impact degrees of multiple sub-paths, accurately evaluating the congestion situation, quickly and accurately perceiving and responding to sudden congestion in the network, and when there are potential congestion risks in the task path of the holographic media data transmission, based on the shortest path optimization algorithm, quickly selecting the best transmission path, avoiding congested areas, and improving transmission efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of transmission congestion prediction, and more specifically, to a holographic media transmission congestion prediction system based on a programmable network. Background Art

[0002] In modern network communications, holographic media transmission has become one of the most demanding applications in terms of network performance due to its high bandwidth and low latency requirements. Holographic media transmission not only requires the transmission of large amounts of image and video data, but also requires the real-time and synchronization of data during transmission to ensure that users have an immersive experience. However, with the sharp increase in network traffic, network congestion has become the main bottleneck affecting the quality of holographic media transmission. Traditional congestion control methods are unable to quickly and accurately perceive and respond to sudden congestion in the network, resulting in increased transmission delays, data loss, and decreased user experience. Summary of the invention

[0003] In order to overcome the above defects of the prior art, the embodiments of the present invention provide a holographic media transmission congestion prediction system based on a programmable network to solve the problems raised in the above background technology.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A holographic media transmission congestion prediction system based on a programmable network, including a path division module, a switch network monitoring module, a holographic media data monitoring module, a congestion model building module, a path congestion perception module, and a path selection optimization module;

[0006] A path division module is used to divide the task path of holographic media data transmission into sub-paths and number the sub-paths according to the data transmission direction;

[0007] A switch network monitoring module is used to obtain abnormal network performance information of the switch and analyze the impact of abnormal network performance information of the switch on the congestion of the holographic media data transmission sub-path;

[0008] A holographic media data monitoring module is used to obtain quality information of holographic media data and analyze the impact of the quality information of holographic media data on the congestion of the holographic media data transmission subpath;

[0009] A congestion model building module is used to build a path congestion model based on the network performance anomaly information of the switch and the quality information of the holographic media data, generate a path congestion impact index, and evaluate the impact of the network performance anomaly information of the switch and the quality information of the holographic media data on the congestion of the holographic media data transmission sub-path;

[0010] A path congestion perception module is used to determine the potential congestion risk of the holographic media data transmission task path according to the congestion impact degree of multiple sub-paths;

[0011] The path selection optimization module is used to select the optimal path for holographic media data transmission based on the shortest path optimization algorithm when there is a potential congestion risk in the holographic media data transmission task path.

[0012] In a preferred implementation, the network performance anomaly information includes a port traffic anomaly coefficient, a bandwidth anomaly utilization coefficient, and a retransmission confirmation times increase coefficient.

[0013] In a preferred embodiment, the logic for obtaining the port flow anomaly coefficient is as follows:

[0014] Get the actual traffic of the switch port , represents the actual traffic of the switch port obtained at time t, t={1,2,...,T}, according to the actual traffic of the switch port , calculate the average flow rate , the expression is as follows ;

[0015] Get the flow threshold value based on the average flow rate , the expression is as follows , where The floating coefficient representing the average flow rate. is a constant factor;

[0016] Compare the actual traffic of the switch port obtained at time t with the traffic threshold, and when the actual traffic of the switch port obtained at time t is greater than the traffic threshold, mark the actual traffic of the switch port obtained at time t as abnormal port traffic;

[0017] Calculate port traffic anomaly coefficient , the expression is as follows , where d represents the occurrence order of abnormal port traffic, d={1,2,...,D}.

[0018] In a preferred embodiment, the logic for obtaining the bandwidth abnormal utilization coefficient is as follows:

[0019] Get the bandwidth utilization of the switch, build a bandwidth utilization data set, and mark the bandwidth utilization data set as ,in represents the f-th acquired bandwidth utilization data point, f={1,2,...,F}, where F is a positive integer;

[0020] Use the bandwidth utilization dataset to train the isolation forest model and calculate the average path length , the expression is as follows , where Represents the path length of the nth isolated tree in the isolation forest model. The path length refers to the distance from the root node of the isolated tree to the bandwidth utilization data point. The path length, n={1,2,...,N}, N is a positive integer;

[0021] Calculate the adjustment factor , the expression is as follows ;

[0022] Calculate the anomaly score of the bandwidth utilization data point. The expression is as follows ;

[0023] Compare the abnormal score of the bandwidth utilization data point with the preset score threshold, mark the bandwidth utilization data point whose abnormal score is greater than the score threshold as the bandwidth utilization abnormal data point, and calculate the proportion of bandwidth utilization abnormal data points , the expression is as follows , where Indicates the number of abnormal bandwidth utilization data points;

[0024] Calculate the average anomaly score of bandwidth utilization anomaly data points , the expression is as follows , ;

[0025] Calculate the bandwidth abnormal utilization coefficient , the expression is as follows .

[0026] In a preferred implementation, the logic for obtaining the retransmission confirmation times increase coefficient is as follows:

[0027] In fixed time period Get the number of holographic media data packet retransmission confirmations that occurred in the switch , calculate the average number of retransmission confirmation times of holographic media data packets , the expression is as follows , where Indicates the jcth fixed time period The number of retransmission confirmations of the holographic media data packet obtained within , is a positive integer;

[0028] The average number of holographic media data packet retransmission confirmation times is used as the initial exponential moving average , the expression obtained is as follows , the exponential moving average algorithm is used to calculate the exponential moving average of different fixed time periods. The calculation expression is as follows ,in Indicates exponential moving average of a fixed time period, Indicates the preset initial smoothing factor;

[0029] Calculate the dynamic smoothing factor , the expression is as follows ,in ;

[0030] The dynamic smoothing factor obtained Replace the preset initial smoothing factor;

[0031] Calculate the peak ratio coefficient of the number of retransmission confirmations of holographic media data packets , , where represents a maximum value acquisition function, which is used to obtain the maximum value of the number of retransmission confirmations of the holographic media data packet within multiple fixed time periods;

[0032] Calculate the increase coefficient of retransmission confirmation times , the expression is as follows .

[0033] In a preferred implementation, the quality information of the holographic media data includes a frame rate fluctuation coefficient and a resolution surge rate.

[0034] In a preferred embodiment, the logic for obtaining the frame rate fluctuation coefficient is as follows:

[0035] Obtain the frame rate data of the holographic media data during the holographic media data transmission process, and construct a frame rate data time series, marking the frame rate data time series as , represents the frame rate value of the holographic media data obtained at time point m, m={0,1,2,...,M-1}, M is a positive integer;

[0036] Perform discrete Fourier transform on the frame rate data time series to convert the time domain data into frequency domain data. The transformation formula is as follows , where Represents the frequency component corresponding to frequency k, where k represents the frequency index. , is an imaginary unit;

[0037] The frequency components Converted into energy value , the conversion formula is as follows ;

[0038] Compare the energy values ​​at different frequencies with the preset energy threshold, and mark the energy values ​​greater than the energy threshold as high-frequency energy ;

[0039] Calculating the frame rate fluctuation coefficient , the expression is as follows , where Indicates High frequency energy, ={1,2,...,GP}, GP is a positive integer;

[0040] The logic for obtaining the resolution surge rate is as follows:

[0041] Obtain the resolution of holographic media data during holographic media transmission and calculate the resolution surge rate , the expression is as follows , where Indicates the resolution of the previous holographic media data when the resolution changes. Indicates the resolution of the next holographic media data when the resolution changes. Indicates the preset resolution change threshold.

[0042] In a preferred embodiment, the port traffic anomaly coefficient, bandwidth anomaly utilization coefficient, retransmission confirmation number increase coefficient, frame rate fluctuation coefficient, and resolution surge rate are normalized, and a path congestion model is constructed based on the normalized port traffic anomaly coefficient, bandwidth anomaly utilization coefficient, retransmission confirmation number increase coefficient, frame rate fluctuation coefficient, and resolution surge rate to generate a path congestion impact index. , the formula based on which is as follows , where Indicates the port traffic anomaly coefficient, Indicates the bandwidth abnormal utilization coefficient, Indicates the coefficient of increase in the number of retransmission confirmations. Represents the frame rate fluctuation coefficient, Indicates the resolution surge rate, , , They represent the preset proportional coefficients of the port traffic anomaly coefficient, bandwidth anomaly utilization coefficient, retransmission confirmation times increase coefficient, frame rate fluctuation coefficient, and resolution sudden increase rate, respectively. , , Both are greater than 0.

[0043] In a preferred embodiment, the path congestion impact index of different sub-paths is obtained according to the path congestion model, a path congestion data analysis set is established, and the path congestion data analysis set is marked as ,in represents the path congestion impact index of the qth sub-path, , is a positive integer;

[0044] Calculate the standard deviation of the path congestion impact index in the path congestion data analysis set , the expression is as follows , where It represents the average value of the path congestion impact index in the path congestion data analysis set. The expression is as follows ;

[0045] The standard deviation of the path congestion impact index in the path congestion data analysis set is compared with the preset standard deviation threshold to determine the potential congestion risk of the holographic media data transmission task path, as follows:

[0046] If the standard deviation of the path congestion impact index in the path congestion data analysis set is greater than the standard deviation threshold, a potential congestion risk signal is generated;

[0047] If the standard deviation of the path congestion impact index in the path congestion data analysis set is less than or equal to the standard deviation threshold, a task path stability signal is generated.

[0048] In a preferred embodiment, the path selection optimization module is used to select the optimal path for holographic media data transmission based on the shortest path optimization algorithm when there is a potential congestion risk on the holographic media data transmission task path, as follows:

[0049] Step A1: Convert all switches and sub-paths in the network into a weighted graph. Each node in the weighted graph represents a switch in the network. Each edge in the weighted graph represents a sub-path between two switches. Assign a preference coefficient to each edge. , Indicates the subpath preference coefficient from node u to node v. Its preference coefficient The expression for obtaining is as follows , where It refers to the path congestion impact index from node u to node v generated by the path congestion model;

[0050] Step A2: Mark the starting node of holographic media data transmission as the source node , marking the target node of holographic media data transmission as the target node , create a distance array And initialize, the source node The initial distance is , indicating that the distance from the source node to itself is 0, and the initial distance of other nodes u in the weighted graph is , indicating that there is currently no path from the source node to these nodes; establish a predecessor node array , and initialized to , indicating that node u has no predecessor node temporarily; establish an unvisited node set QJ, which initially contains all nodes;

[0051] Step A3: Select a node from the unvisited node set QJ that is close to the source node. The nearest node u, that is, is the smallest node, and removes node u from the unvisited node set QJ;

[0052] Step A4: For each node v adjacent to u, calculate the subpath distance from node u to node v. , the subpath distance Distance from current record For comparison, if , then update , indicating that the path from node u to node v is the known shortest subpath distance; update the predecessor node , indicating that from the source node Going through node u to reach node v is the shortest subpath currently known;

[0053] Step A5, repeat step A4 until there is no node in the unvisited node set QJ or the target node is found stop;

[0054] Step A6: According to the predecessor node array To reconstruct the source node To the target node The optimal path from the target node Start by tracing the predecessor nodes forward until you reach the source node , generate the optimal path list , , and according to the optimal path list Perform holographic media data transmission.

[0055] Technical effects and advantages of the present invention:

[0056] 1. The present invention divides the task path of holographic media data transmission into sub-paths, obtains the network performance information of the switch and the quality information of the holographic media data to construct a path congestion model, generates a path congestion impact index, evaluates the impact of the network performance information of the switch and the quality information of the holographic media data on the congestion of the holographic media data transmission sub-path, and responds to the potential congestion risks in the holographic media data transmission task path in a timely manner according to the congestion impact of multiple sub-paths, accurately evaluates the congestion situation, reflects the subtle changes in the network status of the holographic media data transmission task path, helps identify potential congestion risks, quickly and accurately perceives and responds to sudden congestion in the network, and when there is a potential congestion risk in the holographic media data transmission task path, based on the shortest path optimization algorithm, when potential congestion is found, it can quickly select the best transmission path to avoid congested areas and improve transmission efficiency; BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0058] Figure 1 Schematic diagram of the structure of the system of the embodiment of the present invention. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] Embodiment: The present invention provides as Figure 1 The holographic media transmission congestion prediction system based on programmable network shown includes a path division module, a switch network monitoring module, a holographic media data monitoring module, a congestion model building module, a path congestion perception module, and a path selection optimization module;

[0061] A path division module is used to divide the task path of holographic media data transmission into sub-paths and number the sub-paths according to the data transmission direction;

[0062] A switch network monitoring module is used to obtain abnormal network performance information of the switch and analyze the impact of abnormal network performance information of the switch on the congestion of the holographic media data transmission sub-path;

[0063] A holographic media data monitoring module is used to obtain quality information of holographic media data and analyze the impact of the quality information of holographic media data on the congestion of the holographic media data transmission subpath;

[0064] A congestion model building module is used to build a path congestion model based on the network performance anomaly information of the switch and the quality information of the holographic media data, generate a path congestion impact index, and evaluate the impact of the network performance anomaly information of the switch and the quality information of the holographic media data on the congestion of the holographic media data transmission sub-path;

[0065] A path congestion perception module is used to determine the potential congestion risk of the holographic media data transmission task path according to the congestion impact degree of multiple sub-paths;

[0066] A path selection optimization module is used to select the optimal path for holographic media data transmission based on the shortest path optimization algorithm when there is a potential congestion risk on the holographic media data transmission task path;

[0067] A path division module is used to divide the task path of holographic media data transmission into sub-paths and number the sub-paths according to the data transmission direction;

[0068] Holographic media data usually needs to be forwarded between multiple switches to achieve the holographic media data transmission task from the source device to the target device. The holographic media data transmission task path refers to the complete network transmission path through multiple switches in the process of holographic media data transmission from the source device to the target device. It is a collection of communication links and intermediate nodes that holographic media data passes through in the network to complete the transmission task. In network data transmission, a switch is a network device used to forward data packets in a local area network. It accurately sends data to the destination device according to the MAC address (media access control address) of the target device, and plays the role of connection and forwarding. Its main task is to decide which port to forward the data packet to based on the target MAC address in the data packet.

[0069] A subpath refers to a holographic media data transmission path between different switches in a task path. The path division module divides the entire task path into multiple subpaths according to each switch. For example, every two consecutive switches in the task path can be regarded as a subpath, and each divided subpath is numbered according to the data transmission direction. The subpaths are numbered in sequence from the source device to the target device. For example, the subpath between the first switch and the second switch is numbered 1, the subpath between the second switch and the third switch is numbered 2, and so on.

[0070] A switch network monitoring module is used to obtain abnormal network performance information of the switch and analyze the impact of abnormal network performance information of the switch on the congestion of the holographic media data transmission sub-path;

[0071] It should be noted that the network performance abnormality information of the switch obtained refers to the network performance abnormality information of the switch participating in the sub-path holographic media data transmission;

[0072] The network performance abnormality information includes the port traffic abnormality coefficient, the bandwidth abnormality utilization coefficient, and the retransmission confirmation number increase coefficient;

[0073] The port traffic anomaly coefficient is used to measure the difference between the actual traffic processed by the switch port and the expected traffic. Holographic media transmission usually requires large traffic and a continuous and stable transmission rate. By calculating the port traffic anomaly coefficient, it can be used to monitor whether the traffic load of the switch port exceeds the normal level; a higher port traffic anomaly coefficient indicates that the difference between the actual traffic processed by the switch port and the expected traffic is greater, and the greater the impact on the congestion of the holographic media data transmission sub-path, and the smaller the port traffic anomaly coefficient, the more the traffic load of the switch port tends to the normal level;

[0074] The logic for obtaining the port traffic anomaly coefficient is as follows:

[0075] Get the actual traffic of the switch port , represents the actual traffic of the switch port obtained at time t, t={1,2,...,T}, according to the actual traffic of the switch port , calculate the average flow rate , the expression is as follows ;

[0076] It should be noted that the actual traffic of the switch port can be monitored and obtained through network management tools such as SolarWinds, Zabbix, and Nagios;

[0077] Get the flow threshold value based on the average flow rate , the expression is as follows , where The floating coefficient representing the average flow rate. is a constant factor;

[0078] It should be noted that , It can be preset by technicians in this field according to network management strategies and requirements, for example, If it is set to 80%, it means that 80% of the flow threshold will be determined based on the average value of the current flow, which makes it easier to perceive slight fluctuations in flow anomalies. The constant factor b can be set to a fixed value to provide a minimum benchmark value and fine-tune the benchmark value of the flow threshold. For example, setting b to an empirical value can ensure that even in a light-load state, an abnormal alarm will not be triggered due to lower-than-expected flow.

[0079] Compare the actual traffic of the switch port obtained at time t with the traffic threshold, and when the actual traffic of the switch port obtained at time t is greater than the traffic threshold, mark the actual traffic of the switch port obtained at time t as abnormal port traffic;

[0080] Calculate port traffic anomaly coefficient , the expression is as follows , where d represents the occurrence order of abnormal port traffic, d={1,2,...,D};

[0081] The bandwidth abnormal utilization coefficient is used to measure the difference between the actual bandwidth used by the switch and the theoretically available bandwidth. By calculating the bandwidth abnormal utilization coefficient, it is possible to evaluate whether the bandwidth usage of the switch reaches or exceeds the expected level, thereby monitoring whether there is abnormal occupancy or overuse of the network bandwidth. A higher bandwidth abnormal utilization coefficient indicates that the difference between the actual bandwidth used by the switch and the theoretically available bandwidth is greater, and the greater the impact on the congestion of the holographic media data transmission sub-path. A smaller bandwidth abnormal utilization coefficient indicates that the bandwidth utilization of the switch is within the expected range.

[0082] The logic for obtaining the bandwidth abnormal utilization coefficient is as follows:

[0083] Get the bandwidth utilization of the switch, build a bandwidth utilization data set, and mark the bandwidth utilization data set as ,in represents the f-th acquired bandwidth utilization data point, f={1,2,...,F}, where F is a positive integer;

[0084] It should be noted that the bandwidth utilization of the switch can be monitored and obtained through network management tools such as SolarWinds, Zabbix, and Nagios;

[0085] Use the bandwidth utilization dataset to train the isolation forest model and calculate the average path length , the expression is as follows , where Represents the path length of the nth isolated tree in the isolation forest model. The path length refers to the distance from the root node of the isolated tree to the bandwidth utilization data point. The path length, n={1,2,...,N}, N is a positive integer;

[0086] It should be noted that the Isolation Forest Model is an unsupervised anomaly detection algorithm based on a tree structure. It "isolates" data points by constructing a random forest to identify anomalies. Anomalies usually require fewer segmentation operations to be isolated because they are relatively isolated in the data distribution and appear different from other data points. Due to its fast and accurate characteristics, the Isolation Forest Model can effectively identify anomalies in bandwidth utilization, and is particularly suitable for anomaly monitoring in large-scale network environments. The Isolation Forest Model is an existing mature technology and will not be described in detail here.

[0087] Calculate the adjustment factor , the expression is as follows ;

[0088] The adjustment coefficient is used in the isolation forest model to normalize the path length and the size of the bandwidth utilization dataset;

[0089] Calculate the anomaly score of the bandwidth utilization data point. The expression is as follows ;

[0090] Compare the abnormal score of the bandwidth utilization data point with the preset score threshold, mark the bandwidth utilization data point whose abnormal score is greater than the score threshold as the bandwidth utilization abnormal data point, and calculate the proportion of bandwidth utilization abnormal data points , the expression is as follows , where Indicates the number of abnormal bandwidth utilization data points;

[0091] Calculate the average anomaly score of bandwidth utilization anomaly data points , the expression is as follows , ;

[0092] Calculate the bandwidth abnormal utilization coefficient , the expression is as follows ;

[0093] The retransmission confirmation number increase coefficient is used to measure the abnormal increase in the number of retransmission confirmations of holographic media data packets on the switch. By obtaining the retransmission confirmation number increase coefficient, the potential congestion, packet loss and performance degradation in the network can be effectively evaluated. Holographic media transmission has extremely high requirements for network stability and low latency. Therefore, the increase in the number of retransmission confirmations often significantly affects the transmission quality and user experience. Some switches in the network may be congested due to insufficient processing capacity, resulting in packet delays or losses, forcing the sender to retransmit unconfirmed packets. As congestion increases, the number of retransmission confirmations will also increase, affecting the overall transmission efficiency. A higher retransmission confirmation number increase coefficient indicates that the greater the abnormal increase in the number of retransmission confirmations of holographic media data packets on the switch, the deeper the impact on the congestion of the holographic media data transmission sub-path, and the smaller the retransmission confirmation number increase coefficient, the number of retransmission confirmations is within the expected range;

[0094] The logic for obtaining the retransmission confirmation number increase coefficient is as follows:

[0095] In fixed time period Get the number of holographic media data packet retransmission confirmations that occurred in the switch , calculate the average number of retransmission confirmation times of holographic media data packets , the expression is as follows , where Indicates the jcth fixed time period The number of retransmission confirmations of the holographic media data packet obtained within , is a positive integer;

[0096] It should be noted that the number of retransmission confirmations of holographic media data packets on the switch can be monitored and obtained through network management tools such as SolarWinds, Zabbix, and Nagios;

[0097] The average number of holographic media data packet retransmission confirmation times is used as the initial exponential moving average , the expression obtained is as follows , the exponential moving average algorithm is used to calculate the exponential moving average of different fixed time periods. The calculation expression is as follows ,in Indicates exponential moving average of a fixed time period, Indicates the preset initial smoothing factor;

[0098] It should be noted that the smoothing factor determines the weight ratio of the new observation value to the past EMA value and defines the sensitivity of EMA to the latest data point. A larger value will make the EMA more sensitive to the latest data changes, while a smaller value will It will increase the dependence on historical data and make the smoothing effect stronger. In an optional example, Fixed time period To obtain, that is ;

[0099] Since holographic media transmission has extremely high requirements for network stability and low latency, and the fluctuations in the network are relatively frequent, the preset initial smoothing factor may not be able to capture short-term fluctuations in time. Therefore, the smoothing factor is dynamically corrected by the difference in the number of retransmission confirmations of holographic media data packets within adjacent fixed time periods;

[0100] Calculate the dynamic smoothing factor , the expression is as follows ,in ;

[0101] The dynamic smoothing factor obtained Replace the preset initial smoothing factor;

[0102] Calculate the peak ratio coefficient of the number of retransmission confirmations of holographic media data packets , , where represents a maximum value acquisition function, which is used to obtain the maximum value of the number of retransmission confirmations of the holographic media data packet within multiple fixed time periods;

[0103] Calculate the increase coefficient of retransmission confirmation times , the expression is as follows ;

[0104] A holographic media data monitoring module is used to obtain quality information of holographic media data and analyze the impact of the quality information of holographic media data on the congestion of the holographic media data transmission subpath;

[0105] The quality information of the holographic media data includes a frame rate fluctuation coefficient and a resolution surge rate;

[0106] The frame rate fluctuation coefficient is used to measure the degree of frame rate change during the transmission of holographic media data. Frame rate fluctuation directly affects the flow and transmission stability of holographic media data. The higher the frame rate fluctuation coefficient, the more drastic the frame rate fluctuation. When the frame rate fluctuates too much, it may lead to unbalanced use of network resources, causing excessive bandwidth load at certain times and increasing congestion risks. When the frame rate rises sharply, the amount of holographic media data increases dramatically, which puts additional pressure on the processing capacity of network equipment, and the greater the impact on the congestion of the holographic media data transmission sub-path. On the contrary, the lower the frame rate fluctuation coefficient, the smaller the fluctuation of the frame rate, the more stable it tends to be, and the less impact on the congestion of the holographic media data transmission sub-path.

[0107] The logic for obtaining the frame rate fluctuation coefficient is as follows:

[0108] Obtain the frame rate data of the holographic media data during the holographic media data transmission process, and construct a frame rate data time series, marking the frame rate data time series as , represents the frame rate value of the holographic media data obtained at time point m, m={0,1,2,...,M-1}, M is a positive integer;

[0109] It should be noted that the frame rate data of the holographic media data can be obtained by querying the encoder API;

[0110] Perform discrete Fourier transform on the frame rate data time series to convert the time domain data into frequency domain data. The transformation formula is as follows , where Represents the frequency component corresponding to frequency k, where k represents the frequency index. , is an imaginary unit;

[0111] The frequency components Converted into energy value , the conversion formula is as follows ;

[0112] Compare the energy values ​​at different frequencies with the preset energy threshold, and mark the energy values ​​greater than the energy threshold as high-frequency energy ;

[0113] Calculating the frame rate fluctuation coefficient , the expression is as follows , where Indicates High frequency energy, ={1,2,...,GP}, GP is a positive integer;

[0114] The resolution surge rate is used to measure the extent to which the resolution increases rapidly in a short period of time during holographic media transmission. A sudden resolution surge will directly lead to an increase in the size of the holographic media data packet, thereby causing a sudden increase in the transmission bandwidth demand. The greater the impact on the congestion of the holographic media data transmission subpath, the greater the impact.

[0115] The logic for obtaining the resolution surge rate is as follows:

[0116] Obtain the resolution of holographic media data during holographic media transmission and calculate the resolution surge rate , the expression is as follows , where Indicates the resolution of the previous holographic media data when the resolution changes. Indicates the resolution of the next holographic media data when the resolution changes. Indicates the preset resolution change threshold;

[0117] It should be noted that the resolution of holographic media data can be obtained by querying the encoder API;

[0118] A congestion model building module is used to build a path congestion model based on the network performance anomaly information of the switch and the quality information of the holographic media data, generate a path congestion impact index, and evaluate the impact of the network performance anomaly information of the switch and the quality information of the holographic media data on the congestion of the holographic media data transmission sub-path;

[0119] The port traffic anomaly coefficient, bandwidth anomaly utilization coefficient, retransmission confirmation number increase coefficient, frame rate fluctuation coefficient, and resolution surge rate are normalized, and a path congestion model is constructed based on the normalized port traffic anomaly coefficient, bandwidth anomaly utilization coefficient, retransmission confirmation number increase coefficient, frame rate fluctuation coefficient, and resolution surge rate to generate a path congestion impact index. , the formula based on which is as follows , where Indicates the port traffic anomaly coefficient, Indicates the bandwidth abnormal utilization coefficient, Indicates the coefficient of increase in the number of retransmission confirmations. Represents the frame rate fluctuation coefficient, Indicates the resolution surge rate, , , They represent the preset proportional coefficients of the port traffic anomaly coefficient, bandwidth anomaly utilization coefficient, retransmission confirmation times increase coefficient, frame rate fluctuation coefficient, and resolution sudden increase rate, respectively. , , All are greater than 0;

[0120] It should be noted that , , Set it according to the actual situation. For example, adopt the expert empowerment method, that is, invite experts in relevant fields to determine the preset proportion coefficients of various indicators through professional opinion surveys and comprehensive evaluations;

[0121] It can be seen from the above calculation expression that the larger the port traffic anomaly coefficient, the larger the bandwidth anomaly utilization coefficient, the larger the retransmission confirmation times increase coefficient, the larger the frame rate fluctuation coefficient, and the larger the resolution surge rate, the larger the path congestion impact index, indicating that the network performance anomaly information of the switch and the quality information of the holographic media data have a deeper impact on the congestion of the holographic media data transmission sub-path; conversely, the smaller the port traffic anomaly coefficient, the smaller the bandwidth anomaly utilization coefficient, the smaller the retransmission confirmation times increase coefficient, the smaller the frame rate fluctuation coefficient, and the smaller the resolution surge rate, the smaller the path congestion impact index, indicating that the network performance anomaly information of the switch and the quality information of the holographic media data have a shallower impact on the congestion of the holographic media data transmission sub-path;

[0122] A path congestion perception module is used to determine the potential congestion risk of the holographic media data transmission task path according to the congestion impact degree of multiple sub-paths;

[0123] According to the path congestion model, the path congestion impact index of different sub-paths is obtained, and a path congestion data analysis set is established. The path congestion data analysis set is marked as ,in represents the path congestion impact index of the qth sub-path, , is a positive integer;

[0124] Calculate the standard deviation of the path congestion impact index in the path congestion data analysis set , the expression is as follows , where It represents the average value of the path congestion impact index in the path congestion data analysis set. The expression is as follows ;

[0125] The standard deviation of the path congestion impact index in the path congestion data analysis set is compared with the preset standard deviation threshold to determine the potential congestion risk of the holographic media data transmission task path, as follows:

[0126] If the standard deviation of the path congestion impact index in the path congestion data analysis set is greater than the standard deviation threshold, it means that the congestion conditions of different sub-paths are quite different, and there are potential congestion risks. Some sub-paths of the holographic media data transmission task path may cause bottlenecks or delays in data transmission, and it is necessary to further optimize the path selection and generate potential congestion risk signals;

[0127] If the standard deviation of the path congestion impact index in the path congestion data analysis set is less than or equal to the standard deviation threshold, it means that the congestion of each sub-path is relatively balanced, the transmission task path is relatively stable, and there is no obvious congestion risk for the time being, and a task path stability signal is generated;

[0128] A path selection optimization module is used to select the optimal path for holographic media data transmission based on the shortest path optimization algorithm when there is a potential congestion risk on the holographic media data transmission task path;

[0129] Step A1: Convert all switches and sub-paths in the network into a weighted graph. Each node in the weighted graph represents a switch in the network. Each edge in the weighted graph represents a sub-path between two switches. Assign a preference coefficient to each edge. , Indicates the subpath preference coefficient from node u to node v. Its preference coefficient The expression for obtaining is as follows , where It refers to the path congestion impact index from node u to node v generated by the path congestion model;

[0130] Step A2: Mark the starting node of holographic media data transmission as the source node , marking the target node of holographic media data transmission as the target node , create a distance array And initialize, the source node The initial distance is , indicating that the distance from the source node to itself is 0, and the initial distance of other nodes u in the weighted graph is , indicating that there is currently no path from the source node to these nodes; establish a predecessor node array , and initialized to , indicating that node u has no predecessor node temporarily; establish an unvisited node set QJ, which initially contains all nodes;

[0131] Step A3: Select a node from the unvisited node set QJ that is close to the source node. The nearest node u, that is, is the smallest node, and removes node u from the unvisited node set QJ;

[0132] Step A4: For each node v adjacent to u, calculate the subpath distance from node u to node v. , the subpath distance Distance from current record For comparison, if , then update , indicating that the path from node u to node v is the known shortest subpath distance; update the predecessor node , indicating that from the source node Going through node u to reach node v is the shortest subpath currently known;

[0133] Step A5, repeat step A4 until there is no node in the unvisited node set QJ or the target node is found stop;

[0134] Step A6: According to the predecessor node array To reconstruct the source node SY to the target node The optimal path from the target node Start by tracing the predecessor nodes forward until you reach the source node , generate the optimal path list , , and according to the optimal path list Perform holographic media data transmission;

[0135] The present invention divides the task path of holographic media data transmission into sub-paths, obtains the network performance information of the switch and the quality information of the holographic media data to construct a path congestion model, generates a path congestion impact index, evaluates the impact of the network performance information of the switch and the quality information of the holographic media data on the congestion of the holographic media data transmission sub-path, and responds to the potential congestion risks existing in the holographic media data transmission task path in a timely manner according to the congestion impact degrees of multiple sub-paths, accurately evaluates the congestion situation, reflects the subtle changes in the network status of the holographic media data transmission task path, helps identify potential congestion risks, quickly and accurately perceives and responds to sudden congestion in the network, and when there are potential congestion risks in the holographic media data transmission task path, based on the shortest path optimization algorithm, when potential congestion is found, the best transmission path can be quickly selected to avoid congested areas and improve transmission efficiency;

[0136] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0137] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0138] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0139] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0140] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A holographic media transmission congestion prediction system based on a programmable network, characterized by: It includes path division module, switch network monitoring module, holographic media data monitoring module, congestion model building module, path congestion perception module, and path selection optimization module; A path division module is used to divide the task path of holographic media data transmission into sub-paths and number the sub-paths according to the data transmission direction; A switch network monitoring module is used to obtain abnormal network performance information of the switch and analyze the impact of abnormal network performance information of the switch on the congestion of the holographic media data transmission sub-path; A holographic media data monitoring module is used to obtain quality information of holographic media data and analyze the impact of the quality information of holographic media data on the congestion of the holographic media data transmission subpath; A congestion model building module is used to build a path congestion model based on the network performance anomaly information of the switch and the quality information of the holographic media data, generate a path congestion impact index, and evaluate the impact of the network performance anomaly information of the switch and the quality information of the holographic media data on the congestion of the holographic media data transmission sub-path; A path congestion perception module is used to determine the potential congestion risk of the holographic media data transmission task path according to the congestion impact degree of multiple sub-paths; The path selection optimization module is used to select the optimal path for holographic media data transmission based on the shortest path optimization algorithm when there is a potential congestion risk in the holographic media data transmission task path.

2. The programmable network-based holographic media transmission congestion prediction system according to claim 1, characterized in that: The network performance abnormality information includes a port flow abnormality coefficient, a bandwidth abnormality utilization coefficient, and a retransmission confirmation times increase coefficient.

3. The programmable network-based holographic media transmission congestion prediction system according to claim 2, characterized in that: The logic for obtaining the port traffic anomaly coefficient is as follows: Get the actual traffic of the switch port , represents the actual traffic of the switch port obtained at time t, t={1,2,...,T}, according to the actual traffic of the switch port , calculate the average flow rate , the expression is as follows ; Get the flow threshold value based on the average flow rate , the expression is as follows , where The floating coefficient representing the average flow rate. is a constant factor; Compare the actual traffic of the switch port obtained at time t with the traffic threshold, and when the actual traffic of the switch port obtained at time t is greater than the traffic threshold, mark the actual traffic of the switch port obtained at time t as abnormal port traffic; Calculate port traffic anomaly coefficient , the expression is as follows , where d represents the occurrence order of abnormal port traffic, d={1,2,...,D}.

4. The programmable network-based holographic media transmission congestion prediction system according to claim 2, characterized in that: The logic for obtaining the bandwidth abnormal utilization coefficient is as follows: Get the bandwidth utilization of the switch, build a bandwidth utilization data set, and mark the bandwidth utilization data set as ,in represents the f-th acquired bandwidth utilization data point, f={1,2,...,F}, where F is a positive integer; Use the bandwidth utilization dataset to train the isolation forest model and calculate the average path length , the expression is as follows , where Represents the path length of the nth isolated tree in the isolation forest model. The path length refers to the distance from the root node of the isolated tree to the bandwidth utilization data point. The path length, n={1,2,...,N}, N is a positive integer; Calculate the adjustment factor , the expression is as follows ; Calculate the anomaly score of the bandwidth utilization data point. The expression is as follows ; Compare the abnormal score of the bandwidth utilization data point with the preset score threshold, mark the bandwidth utilization data point whose abnormal score is greater than the score threshold as the bandwidth utilization abnormal data point, and calculate the proportion of bandwidth utilization abnormal data points , the expression is as follows , where Indicates the number of abnormal bandwidth utilization data points; Calculate the average anomaly score of bandwidth utilization anomaly data points , the expression is as follows , ; Calculate the bandwidth utilization factor , the expression is as follows .

5. The programmable network-based holographic media transmission congestion prediction system according to claim 2, characterized in that: The logic for obtaining the retransmission confirmation number increase coefficient is as follows: In fixed time period Get the number of holographic media data packet retransmission confirmations that occurred in the switch , calculate the average number of retransmission confirmation times of holographic media data packets , the expression is as follows , where Indicates the jcth fixed time period The number of retransmission confirmations of the holographic media data packet obtained within , is a positive integer; The average number of holographic media data packet retransmission confirmation times is used as the initial exponential moving average , the expression obtained is as follows , the exponential moving average algorithm is used to calculate the exponential moving average of different fixed time periods. The calculation expression is as follows ,in Indicates exponential moving average of a fixed time period, Indicates the preset initial smoothing factor; Calculate the dynamic smoothing factor , the expression is as follows ,in ; The dynamic smoothing factor obtained Replace the preset initial smoothing factor; Calculate the peak ratio coefficient of the number of retransmission confirmations of holographic media data packets , , where represents a maximum value acquisition function, which is used to obtain the maximum value of the number of retransmission confirmations of the holographic media data packet within multiple fixed time periods; Calculate the increase coefficient of retransmission confirmation times , the expression is as follows .

6. The programmable network-based holographic media transmission congestion prediction system according to claim 1, characterized in that: The quality information of the holographic media data includes a frame rate fluctuation coefficient and a resolution surge rate.

7. The programmable network-based holographic media transmission congestion prediction system according to claim 6, characterized in that: The logic for obtaining the frame rate fluctuation coefficient is as follows: Obtain the frame rate data of the holographic media data during the holographic media data transmission process, and construct a frame rate data time series, marking the frame rate data time series as , represents the frame rate value of the holographic media data obtained at time point m, m={0,1,2,...,M-1}, M is a positive integer; Perform discrete Fourier transform on the frame rate data time series to convert the time domain data into frequency domain data. The transformation formula is as follows , where Represents the frequency component corresponding to frequency k, where k represents the frequency index. , is an imaginary unit; The frequency components Converted into energy value , the conversion formula is as follows ; Compare the energy values ​​at different frequencies with the preset energy threshold, and mark the energy values ​​greater than the energy threshold as high-frequency energy ; Calculating the frame rate fluctuation coefficient , the expression is as follows , where Indicates High frequency energy, ={1,2,...,GP}, GP is a positive integer; The logic for obtaining the resolution surge rate is as follows: Obtain the resolution of holographic media data during holographic media transmission and calculate the resolution surge rate , the expression is as follows , where Indicates the resolution of the previous holographic media data when the resolution changes. Indicates the resolution of the next holographic media data when the resolution changes. Indicates the preset resolution change threshold.

8. The programmable network-based holographic media transmission congestion prediction system according to claim 7, characterized in that: The port traffic anomaly coefficient, bandwidth anomaly utilization coefficient, retransmission confirmation number increase coefficient, frame rate fluctuation coefficient, and resolution surge rate are normalized, and a path congestion model is constructed based on the normalized port traffic anomaly coefficient, bandwidth anomaly utilization coefficient, retransmission confirmation number increase coefficient, frame rate fluctuation coefficient, and resolution surge rate to generate a path congestion impact index. , the formula based on which is as follows , where Indicates the port traffic anomaly coefficient, Indicates the bandwidth abnormal utilization coefficient, Indicates the coefficient of increase in the number of retransmission confirmations. Represents the frame rate fluctuation coefficient, Indicates the resolution surge rate, , , They represent the preset proportional coefficients of the port traffic anomaly coefficient, bandwidth anomaly utilization coefficient, retransmission confirmation times increase coefficient, frame rate fluctuation coefficient, and resolution sudden increase rate, respectively. , , Both are greater than 0.

9. The programmable network-based holographic media transmission congestion prediction system according to claim 8, characterized in that: According to the path congestion model, the path congestion impact index of different sub-paths is obtained, and a path congestion data analysis set is established. The path congestion data analysis set is marked as ,in represents the path congestion impact index of the qth sub-path, , is a positive integer; Calculate the standard deviation of the path congestion impact index in the path congestion data analysis set , the expression is as follows , where It represents the average value of the path congestion impact index in the path congestion data analysis set. The expression is as follows ; The standard deviation of the path congestion impact index in the path congestion data analysis set is compared with the preset standard deviation threshold to determine the potential congestion risk of the holographic media data transmission task path, as follows: If the standard deviation of the path congestion impact index in the path congestion data analysis set is greater than the standard deviation threshold, a potential congestion risk signal is generated; If the standard deviation of the path congestion impact index in the path congestion data analysis set is less than or equal to the standard deviation threshold, a task path stability signal is generated.

10. The programmable network-based holographic media transmission congestion prediction system according to claim 9, characterized in that: The path selection optimization module is used to select the optimal path for holographic media data transmission based on the shortest path optimization algorithm when there is a potential congestion risk on the holographic media data transmission task path. The details are as follows: Step A1: Convert all switches and sub-paths in the network into a weighted graph. Each node in the weighted graph represents a switch in the network. Each edge in the weighted graph represents a sub-path between two switches. Assign a preference coefficient to each edge. , Indicates the subpath preference coefficient from node u to node v. Its preference coefficient The expression for obtaining is as follows , where It refers to the path congestion impact index from node u to node v generated by the path congestion model; Step A2: Mark the starting node of holographic media data transmission as the source node , marking the target node of holographic media data transmission as the target node , create a distance array And initialize, the source node The initial distance is , indicating that the distance from the source node to itself is 0, and the initial distance of other nodes u in the weighted graph is , indicating that there is currently no path from the source node to these nodes; establish a predecessor node array , and initialized to , indicating that node u has no predecessor node temporarily; establish an unvisited node set QJ, which initially contains all nodes; Step A3: Select a node from the unvisited node set QJ that is close to the source node. The nearest node u, that is, is the smallest node, and removes node u from the unvisited node set QJ; Step A4: For each node v adjacent to u, calculate the subpath distance from node u to node v. , the subpath distance Distance from current record For comparison, if , then update , indicating that the path from node u to node v is the known shortest subpath distance; update the predecessor node , indicating that from the source node Going through node u to reach node v is the shortest subpath currently known; Step A5, repeat step A4 until there is no node in the unvisited node set QJ or the target node is found stop; Step A6: According to the predecessor node array To reconstruct the source node To the target node The optimal path from the target node Start by tracing the predecessor nodes forward until you reach the source node , generate the optimal path list , , and according to the optimal path list Perform holographic media data transmission.

Images