CN112822268B - A method for multi-service coexistence and IP layer data packet scheduling in the Industrial Internet of Things - Google Patents

Abstract

An IP layer data packet scheduling algorithm for multi-service coexistence and deterministic network requirements in an industrial Internet of things belongs to the technical field of real-time data transmission and multi-network fusion of the industrial Internet of things. The current data packet scheduling algorithm lacks analysis on the problem of coexistence of continuous periodic data and discrete control data in the industrial Internet of things. The algorithm considers the priority and the transmission delay of the data packet at the same time, obtains partial switching conditions of the hybrid switching system through a calculation formula of the maximum transmission delay of the data packet of each priority, and analyzes the scheduling process of the data packet by using a hybrid switching system model, thereby designing the data packet scheduling algorithm of an IP layer. The algorithm can be used for scheduling the data packets from different types of networks, so that multi-network fusion is realized, the real-time requirement of the data packets in the industrial Internet of things can be met, and the packet loss rate is reduced.

Description

A method for multi-service coexistence and IP layer data packet scheduling in the Industrial Internet of Things

technical field

The invention belongs to the technical field of real-time data transmission and multi-network integration of industrial internet of things, and relates to a scheduling algorithm for data packet scheduling based on data packet priority, transmission time and hybrid switching system model at the IP layer.

Background technique

With the development of wireless networks and 5G technologies, multi-network integration and multi-service coexistence have become issues that must be overcome in the current industrial IoT research field. At the same time, with the introduction of deterministic network application requirements, how to ensure that each data packet can be sent to the network in a timely manner as required in this hybrid network application scenario has become the key. In order to meet the end-to-end low-latency requirements of industrial IoT data packets, the IEEE802.1 working group has released a series of time-sensitive networking (TSN) related standards. The real-time nature of packet transmission. However, the relevant standards for time-sensitive networks only apply to the data link layer, and in order to extend the technology developed in TSN to routers, the Internet Engineering Task Force (IETF) proposed Deterministic Networking (DetNet). Deterministic network is a technology that helps to realize IP network from providing "best effort" network service to providing "punctual, accurate and fast" network service, controlling and reducing end-to-end delay. Deterministic transmission paths are implemented at the IP layer that provide worst-case bounds on delay, packet loss, and jitter to provide deterministic latency. At present, between the IP layer and the data link layer, how to design a data packet scheduling algorithm according to requirements is still immature, especially for data packet scheduling according to the priority and transmission time of the data packets, there is no relevant report yet.

SUMMARY OF THE INVENTION

The purpose of the present invention is to overcome the deficiencies of the existing technology, and to provide a method for when data packets from multiple different types or multiple networks of the same type and data packets from multiple application services are forwarded in the IP layer in the Industrial Internet of Things. scheduling algorithm. The method considers the priority and transmission time of the data packets at the same time, and obtains some switching conditions of the hybrid switching system through the calculation formula of the maximum transmission delay of the data packets of each priority, and uses the hybrid switching system model to analyze the packet scheduling process. , thereby designing a packet scheduling algorithm at the IP layer to overcome the problem that the real-time performance of data packets with high requirements on delay such as the control class in the current industrial Internet of Things cannot be guaranteed, and at the same time, it also reduces the packet loss rate of the network.

In order to achieve the above purpose, the present invention provides a data packet scheduling algorithm based on data priority, transmission time and maximum transmission delay oriented to the Industrial Internet of Things. The method is based on the IP protocol layer in the existing TCP/IP protocol architecture. An extension to the base. During installation, this method needs to be added to the IP layer. At the same time, the three parameter information of data priority, transmission time and maximum transmission delay required by the method can be obtained directly through the network, or obtained through simple calculation. Among them, the data priority can be obtained according to the fixed field information in the data packet, which is specified in the protocol. Under the premise of clock synchronization, the transmission time can be obtained by simply calculating the synchronized clock time and the time recorded by the time stamp option in the optional field of the IP packet specified in the protocol. The maximum transmission delay can be calculated from the preset sampling period value required by industrial production and the delay specified in the protocol.

The technical scheme of the present invention is:

First, the packet scheduling process is designed as a hybrid switching model including two subsystems, the two subsystems are the priority-priority subsystem and the time-priority subsystem. The data packets in the to-be-sent column in the priority subsystem are sorted by priority, that is, the priority of the data packets from the head of the queue to the end of the queue is gradually reduced, and the data packets with the same priority are arranged in descending order according to the transmission time. Sequence sorting, the data packets in the queue to be sent in the time priority subsystem adopt the time priority sorting method, that is, the remaining transmission time (the difference between the maximum transmission delay and the transmission time) of the data packets from the head of the queue to the end of the queue gradually increases, The data packets with the same remaining transmission time are sorted according to the order of priority from high to low; secondly, the maximum value of the average transmission time of the queue data packets to be sent in the switching condition is calculated by the relevant parameters; finally, according to the state parameters of the subsystem ( Whether the average elapsed time (average elapsed time) meets the switching conditions and whether discrete events occur together determine the switching between subsystems.

The specific steps of the packet scheduling algorithm are as follows:

Step 1. Initialization: Set the current subsystem as the priority subsystem, and calculate each priority through the preset sampling period of each priority data packet in the industrial network and the maximum transmission delay of the data packet with the priority of IPP specified in the network protocol. Maximum transmission delay of priority packets;

表示当前工业网络中优先级为IPP的数据包的最大传输延迟；其表达式如下：maximum transmission delay

Represents the maximum transmission delay of the data packet with the priority of IPP in the current industrial network; its expression is as follows:

为当前工业网络中优先级为IPP的数据包的预设采样周期，该工业网络中该优先级的数据包理论传输最大传输时延等于预设采样周期值；

表示网络协议中规定的优先级为IPP的数据包的最大传输延迟；取当前工业网络中优先级为IPP的数据包的预设采样周期和网络协议中规定的优先级为IPP的数据包的最大传输延迟中的较小值作为网络资源调度的最大传输时延；Among them, IPP∈{0,1,2,3,4,5,6,7}, indicating the sending priority of IP packets;

is the preset sampling period of the data packet with the priority of IPP in the current industrial network, and the theoretical maximum transmission delay of the data packet of this priority in the industrial network is equal to the preset sampling period value;

Indicates the maximum transmission delay of the data packet with the priority of IPP specified in the network protocol; take the preset sampling period of the data packet with the priority of IPP in the current industrial network and the maximum transmission delay of the data packet with the priority of IPP specified in the network protocol The smaller value of the transmission delay is used as the maximum transmission delay of network resource scheduling;

的表达式如下：If the preset sampling period values of data packets of different applications of the same priority in the current industrial network are different, the maximum transmission delay

The expression is as follows:

Among them, i≤A _num , A _num is the number of applications in the current industrial network that generate data packets of the same priority but with different preset sampling periods;

Step 2. Obtain the data type: when a data packet arrives at the IP layer, extract the priority of the data packet and the sending time recorded by the timestamp option; obtain the round-trip delay RTT of the link between the forwarding node and the receiving end;

The data priority β is used to represent the priority weight of the current data packet in the scheduling process, and the expression is as follows:

Among them, IPP∈{0,1,2,3,4,5,6,7}, indicating the sending priority of IP packets, the larger the value, the higher the priority; IPP _um is the priority type, in the RFC standard There are 8 priorities, namely IPP _um = 8; therefore IPP<IPP _um , the data priority β∈[0,1); the data priority is judged by β, the larger the β, the higher the data priority;

RFC 1323 introduces the timestamp option in the optional field of the IP packet; when the flag field FL of the timestamp option is 3, the IP address of the sender and the time when the data packet is generated are recorded; when a data packet arrives at the IP layer, it is extracted. The priority of the packet and the timestamp corresponding to the same IP address as the source address in the timestamp option of the packet; under the premise of clock synchronization, the sender or forwarding node sends the packet through the current clock time and timestamp option. The time difference is used to obtain the transmission time of the data packet, which is represented by T _trans ; the sending end or forwarding node obtains the round-trip delay of the link between the node and the receiving end by communicating with the receiving end, which is represented by RTT;

Step 3. Calculate the state parameters: use the round-trip delay and the maximum transmission delay of the priority data packets to calculate the maximum value of the average elapsed time of the queue data packets to be sent in the switching condition; calculate the current average elapsed time of the queue data packets to be sent. The transmission time, the expression is as follows:

为该数据包对应优先级数据包的最大传输时延，RTT为该节点与接收端之间的往返时延，m为待发送队列中数据包数，m_IPP为待发送队列中该数据包对应优先级的数据包个数，R_max、R_min为系数；其中：m≤M，M为待发送队列总长度；in,

is the maximum transmission delay of the data packet corresponding to the priority data packet, RTT is the round-trip delay between the node and the receiver, m is the number of data packets in the queue to be sent, m _IPP is the corresponding data packet in the queue to be sent The number of data packets of priority, R _max and R _min are coefficients; among them: m≤M, M is the total length of the queue to be sent;

Then calculate the average transmission time of the data packets in the queue to be sent. The average transmission time T is used to represent the delay of the data packets in the queue to be sent. The expression is as follows:

The current average transmission time T of the IP packets in the queue to be sent is determined by the data priority β, the transmission time T _trans and the number of packets m in the queue to be sent. The transmission time of high-priority packets is accounted for in the calculation. The ratio is greater than that of low-priority data packets; the larger the T, the more data packets that are close to the maximum transmission delay in the current queue to be sent, and the more urgent it is to be sent;

Step 4. Subsystem switching judgment: according to the current subsystem to judge whether the switching condition is satisfied, the switching condition for the priority priority subsystem to switch to the time priority subsystem is the occurrence of the discrete event that there is a timeout packet in the queue to be sent, or The state parameter of the system is that the average transmission time of the queue to be sent is higher than the maximum value; the switching condition for the time priority subsystem to switch to the priority priority subsystem is that the average transmission time is lower than the minimum value, or the average transmission time is lower than the maximum value. In the case of , the current queue to be sent contains control packets;

When the transmission time T _trans of the data packet satisfies the following expression, it is considered to be about to time out:

为该数据包对应优先级数据包的最大传输时延，RTT为该节点与接收端之间的往返时延；in,

is the maximum transmission delay of the data packet corresponding to the priority data packet, and RTT is the round-trip delay between the node and the receiver;

If the switching conditions are met, the subsystem will be switched, and then the newly received data packets will be inserted into the appropriate position of the queue to be sent according to the ordering rules of the packets in the queue to be sent after the switching; if the switching conditions are not met, directly Insert the newly received data packet into the appropriate position of the to-be-sent queue according to the ordering rules of the data packets in the to-be-sent queue of the current subsystem.

The data packet scheduling process needs to be designed as a hybrid switching model including the priority priority subsystem and the time priority subsystem. The data packets in the to-be-sent columns in the two subsystems are sorted as follows: The to-be-sent column in the priority priority subsystem The data packets in the queue are sorted by priority, that is, the priority of the data packets from the head of the queue to the tail of the queue is gradually reduced, and the data packets with the same priority are sorted according to the order of transmission time from large to small. The data packets in the sending queue are sorted by time priority, that is, the remaining transmission time of the data packets from the head of the queue to the end of the queue increases gradually, and the data packets with the same remaining transmission time are sorted in order of priority from high to low; the The remaining transmission time is the difference between the maximum transmission delay and the transmission time.

The effects and benefits of the present invention are: the current data packet scheduling algorithm lacks analysis on the coexistence of continuous periodic data and discrete control data in the Industrial Internet of Things. The algorithm considers the priority and transmission delay of the data packets at the same time, and obtains some switching conditions of the hybrid switching system through the calculation formula of the maximum transmission delay of the data packets of each priority. Analyzed and designed the packet scheduling algorithm of IP layer. This method is a method for determining the maximum transmission delay of each priority data packet considering industrial production requirements and protocol requirements; and uniformly schedules data packets from different types of networks in the IP layer to achieve multi-network fusion; this method uses The hybrid switching system model analyzes the data packet scheduling process, overcomes the problem that the real-time performance of data packets with high delay requirements such as the control class in the current industrial Internet of Things cannot be guaranteed, and also reduces the network packet loss rate.

Description of drawings

Figure 1 is a flowchart of the scheduling algorithm.

Figure 2 is an applicable network topology diagram, and the scheduling algorithm can be run on any node that needs to send data packets.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings.

Referring to FIG. 2, an applicable network topology diagram of a data packet scheduling algorithm based on data priority, transmission time and maximum transmission delay of the present invention is introduced. The algorithm can be run on network terminals and forwarding nodes that need to send data packets . The invention uniformly schedules data packets from different types of networks in the IP layer, and realizes multi-network fusion. At the same time, the present invention uses the hybrid switching system model to analyze the data packet scheduling process, while ensuring the service quality of various types of data in the network, and overcomes the fact that the real-time performance of data packets that have high requirements on delay in the current industrial Internet of Things, such as control classes, cannot be guaranteed. It also reduces the packet loss rate of the network.

Referring to Fig. 1, in conjunction with the following specific operation steps of a link selection algorithm based on data priority, transmission time and maximum transmission delay of the present invention:

Step 1. Initialize. Set the current subsystem as the priority subsystem, the data packets in the queue to be sent are sorted by priority, that is, the priority of the data packets from the head of the queue to the tail of the queue is gradually reduced, and the data packets with the same priority are transmitted according to the order of priority. Times are sorted in descending order. The maximum transmission delay of each priority data packet is calculated according to the preset sampling period of each priority data packet in the industrial network and the maximum transmission delay of the data packet with the priority of IPP specified in the network protocol.

表示当前工业网络中优先级为IPP的数据包的最大传输延迟。其表达式如下：Definition 1 Maximum transmission delay

Indicates the maximum transmission delay of the data packet with the priority of IPP in the current industrial network. Its expression is as follows:

为当前工业网络中优先级为IPP的数据包的预设采样周期。由于工业生产中网络中优先级为IPP的数据包的传输延迟小于等于预设采样周期值时，才能保证工业生产的正常进行，所以该工业网络中该优先级的数据包理论传输最大传输时延等于预设采样周期值。

表示网络协议中规定的优先级为IPP的数据包的最大传输延迟。取当前工业网络中优先级为IPP的数据包的预设采样周期和网络协议中规定的优先级为IPP的数据包的最大传输延迟中的较小值作为网络资源调度的最大传输时延。Among them, IPP∈{0,1,2,3,4,5,6,7}, indicating the sending priority of IP packets.

It is the preset sampling period of the data packets with the priority of IPP in the current industrial network. Since the normal operation of industrial production can only be guaranteed when the transmission delay of the data packet with the priority of IPP in the network in industrial production is less than or equal to the preset sampling period value, the theoretical transmission delay of the data packet of this priority in the industrial network is the maximum transmission delay. Equal to the preset sampling period value.

Indicates the maximum transmission delay of a packet with a priority of IPP specified in the network protocol. The smaller value of the preset sampling period of the data packet with the priority of IPP in the current industrial network and the maximum transmission delay of the data packet of the priority of IPP specified in the network protocol is taken as the maximum transmission delay of the network resource scheduling.

的表达式如下：Extension If the preset sampling period values of data packets of different applications of the same priority in the current industrial network are different, the maximum transmission delay

The expression is as follows:

Wherein, i≤A _num , and A _num is the number of application programs that generate data packets of the same priority but with different preset sampling periods in the current industrial network.

在数据链路层协议IEEE802.1Q中指出优先级为4的视频数据的延迟和抖动小于100ms，则

那么，代入表达式(1)，调度过程中优先级为4的数据包最大传输时延

计算如下：For example: Assuming that the preset sampling period of the data packets with priority 4 in an industrial network is the same, which is 120ms, then

In the data link layer protocol IEEE802.1Q, it is pointed out that the delay and jitter of the video data with priority 4 are less than 100ms, then

Then, substituting into expression (1), the maximum transmission delay of the data packet with priority 4 in the scheduling process

The calculation is as follows:

To sum up, from expression (1) or (2), the theoretical maximum value of the transmission delay in the data packet scheduling process can be calculated for an industrial network, which is used for the calculation of the switching conditions of the subsequent hybrid switching system model and the data packet Scheduling provides the basis.

Step 2. Get the data type. When a packet arrives at the IP layer, extract the priority of the packet.

Definition 2 Data priority (β) is used to represent the priority weight of the current data packet in the scheduling process, and the expression is as follows:

Among them, IPP∈{0,1,2,3,4,5,6,7} represents the sending priority of IP packets. The larger the value, the higher the priority. IPP _um is a priority type. In the RFC standard, there are 8 priorities, so IPP _um =8. It can be known from the definition that the range of IPP is 0 to 7, and the value of IPP _um is 8, so IPP<IPP _um , and the data priority is β∈[0,1). The data priority can be judged by β. The larger the β, the higher the data priority.

RFC 1323 introduced the TimeStamp option in the optional field of IP packets. When the flag field (FL) of the timestamp option is 3, the IP address of the sender and the time when the data packet is generated can be recorded. When a packet arrives at the IP layer, in addition to extracting the priority of the packet, it is also necessary to extract the timestamp corresponding to the IP address that is the same as the source address in the timestamp option of the packet. Under the premise of clock synchronization, the sender or forwarding node obtains the transmission time of the data packet by using the difference between the current clock time and the data packet transmission time recorded by the timestamp option, which is represented by T _trans . The sending end or the forwarding node obtains the round-trip delay (Round-Trip Time, RTT) of the link between the node and the receiving end by communicating with the receiving end.

Step 3. Calculate the state parameters. According to the round-trip delay obtained in step 2 and the maximum transmission delay of priority data packets obtained in step 1, the maximum value of the average elapsed time of the queue data packets to be sent in the switching condition is calculated, and the expression is as follows:

为该数据包对应优先级数据包的最大传输时延，RTT为该节点与接收端之间的往返时延，m为待发送队列中数据包数(m≤M，M为待发送队列总长度)，m_IPP为待发送队列中该数据包对应优先级的数据包个数，R_max、R_min为系数，经实验确定，其取值范围为80％≤R_max≤90％，40％≤R_min≤60％时性能较佳。in,

is the maximum transmission delay of the data packet corresponding to the priority data packet, RTT is the round-trip delay between the node and the receiver, m is the number of data packets in the queue to be sent (m≤M, M is the total length of the queue to be sent) ), m _IPP is the number of data packets corresponding to the priority of the data packet in the queue to be sent, R _max and R _min are coefficients, which are determined by experiments, and their value ranges are 80% _≤Rmax≤90 %, 40%≤ The performance is better when R _min ≤ 60%.

Then calculate the average elapsed time of the queue data packets to be sent.

Definition 3 The average transmission time (T) is used to represent the delay of the data packets in the queue to be sent currently. Its expression is as follows:

The current average transmission time T of IP packets in the queue to be sent is jointly determined by the data priority β, the transmission time T _trans , and the number of packets m in the queue to be sent (m≤M, M is the total length of the queue to be sent), and the high The elapsed time of priority packets accounts for a larger proportion of the calculation than lower priority packets. The larger T is, the more data packets in the queue to be sent are close to the maximum transmission delay, and the more urgent the waiting is.

Step 4. Subsystem switching judgment. According to the current subsystem to determine whether the switching condition is satisfied, the switching condition for the priority priority subsystem to switch to the time priority subsystem is the occurrence of the discrete event of a packet about to timeout in the queue to be sent, or the status parameter of the system to the queue to be sent. The average transmitted time is higher than the maximum value; the switching condition for the time priority subsystem to switch to the priority priority subsystem is that the average transmitted time is lower than the minimum value, or when the average transmitted time is lower than the maximum value, the current queue to be sent contains control packets.

When the transmission time (T _trans ) of a data packet satisfies the following expression, it is considered to be about to time out:

为该数据包对应优先级数据包的最大传输时延，RTT为该节点与接收端之间的往返时延。in,

is the maximum transmission delay of the data packet corresponding to the priority data packet, and RTT is the round-trip delay between the node and the receiver.

If the switching conditions are met, the subsystem will be switched, and then the newly received data packets will be inserted into the appropriate position of the queue to be sent according to the ordering rules of the packets in the queue to be sent after the switching; if the switching conditions are not met, directly Insert the newly received data packet into the appropriate position of the to-be-sent queue according to the ordering rules of the data packets in the to-be-sent queue of the current subsystem.

Claims (2)

1. An IP layer data packet scheduling method for multi-service coexistence and deterministic network requirements in the Industrial Internet of Things, characterized in that: first, the data packet scheduling process is designed as a hybrid switching model comprising two subsystems, and the two subsystems are They are the priority priority subsystem and the time priority subsystem, respectively; secondly, the maximum value of the average transmission time of the queue data packets to be sent in the switching conditions is calculated through the relevant parameters; finally, according to whether the state parameters of the subsystems meet the switching conditions and discrete Whether the event occurs or not jointly determines the switching between subsystems. Discrete events include the received control data packet and the existence of the time-out data packet in the queue to be sent; The specific steps of the packet scheduling algorithm are as follows: Step 1. Initialization: Set the current subsystem as the priority subsystem, and calculate each priority through the preset sampling period of each priority data packet in the industrial network and the maximum transmission delay of the data packet with the priority of IPP specified in the network protocol. Maximum transmission delay of priority packets; maximum transmission delay

Represents the maximum transmission delay of the data packet with the priority of IPP in the current industrial network; its expression is as follows:

Among them, IPP∈{0,1,2,3,4,5,6,7}, indicating the sending priority of IP packets;

is the preset sampling period of the data packet with the priority of IPP in the current industrial network, and the theoretical maximum transmission delay of the data packet of this priority in the industrial network is equal to the preset sampling period value;

Indicates the maximum transmission delay of the data packet with the priority of IPP specified in the network protocol; take the preset sampling period of the data packet with the priority of IPP in the current industrial network and the maximum transmission delay of the data packet with the priority of IPP specified in the network protocol The smaller value of the transmission delay is used as the maximum transmission delay of network resource scheduling; If the preset sampling period values of data packets of different applications of the same priority in the current industrial network are different, the maximum transmission delay

The expression is as follows:

Among them, 1≤i≤A _num , A _num is the number of applications that generate the same priority data packets but different preset sampling periods in the current industrial network,

Indicates the preset sampling period of the data packet whose priority is IPP in the ith application; Step 2. Obtain the data type: when a data packet arrives at the IP layer, extract the priority of the data packet and the sending time recorded by the timestamp option; obtain the round-trip delay RTT of the link between the forwarding node and the receiving end; The data priority β is used to represent the priority weight of the current data packet in the scheduling process, and the expression is as follows:

Among them, IPP∈{0,1,2,3,4,5,6,7}, indicating the sending priority of IP packets, the larger the value, the higher the priority; IPP _um is the priority type, in the RFC standard There are 8 priorities, namely IPP _um = 8; therefore IPP<IPP _um , the data priority β∈[0,1); the data priority is judged by β, the larger the β, the higher the data priority; RFC 1323 introduces the timestamp option in the optional field of the IP packet; when the flag field FL of the timestamp option is 3, the IP address of the sender and the time when the data packet is generated are recorded; when a data packet arrives at the IP layer, it is extracted. The priority of the packet and the timestamp corresponding to the same IP address as the source address in the timestamp option of the packet; under the premise of clock synchronization, the sender or forwarding node sends the packet through the current clock time and timestamp option. The time difference is used to obtain the transmission time of the data packet, which is represented by T _trans ; the sending end or forwarding node obtains the round-trip delay of the link between the node and the receiving end by communicating with the receiving end, which is represented by RTT; Step 3. Calculate the state parameters: use the round-trip delay and the maximum transmission delay of the priority data packet to calculate the maximum and minimum values of the average elapsed time of the queue data packets to be sent in the switching condition. The expressions are as follows:

in,

is the maximum transmission delay of the data packet corresponding to the priority data packet, RTT is the round-trip delay between the node and the receiver, m is the number of data packets in the queue to be sent, m _IPP is the corresponding data packet in the queue to be sent The number of data packets of priority, R _max and R _min are coefficients; among them: m≤M, M is the total length of the queue to be sent; Then calculate the average transmission time of the data packets in the queue to be sent. The average transmission time T is used to represent the delay of the data packets in the queue to be sent. The expression is as follows:

The current average transmission time T of the IP packets in the queue to be sent is determined by the data priority β, the transmission time T _trans and the number of packets m in the queue to be sent. The transmission time of high-priority packets is accounted for in the calculation. The ratio is greater than that of low-priority data packets; the larger the T, the more data packets that are close to the maximum transmission delay in the current queue to be sent, and the more urgent it is to be sent; Step 4. Subsystem switching judgment: according to the current subsystem to judge whether the switching condition is satisfied, the switching condition for the priority priority subsystem to switch to the time priority subsystem is the occurrence of the discrete event that there is a timeout packet in the queue to be sent, or The state parameter of the system is that the average transmitted time of the queue to be sent is higher than the maximum value T _max ; the switching condition for the time priority subsystem to switch to the priority priority subsystem is that the average transmitted time is lower than the minimum value T _min , or the average transmitted time is lower than the minimum value T min . When the time is lower than the maximum value T _max , the current queue to be sent contains control data packets; When the transmission time T _trans of the data packet satisfies the following expression, it is considered to be about to time out:

in,

is the maximum transmission delay of the data packet corresponding to the priority data packet, and RTT is the round-trip delay between the node and the receiver; If the switching conditions are met, the subsystem will be switched, and then the newly received data packets will be inserted into the appropriate position of the queue to be sent according to the ordering rules of the packets in the queue to be sent after the switching; if the switching conditions are not met, directly Insert the newly received data packet into the appropriate position of the to-be-sent queue according to the ordering rules of the data packets in the to-be-sent queue of the current subsystem. 2. The IP layer data packet scheduling method for multi-service coexistence and deterministic network requirements in an industrial Internet of Things according to claim 1, wherein: the data in the to-be-sent column in the priority priority subsystem Packets are sorted as follows: the packets in the to-be-sent column in the priority priority subsystem are sorted by priority, that is, the priority of packets from the head of the queue to the tail of the queue is gradually reduced, and the packets with the same priority are sorted by priority. The transmission time is sorted in descending order; the data packets in the queue to be sent in the time priority subsystem are sorted by time priority, that is, the remaining transmission time of the data packets from the head of the queue to the end of the queue increases gradually, and the remaining transmission time is the same The data packets are sorted in descending order of priority; the remaining transmission time is the difference between the maximum transmission delay and the transmission time.

Images