An asynchronous traffic optimization scheduling method of a 5G+TSN fusion network with high schedulability

Abstract

The application discloses a 5G+TSN fusion network asynchronous flow optimization scheduling method with high schedulability, comprising the following steps: S1, in the fusion network of 5G and TSN, first, the overall information of the network is collected by the central controller through the network control protocol, the information includes the channel condition and frequency resource in the 5G network, the network topology and link rate of the TSN network, and the communication demand of the flow; S2, the asynchronous flow optimization scheduling model of the 5G+TSN hybrid network is run through the central controller, the routing decision and transmission scheduling of the flow collected in the above step in the hybrid network are carried out, the scheduling result is obtained, and the consumption of the network resource is minimized while meeting the transmission demand of the hard real-time flow. The application significantly improves the schedulability of the TSN gateway, expands the application range of the asynchronous deterministic access mechanism, and realizes end-to-end flow scheduling.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of deterministic real-time transmission of computer networks, in particular to an asynchronous traffic optimization scheduling method for a 5G+TSN fusion network with high schedulability. BACKGROUND

[0002] The fusion networking of the fifth generation mobile communication technology (5G) and the time sensitive network (TSN) is an important technology to realize flexible manufacturing and mobile applications in industry 4.0. The TSN is based on standard Ethernet and introduces a time-triggered transmission mechanism in IEEE 802.1Qbv. In industrial control, hard real-time traffic is generally modeled as periodic traffic. This mechanism periodically reserves transmission resources for such periodic data streams on the transmission link, i.e., each data packet is scheduled to be sent at a specific time on the routing device. Therefore, the clock synchronization of the entire network is required. One of the three major technical visions of 5G is to achieve ultra-reliable and low-latency transmission. For this purpose, semi-persistent scheduling is introduced, which is similar to the time-triggered transmission of TSN and also reserves periodic transmission resources for data packets. Therefore, it is very suitable for co-networking with TSN to achieve more flexible data transmission. In the existing traffic scheduling of 5G and TSN fusion network, the common practice is to estimate the worst-case latency of the traffic in the 5G network and reserve transmission resources in the TSN network according to the worst-case latency of the 5G. Therefore, the traffic accesses the TSN network according to the worst-case latency. This scheme has the following two problems: (1) it is difficult to estimate the worst-case latency, and (2) it may be difficult to meet the end-to-end latency requirement of the traffic. Treating the traffic from the 5G network as asynchronous terminal traffic and performing asynchronous traffic scheduling is a promising solution.

[0003] In the prior art, such as the "transmission deterministic access method of asynchronous terminal" with Chinese publication number CN202110583698.4, two mechanisms are included, one is the asynchronous-to-synchronous mechanism, and the other is the latency adaptation mechanism. Specifically, the first switch receives the message and resides at the nearest sending time point before sending it out. The message is transmitted to the edge switch according to the time-triggered transmission mechanism of TSN in the synchronous domain. The edge switch sends the message to the receiving terminal after maintaining it for a period of time. The time of this delay sending is equal to the difference between the sending period of the data stream and the residence time of the message, so as to eliminate the uncertainty of the end-to-end latency.

[0004] In the prior art, directly applying the deterministic access method of such asynchronous terminals to the networking and scheduling of 5G and TSN will cause a thorny problem: the scheduling of traffic at the TSN gateway is prone to failure. Because: (1) the TSN gateway plays the role of a bridge between two networks, and all traffic entering the TSN network from the 5G network needs to be converged and distributed at the TSN gateway, so the gateway is highly loaded; (2) the asynchronous deterministic mechanism often needs to provide more transmission resources for traffic; these two factors together are prone to cause the scheduling of traffic at the gateway to fail. In traditional TSN traffic scheduling, typical methods for improving the schedulability of traffic can be divided into two categories: relaxation in the time domain and relaxation in the spatial domain; however, introducing the relaxation in the time domain and the spatial domain into the asynchronous deterministic access mechanism will face the following three unsolved problems:

[0005] (1) How to balance the end-to-end latency of traffic and the consumption of network resources: according to the principle of the asynchronous deterministic access mechanism, the physical meaning of the alignment time is the maximum waiting time of the data packet at the gateway for transmission. Assuming that the average resource period allocated to a flow before and after the introduction of the time domain relaxation is the same, the alignment time will necessarily be greater than that without the introduction of the time domain relaxation due to the non-periodicity of the transmission time slot caused by the time domain relaxation. In order to reduce the alignment time of a flow, more transmission resources can be provided, but the resources left for other traffic will also correspondingly decrease. In the face of complex heterogeneous traffic and its latency requirements in the network, how to comprehensively consider the appropriate amount of resources allocated to each flow?

[0006] (2) How to model the alignment time in the case of relaxation in the time domain and the spatial domain: for the original asynchronous deterministic access mechanism, the alignment time is the resource period; but after the introduction of the relaxation in the time domain and the spatial domain, the alignment time becomes more difficult to calculate; how to model the alignment time of the asynchronous deterministic access mechanism in the case of the introduction of the relaxation in the time domain and the spatial domain?

[0007] (3) How to coordinate the allocation of resources in 5G and TSN networks: in the heterogeneous network of 5G and TSN, the end-to-end latency includes both the latency of TSN and the latency of 5G; if the allocation of resources in the 5G network changes, it will also affect the scheduling of resources in TSN, such as the data packet occupying more frequency resources in order to consume less time domain resources, the latency of 5G will become smaller, and the scheduling pressure on the TSN network will become smaller, but more frequency resources will be consumed. In the face of complex traffic in the network, how to coordinate the allocation of resources in 5G and TSN networks? SUMMARY

[0008] The purpose of the present application is to provide an asynchronous traffic optimization scheduling method for a 5G+TSN fusion network with high schedulability, to solve the problems of how to balance the end-to-end delay of traffic and the consumption of network resources, how to model the alignment time in the case of time domain and space domain relaxation, and how to coordinate the allocation of 5G and TSN network resources.

[0009] To achieve the above purpose, the present application provides the following technical scheme: an asynchronous traffic optimization scheduling method for a 5G+TSN fusion network with high schedulability, comprising the following steps:

[0010] S1, in the fusion network of 5G and TSN, first collect the overall information of the network by the central controller through the network control protocol, which includes the channel conditions and frequency resources in the 5G network, the network topology and link rate of the TSN network, and the communication requirements of the traffic;

[0011] S2, run the asynchronous traffic optimization scheduling model of the 5G+TSN hybrid network by the central controller, make routing decisions and transmission scheduling for the data flow set of the network overall information collected in step S1 in the hybrid network, obtain the scheduling result, and minimize the consumption of network resources while meeting the transmission requirements of hard real-time traffic;

[0012] S3, calculate the configuration information of each network device according to the scheduling result obtained in step S2, which includes the queue gating list of the TSN switch, and the resource time configuration of the gateway and edge switch in the asynchronous access deterministic mechanism;

[0013] S4, in the running phase, when the data packet arrives at the TSN gateway from the 5G network, the gateway identifies which traffic the data packet belongs to, delivers it to the nearest time slot transmission prepared for the traffic, adds the waiting time of the data packet to the back of the data packet payload, and then encapsulates the header and trailer in the standard Ethernet way;

[0014] S5, in the running phase, when the data packet arrives at the edge switch through the transmission of TSN, calculate the residence time, according to the original asynchronous deterministic access mechanism, the residence time should be the difference between the alignment time and the waiting time at the gateway, and an additional residence time needs to be introduced, whose value is the difference between the maximum TSN delay of all candidates of the data packet traffic set and the TSN delay of the actual candidate of the transmitted data packet;

[0015] Among them, the slot resource set that can provide continuous transmission from the gateway to the destination node for a data packet on one route is called a candidate.

[0016] Preferably, in step S2, the asynchronous traffic optimization scheduling model of the 5G+TSN hybrid network comprises an alignment time model under time domain and space domain relaxation conditions, a 5G network resource scheduling model, and a TSN network resource scheduling model.

[0017] Preferably, the alignment time model under time domain and space domain relaxation conditions comprises a resource allocation scheme of an asynchronous deterministic access mechanism under time domain and space domain relaxation conditions and an alignment time model under the scheme.

[0018] The basic resource cycle provided by the TSN in the resource allocation scheme is , which can provide a set of slot resources for continuous transmission of a data packet from the gateway to the destination node on one route, referred to as a candidate, and each set of four slots constitutes a candidate. The scheduling model needs to decide the number of candidates allocated to each flow to implement the resource allocation of the asynchronous deterministic access mechanism and determine the route taken by each candidate and the scheduling of the candidate on the route.

[0019] The alignment time model is a calculation of the alignment time. The physical meaning of the alignment time is the maximum waiting time of a data packet at the gateway. Therefore, a logical link can be abstracted for a flow on the gateway for candidates dispersed on different egress links. On the logical link, the offset of the th candidate of the flow at the gateway is represented as , and the time domain relaxation is calculated on the obtained logical link.

[0020] The maximum number of candidates that can be provided by one is , and the index starts from to . Each candidate has a binary variable representing whether the candidate is selected. To ensure the physical meaning of the candidate, it should complete transmission within the period , and it is easy to obtain:

[0021] ;

[0022] To distinguish each candidate, the relative positions between the candidates are constrained:

[0023] ; ​

[0024] Then, for a flow its alignment time can be calculated by the following theorem:

[0025] Theorem 1:

[0026] ;

[0027] where two sets and are defined as:

[0028] ;

[0029] .

[0030] Preferably, the transmission resource in the 5G network resource scheduling model is located in two dimensions: time domain and frequency domain, and the basic scheduling unit in the 5G resource grid occupies 1 resource block (RB) in frequency and one transmission time interval (TTI) in time domain. One scheduling unit can carry a certain byte transmission.

[0031] The set of RBs available to the 5G network is represented as , where represents the number of RBs.

[0032] In the 5G network, that is, , the resources allocated to a flow are composed of the following decision variables:

[0033] ;

[0034] where the physical meaning of each decision variable is: is the period of the resource equal to the period of the flow, the start time of the resource , the duration of the resource , represents whether the flow occupies the kth RB of the 5G link l, represents that the RB labeled k is allocated to the flow ;

[0035] The time domain resource is allocated continuously, and the frequency domain resource can be allocated discontinuously. Since the periods of the flows can be different, the least common multiple of their periods is defined as the macro period, denoted as , that is:

[0036] .

[0037] Preferably, the decision variables need to satisfy the following constraints:

[0038] Transmission opportunity constraint: a data packet can only start transmission at an integer time of TTI, the duration of transmission must also be an integer of TTI, and the data packet must be transmitted within a period;

[0039] Transmission resource constraint: the resource allocated for a data packet is guaranteed to be sufficient for data transmission;

[0040] OFDMA constraint: the first two constraints ensure the effectiveness of resource allocation for a single flow, and the OFDMA constraint ensures that the resources between multiple flows will not conflict, i.e., a RB-TTI block can be occupied by at most one flow;

[0041] RB constraint: introduce a variable to indicate whether the kth RB of the 5G link l is allocated for traffic transmission.

[0042] Preferably, the TSN network resource scheduling model is in a TSN network, i.e. The resource allocated for a flow is composed of the following decision variables:

[0043]

[0044] ;

[0045] The meaning of each decision variable is as follows: represents whether the jth candidate of the flow is used, represents whether the route of the jth candidate of the flow contains the link l, then determines the transmission of the jth candidate of the flow on the link l, where is a period, the size of which is , is an offset decision variable, is the transmission duration.

[0046] Preferably, the decision variables need to satisfy the transmission constraints, i.e., routing constraint, frame constraint, resource constraint, alignment time constraint, transmission order constraint, TDMA constraint, frame isolation constraint, and end-to-end delay constraint.

[0047] Preferably, in step S2, the transmission requirement of the hard real-time traffic is satisfied while minimizing the consumption of network resources, which includes a 5G network part and a TSN network part.

[0048] The 5G network part, minimize the number of RBs allocated to hard real-time critical traffic, since 5G system only has one uplink, we omit the subscript l for the following formula, that is:

[0049] ;

[0050] The TSN network part, minimize the number of candidates allocated throughout the network, that is:

[0051] ;

[0052] In summary, change the optimization direction and add a weight factor γ, the final optimization goal is:

[0053] .

[0054] Compared with the prior art, the beneficial effects of the present application are:

[0055] (1) Significantly improve the schedulability of TSN gateway: by introducing the relaxation of time domain and space domain, it is easier to meet the non-overlapping of transmission time slots allocated by TSN to different flows, thereby improving the schedulability at the gateway.

[0056] (2) Expand the application scope of asynchronous deterministic access mechanism: the original asynchronous deterministic access mechanism assumes to provide periodic resources for traffic, and the present application breaks this assumption, making the application scope of the asynchronous deterministic access mechanism wider.

[0057] (3) End-to-end traffic scheduling: the present application models the end-to-end transmission, comprehensively considers the network and resource characteristics of 5G and TSN, the delay experienced by traffic in 5G and TSN, and leaves as many resources as possible for low-priority traffic under the condition of meeting the traffic transmission requirements. BRIEF DESCRIPTION OF DRAWINGS

[0058] Figure 1 Fig. 1 is a schematic diagram of time-triggered transmission process in 5G and TSN fusion network under ideal conditions;

[0059] Figure 2 Fig. 2 is a schematic diagram of deterministic access scheme for asynchronous terminals;

[0060] Figure 3 Fig. 3 is a schematic diagram of scheduling failure of two flows on one link;

[0061] Figure 4 Fig. 4 is a schematic diagram of time domain relaxation;

[0062] Figure 5 Fig. 5 is a schematic diagram of space domain relaxation;

[0063] Figure 6Asynchronous deterministic access mechanism diagram with time domain and space domain relaxation for application;

[0064] Figure 7 Alignment time calculation diagram on logical link;

[0065] Figure 8 5G network transmission resource diagram;

[0066] Figure 9 5G+TSN asynchronous traffic optimization scheduling method with high schedulability. DETAILED DESCRIPTION

[0067] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of the present application.

[0068] Please refer to Figures 1-9 The present application provides a technical solution: an asynchronous traffic optimization scheduling method for a 5G+TSN fusion network with high schedulability, comprising the following steps:

[0069] Step one, in the fusion network of 5G and TSN, first collect the overall information of the network by the central controller through the network control protocol, which includes the channel conditions and frequency resources in the 5G network, the network topology and link rate of the TSN network, and the communication requirements of the traffic;

[0070] Step two, run the asynchronous traffic optimization scheduling model of the 5G+TSN hybrid network by the central controller, make routing decisions and transmission scheduling for the data flow set of the overall network information collected in the previous step in the hybrid network, obtain the scheduling result, and minimize the consumption of network resources while meeting the transmission requirements of hard real-time traffic;

[0071] Among them, the asynchronous traffic optimization scheduling model of the 5G+TSN hybrid network includes the alignment time model under the time domain and space domain relaxation condition, the 5G network resource scheduling model and the TSN network resource scheduling model;

[0072] Specifically, the alignment time model under the time domain and space domain relaxation condition includes the resource allocation scheme of the asynchronous deterministic access mechanism under the time domain and space domain relaxation condition and the alignment time model under this scheme; the basic resource period provided by the TSN in the resource allocation scheme is A candidate is a set of slot resources that can provide continuous transmission of a data packet from the gateway to the destination node on a route. Each of the four slots constitutes a set, and each set is a candidate. The scheduling model needs to decide the number of candidates allocated to each flow in order to realize the resource allocation of the asynchronous deterministic access mechanism and determine the route that each candidate traverses and its scheduling on the route.

[0073] The alignment time model models the calculation of alignment time. The physical meaning of alignment time is the maximum waiting time for a data packet to be transmitted at the gateway. Therefore, it can abstract a logical link for candidates of a flow scattered across different outgoing links at the gateway. On this logical link, the flow... The The offset of each candidate at the gateway is represented as Time-domain relaxation is calculated on the obtained logical link;

[0074] Defined as one The maximum number of candidates that can be provided is Its number starts from Start to Each candidate has a binary variable. This indicates whether the candidate has been selected. To ensure the physical meaning of the candidate, it should be within a certain period. The transmission is completed internally, making it easy to obtain:

[0075] ;

[0076] To differentiate between each candidate, constraints are imposed on the relative positions of the candidates:

[0077] ;

[0078] So, for flow Regarding its alignment time The following theorems can be used to calculate:

[0079] Theorem 1:

[0080] ;

[0081] Two sets and The definitions are as follows:

[0082] ;

[0083] .

[0084] Proof:

[0085] After abstracting the logical link, we prove the theorem by taking Figure 7 as an example of heuristic. According to the meaning of , i.e. the maximum interval between adjacent candidates, then,

[0086] ;

[0087] where represents the interval between different and adjacent candidates within a period, which is contained in the set ; and represents the interval between the last candidate and the first candidate in the next period, which is contained in the set , so that is the maximum interval between adjacent candidates in Figure 7 , i.e. the alignment time. Generally, in order to find the maximum interval between adjacent candidates, for the set , the right half of each element is non-negative if and only if the pth and qth candidates exist and there is no candidate between them (i.e. , ), and the left half of the element is 1, in which case the element is non-negative and represents the interval between two existing, different and adjacent candidates, if there is another candidate between p and q, then the element is negative. Similarly, for the set , the right half of each element is easily obtained to be positive if and only if the pth and qth candidates exist and there is no candidate before p and after q (i.e. , ), and the left half of the element is 1, in which case the element is positive and represents the interval between the last candidate in the period and the first candidate in the next period, if there is another candidate before p or after q, then the value of the element is negative; p can be equal to q in order to deal with the case of assigning only one candidate to a flow; since at least one candidate is provided to the flow, the set must have and only have one positive element. Finally, since , It must be positive, which also explains is the maximum interval of adjacent candidates, and the proof is complete.

[0088] Specifically, the transmission resource in the 5G network resource scheduling model is located in two dimensions: time domain and frequency domain. In the 5G resource grid, the basic scheduling unit occupies one resource block (RB) in frequency and one transmission time interval (TTI) in time domain. A scheduling unit can carry a certain number of byte transmissions.

[0089] The set of RBs available to the 5G network is represented as where represents the number of RBs.

[0090] In the 5G network, that is, The resources allocated to a flow are composed of the following decision variables:

[0091] ;

[0092] The physical meaning of each decision variable is as follows: is the start time of the resource , the duration of the resource , represents whether the flow occupies the kth RB of the 5G link l, represents that the RB labeled k is allocated to the flow ;

[0093] The time domain resource is allocated continuously, and the frequency domain resource can be allocated non-continuously. Since the periods of the flows may be different, the least common multiple of their periods is defined as the macro period, denoted as , that is:

[0094] .

[0095] The decision variables need to satisfy the following constraints:

[0096] (1) Transmission opportunity constraint: data packets can only start transmission at integer times of TTI, the duration of transmission must also be an integer of TTI, and the data packets must be transmitted within a period; therefore:

[0097]

[0098]

[0099]

[0100]

[0101] ;

[0102] where variable integer means the flow TTI number that starts to transmit on 5G link , and represents the number of TTIs that continuously transmit;

[0103] (2) Transmission resource constraint: the resource allocated for a data packet guarantees enough transmission data, so that represents the number of bytes that can be transmitted by a TTI on the kth RB of 5G link l, so:

[0104]

[0105] ;

[0106] The above two constraints make the allocated resource just enough to transmit data;

[0107] (3) OFDMA constraint: the first two constraints ensure the effectiveness of allocating resources to a single flow, and the OFDMA constraint ensures that the resources between multiple flows will not conflict, that is, a RB-TTI block can be occupied by at most one flow, so:

[0108]

[0109] wherein indicates that when the ith flow and the jth flow are transmitted through the kth RB of 5G link l, the time domain resources of the two flows are constrained to be non-overlapping;

[0110] (4) RB constraint: introduce variable to indicate whether the kth RB of 5G link l is allocated for traffic transmission, specifically, if any traffic is allocated to the kth RB, the indicating variable is set to 1, so:

[0111] ;

[0112] Specifically, the TSN network resource scheduling model is in a TSN network, i.e. The resource allocated to one flow is composed of the following decision variables:

[0113]

[0114] ;

[0115] The meaning of each decision variable is as follows: represents whether the jth candidate of the flow is used, represents whether the route of the jth candidate of the flow contains the link l, then determines the transmission of the jth candidate of the flow on the link l, wherein is a period (the size of which is , which is priori knowledge), is the offset, which is a decision variable, is the transmission duration;

[0116] Among them, the decision variable needs to meet the transmission constraints, which are route constraints, frame constraints, resource constraints, alignment time constraints, transmission order constraints, TDMA constraints, frame isolation constraints, and end-to-end delay constraints.

[0117] (1) Route constraints: a route needs to be scheduled for each selected candidate, and when a candidate is not selected, no route is allocated. Therefore, the value of needs to be modeled. When the flow uses the jth candidate (i.e. ), we need to allocate a route, and we only consider simple acyclic routes. The following three constraints respectively depict the characteristics of the sending node (in the scenario considered in the present application, it is a TSN gateway), the intermediate forwarding node, and the acyclic constraint in TSN: the gateway node is the sending node of the flow , and only one output link is activated (the link l is activated, i.e. ). The intermediate link of the flow route only bears the functions of receiving and forwarding, so the number of activated input links is equal to the number of activated output links. In order to limit the acyclic, we limit that a node has at most one link activated.

[0118]

[0119] ;

[0120]

[0121]

[0122] ;

[0123] Additionally, in the flow , we want that for this candidate, all the input, output links of the node are not activated by this candidate, i.e.

[0124] .

[0125] (2) Frame constraint: each candidate needs to have non-negative offset on each link, and for modeling convenience, it needs to complete transmission in a period , in addition, if a candidate does not pass through a link l, its offset should be 0, thus:

[0126]

[0127]

[0128]

[0129]

[0130]

[0131]

[0132] ;

[0133] where the last hop we do not schedule (i.e. , same below);

[0134] It is worth noting that we have already obtained in the routing constraint, and we have here, thus we can conclude that if a candidate of a flow is not used, all the links will not be activated by this candidate and the offset on the links will be 0. If it is used, only on the activated links its offset can not be 0;

[0135] ​(3) Resource constraint: enough resource should be allocated for a flow in TSN, i.e. the number of provided candidates should not be less than the number of data packets appeared in , thus:

[0136]

[0137] ;

[0138] (4) Alignment time constraint: the purpose of this constraint is to abstract a logical link for the candidates of a flow which are scattered in the whole network at the gateway, i.e. to get the value of variable . When a candidate is not used, there is no need to get the value. When used, according to the previous constraint, there is and only one output link of the gateway is activated while other links are not activated (offset is 0), at this time is needed to capture its offset at the gateway. Since we do not know which output link of the gateway the candidate is transmitted on, we need to accumulate the offset on all output links, thus:

[0139] ;

[0140] (5) Transmission order constraint: the switch in the network can only transmit on the next link after completely receiving the data packet of the previous link, let represent the propagation delay of link l. Since we do not know which output or input link of a node is activated, we need to accumulate all input or output links (the same below), thus:

[0141]

[0142]

[0143]

[0144] When the jth candidate of flow is not selected (i.e. ), the inequality is naturally established. When selected (i.e. ), if the route does not pass through node v, it is also naturally established; while passing through node v, the constraint represents the transmission order constraint of the sending time of the previous and next links;

[0145] (6) TDMA constraint: On TSN link, reserved transmission slots cannot overlap each other, TDMA constraint needs to be followed. Note that the constraint is only needed when two candidates go through the same link (i.e. ), thus:

[0146] ;

[0147] (7) Frame isolation constraint: due to the use of asynchronous deterministic access mechanism, the transmission resource provided for a flow is over-provisioned, in order to guarantee the determinism of queue queuing in the network, each transmission instance needs to be isolated in the queue, thus:

[0148]

[0149]

[0150]

[0151]

[0152]

[0153]

[0154]

[0155]

[0156] ;

[0157] wherein makes the link l have a preceding link;

[0158] (8) End-to-end latency constraint: let the processing latency of the base station be , then the latency experienced by the 5G system is:

[0159]

[0160] ;

[0161] According to the deterministic asynchronous access mechanism, the latency of each candidate in TSN can be calculated according to the following formula:

[0162]

[0163] ;

[0164]

[0165] ;

[0166] wherein denotes the interval between the time the data packet is sent to the edge switch and the time it is sent at the gateway, is the transmission and propagation delay of the last hop.

[0167] Since the transmission of different candidates assigned to one flow can have different delays in TSN, in order to keep all the candidates used by the flow have consistent TSN delay, we distinguish the data packets from different candidates at the edge switch and add some residence time based on the residence time of the asynchronous deterministic access mechanism, so that the theoretical transmission time of the candidate in TSN is consistent with the TSN delay of the candidate with the largest TSN delay, therefore, the TSN delay experienced by a flow should be the maximum of all candidate TSN delays, so the end-to-end delay should meet the delay requirement of the flow, that is:

[0168]

[0169]

[0170] ;

[0171] Specifically, meeting the transmission requirements of hard real-time traffic while minimizing the consumption of network resources includes 5G network part and TSN network part;

[0172] 5G network part, minimizing the number of RBs allocated for hard real-time critical traffic, since the 5G system only has one uplink, we omit the subscript l for the following formula, that is:

[0173] ;

[0174] TSN network part, minimizing the number of candidates allocated throughout the network, that is:

[0175] ;

[0176] In summary, change the optimization direction and add a weight factor γ, the final optimization goal is:

[0177] .

[0178] Step three, calculate the configuration information of each network device according to the scheduling result obtained in the previous step, including the queue gating list of the TSN switch, the resource time configuration of the gateway and the edge switch in the asynchronous access deterministic mechanism;

[0179] Step four, in the running phase, when the data packet arrives at the TSN gateway from the 5G network, the gateway identifies which flow the data packet belongs to, delivers it to the nearest time slot transmission prepared for the flow, adds the waiting time of the data packet to the back of the data packet payload, and then encapsulates the header and the tail according to the standard Ethernet mode;

[0180] Step five, in the running phase, when the data packet arrives at the edge switch through the transmission of the TSN, the residence time is calculated, and according to the original asynchronous deterministic access mechanism, the residence time should be the difference between the alignment time and the waiting time at the gateway, and an additional residence time needs to be introduced, which is the difference between the maximum TSN delay of all candidates of the data packet flow set and the TSN delay of the actual candidate of the transmitted data packet;

[0181] Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to part of the technical features, and any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. An asynchronous traffic optimization scheduling method for a 5G+TSN fusion network with high schedulability, characterized in that, The method comprises the following steps: S1, in the fusion network of 5G and TSN, first, the overall information of the network is collected by the central controller through the network control protocol, including the channel condition and frequency resource in the 5G network, the network topology and link rate of the TSN network, and the communication demand of the flow; S2, the asynchronous flow optimization scheduling model of the 5G+TSN hybrid network is run by the central controller, the data flow set of the overall information of the network collected in step S1 is routed and scheduled in the hybrid network, the scheduling result is obtained, and the consumption of the network resource is minimized while meeting the transmission demand of the hard real-time flow; S3, the configuration information of each network device is calculated according to the scheduling result obtained in step S2, including the queue gating list of the TSN switch, and the resource time configuration of the gateway and the edge switch in the asynchronous access deterministic mechanism; S4, in the running stage, when the data packet arrives at the TSN gateway from the 5G network, the gateway identifies to which flow the data packet belongs, delivers the data packet to the nearest time slot transmission prepared for the flow, adds the waiting time of the data packet to the rear of the data packet payload, and then encapsulates the header and the trailer in the standard Ethernet mode; S5, in the running stage, when the data packet is transmitted through the TSN and arrives at the edge switch, the residence time is calculated, the residence time is the difference between the alignment time and the waiting time at the gateway according to the original asynchronous deterministic access mechanism, and an additional residence time is also introduced, and the value of the additional residence time is the difference between the maximum TSN delay of all candidates of the data packet flow set and the TSN delay of the actual candidate of the transmitted data packet. Wherein, the slot resource set capable of providing continuous transmission from the gateway to the destination node for a data packet on one route is called a candidate.

2. The asynchronous traffic optimization scheduling method of the 5G+TSN converged network with high schedulability according to claim 1, characterized in that, In step S2, the asynchronous flow optimization scheduling model of the 5G+TSN hybrid network comprises an alignment time model under the time domain and space domain relaxation conditions, a 5G network resource scheduling model and a TSN network resource scheduling model.

3. The asynchronous traffic optimization scheduling method of the 5G+TSN converged network with high schedulability according to claim 2, characterized in that, The alignment time model under the time domain and space domain relaxation conditions comprises a resource allocation scheme of the asynchronous deterministic access mechanism under the time domain and space domain relaxation conditions and an alignment time model under the scheme; In the resource allocation scheme, a period of a basic resource provided by the TSN, i.e., a period of the candidate, is The scheduling model needs to determine the number of candidates allocated to each flow to implement the resource allocation of the asynchronous deterministic access mechanism, and determine the route passed by each candidate and the scheduling of each candidate on the route. The alignment time model is a calculation of alignment time, whose physical meaning is the maximum waiting time of a data packet at the gateway waiting for transmission, so a logical link can be abstracted for a flow at the gateway, which is dispersed in different candidate of the exit link, on which the offset of the first candidate of the flow at the gateway is represented as , and the alignment time calculation model of the obtained logical link is obtained. ​​ one The maximum number of candidates that can be provided is , whose label starts from to , each candidate has a binary variable , representing whether the candidate is selected or not. In order to ensure the physical meaning of the candidate, it should be completed within the transmission period , it is easy to get: ； In order to distinguish each candidate, the relative positions between the candidates are constrained: ； So, for the stream its alignment time can be calculated by the following theorem: Theorem 1: ； where two sets and are defined as: ； 。 4. The asynchronous traffic optimization scheduling method of the 5G+TSN converged network with high schedulability according to claim 2, characterized in that, The transmission resource in the 5G network resource scheduling model is located in two dimensions: time domain and frequency domain, and the basic scheduling unit in the 5G resource grid occupies 1 resource block (RB) in frequency and one transmission time interval (TTI) in time domain, and one scheduling unit can carry a certain byte transmission; The set of RBs available for use by a 5G network is denoted as where represents the number of RBs. In 5G networks, i.e. The resources allocated to one traffic are composed of the following decision variables: ； where the physical meaning of each decision variable is: the period of the resource is equal to the period of the flow, the start time of the resource , the duration of the resource , represents whether the flow occupies the kth RB of the 5G link l, represents that the RB with label k is allocated to the flow ; The time domain resource is continuously allocated, and the frequency domain resource can be discontinuously allocated. Since the periods of the traffics can be different, the least common multiple of their periods is defined as a macro period, denoted as That is: 。 5. The asynchronous traffic optimization scheduling method of a 5G+TSN converged network with high schedulability according to claim 4, characterized in that, The decision variable needs to meet the following constraints: Transmission opportunity constraint: the data packet can only start transmission at an integer time of TTI, the transmission duration must also be an integer of TTI, and the data packet must also be transmitted within a period; Transmission resource constraint: the resource allocated for a data packet is guaranteed to be sufficient for data transmission; OFDMA constraint: the first two constraints ensure the validity of resource allocation for a single stream, and the OFDMA constraint ensures that there is no conflict between multiple streams, i.e., a RB-TTI block can be occupied by at most one stream; RB constraint: introduce variable to indicate whether the k-th RB of the 5G link l is allocated for traffic transmission.

6. The asynchronous traffic optimization scheduling method of the 5G+TSN converged network with high schedulability according to claim 2, characterized in that, The TSN network resource scheduling model is in a TSN network, i.e. The resources allocated to one traffic consist of the following decision variables: ； The meaning of each decision variable is: represents whether the jth candidate of flow is used, represents whether the route of the jth candidate of flow contains link l, then determines the transmission of the jth candidate of flow on link l, where is the period, whose size is , is the offset, which is a decision variable, is the transmission duration.

7. The asynchronous traffic optimization scheduling method of a 5G+TSN converged network with high schedulability according to claim 6, characterized in that, The decision variable needs to satisfy the transmission constraints: routing constraints, frame constraints, resource constraints, alignment time constraints, transmission order constraints, TDMA constraints, frame isolation constraints and end-to-end delay constraints.

8. The asynchronous traffic optimization scheduling method of a 5G+TSN converged network with high schedulability according to claim 1, characterized in that, In step S2, the transmission requirement of the hard real-time traffic is satisfied while the consumption of network resources is minimized, including a 5G network part and a TSN network part; The 5G network part minimizes the number of RBs allocated for hard real-time critical traffic, and since the 5G system has only one uplink, we omit the subscript l for the following formula, i.e., ； The TSN network part minimizes the number of candidates allocated throughout the network, i.e., ； In summary, change the optimization direction and add a weight factor γ, and the final optimization goal is: 。

Images