A time slot allocation method under multi-service flow coexistence

Abstract

The application belongs to the technical field of wireless communication, and particularly relates to a time slot allocation method under multi-service flow coexistence, which comprises the following steps: in a station, data mapping and queue management are performed on generated data according to a destination mac address and a priority attribute; a device node extracts queue state from a current data queue and reports the queue state to an access point; after receiving, the access point performs mathematical modeling on periodic and aperiodic data services to divide time slots; predicted time slot allocation information is sent to the device node via a beacon frame; and data is sent in a time slot allocated to the device node according to a time slot allocation result issued by the access point. The application realizes effective guarantee of time-sensitive data transmission in a network through time division and time slot retransmission mechanism; through feedback of queue information by the device node, controller statistical analysis of transmission data of each device node in the network, higher utilization of network resources, and optimization of end-to-end time delay, the real-time performance and reliability of time-sensitive data are satisfied.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology, specifically relating to a time slot allocation method under the coexistence of multiple service flows. Background Technology

[0002] With the rapid development of commercial spaceflight, the functions of spacecraft that meet people's living and working needs have become more complex. Existing spacecraft internal data transmission buses commonly use SpaceWire (SpW), 1553B, CAN and other buses for data communication. Among them, SpaceWire (SpW) is one of the most reliable wired communication technologies in satellites, providing data rates up to 200Mbps. The European Space Agency, NASA, JAXA and RosCosmos use it for Earth observation, commercial and other flight missions that require medium or high data rates. However, the quality of satellite wiring greatly increases launch costs. At the same time, the mobile communication needs of astronauts are constantly increasing. In order to reduce the weight of spacecraft and meet the needs of mobile communication, efforts are being made to replace wired communication between airborne equipment with wireless links. In addition, the use of wireless networks in spacecraft systems can also bring other substantial benefits, such as: (1) reducing the workload of wiring harness design and wiring, (2) being easier to handle and adapt, and (3) simplifying the assembly, integration and testing (AIT) process.

[0003] Wireless networks have broad application prospects in satellite networks, but they also face significant challenges. Spacecraft systems contain numerous sensors used to monitor and control mission-critical functions. Various types of sensors track on-orbit performance, but sensors are also used in aerospace technology activities to support the detection of various experimental conditions on the spacecraft. Sensors are typically assigned to relevant subsystems of the satellite: 1) Thermal subsystem: Sensors monitoring temperature limits; typical sampling rates are 10-30 seconds; no requirements for latency or determinism. 2) Mechanical subsystem: Sensors primarily used for aerospace information technology experimental activities; typically used for on-orbit use of mechanisms. 3) Attitude and Orbit Control Subsystem (AOCS): Various sensors and actuators controlling attitude; stringent requirements for latency, reliability, and determinism; high sampling rates (up to 100Hz). AOCS is a time-sensitive subsystem, heavily reliant on deterministic and low-latency network communication; other subsystems also exhibit varying degrees of time sensitivity. Real-world needs have driven the development of relevant standards and protocols. The Time-Sensitive Network (TSN) proposed by the IEEE 802.1 task force in 2012 can guarantee that time-sensitive and non-time-sensitive data streams can be transmitted on the same network, and that the transmission of non-time-sensitive data streams will not affect the transmission latency of time-sensitive data streams. However, research on wireless TSN technology is still in its initial stages, and there is no unified standard. Current research involves modifying the IEEE 802.11 protocol from the original Carrier Sense Multiple Access (CSMA) to Time Division Multiple Access (TDMA) to generate low-latency and high-reliability wireless networks. Therefore, wireless TSN technology can be introduced into the internal networks of spacecraft.

[0004] However, when wireless mechanisms are introduced into spacecraft, communication nodes need to wait for their own wireless communication time slots before they can send data. If nodes cannot acquire time slots in time, random delays in data transmission will occur within the nodes, failing to meet the latency requirements of time-sensitive streams. Therefore, the allocation of time slots significantly impacts data transmission latency. Thus, the allocation of communication time slots in TDMA-based wireless networks is a key research focus. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention proposes a time slot allocation method for multiple service flows coexisting, comprising:

[0006] S1: The site internally maps and manages queues for the generated data based on the data's destination MAC address and priority attribute.

[0007] S2: The device node STA continuously extracts the queue status from the current data queue and reports the extracted queue status to the access node AP through feedback frames during the CSMA phase by competing for channel resources;

[0008] The queue status includes the queuing time period and data volume information of time-sensitive business data and non-time-sensitive business data with different time sensitivity and different generation periods;

[0009] The feedback frame consists of four main parts: header, feedback information, padding field and FCS field. The feedback information includes frame identification code, frame length information, device node association identification code and data queue length information.

[0010] S3: After receiving the queue information in the feedback frame, the access node (AP) performs mathematical modeling on periodic and non-periodic data services to divide them into time slots and obtain time slot allocation information.

[0011] S4: The controller of the access node AP adds the time slot allocation information to the extended field of the beacon frame periodically sent by the access node AP, obtains the new time slot allocation result, and sends it to the device node;

[0012] S5: The device node receives the new time slot allocation result issued by the controller in the access node (AP), updates the local time slot allocation table according to the new time slot allocation result, and sends the data in the time slot allocated in the updated local time slot allocation table.

[0013] Preferably, S1 specifically includes:

[0014] Extract the destination MAC address and priority attribute of the data generated within the site, generate a hash value, and store the wireless data frame in the corresponding buffer queue according to the generated hash value: if the hash value of the wireless data frame is a specific set of values, then the data packet is time-sensitive data and is stored in the time-sensitive list; otherwise, that is, the data packet is non-time-sensitive data, then the data packet is stored in the non-time-sensitive list and waits to be sent.

[0015] Preferably, time slots are divided by mathematical modeling for periodic and non-periodic data services, including:

[0016] For non-periodic data services, a Poisson distribution is used to process them, treating them as special periodic data that appears only once in the current superframe during each allocation, and planning them together with periodic data;

[0017] For periodic data services, calculate the time slot allocation information in the next time slot allocation table: total superframe length, unit time slot length, and node transmission time slot position.

[0018] Furthermore, a Poisson distribution is used to process non-periodic data services, including:

[0019]

[0020] in, This indicates that a Poisson distribution is used to process non-periodic data services, f i C (x) represents the number of data frames C included in the service. i,j The characteristic function, f i C (x)=exp[exp(ixC i,j )], C i,j Indicates the number of data frames included in the service, δ i Let t represent the arrival rate of the event, t represent the time period for calculation, x represent the number of times business data arrives within the time period t, i represent business i, and exp represent the expected operation.

[0021] Furthermore, the total superframe length is calculated, including:

[0022] T td =[T i ,...,T N ]+k(2·|T0|+T trans +T SIFS +T ACK )

[0023] Among them, T td T represents the total superframe length. i T represents the time-sensitive data stream cycle period, T0 represents the maximum error that synchronization may produce, and T trans T represents the average propagation time between wireless links. SIFS T represents the shortest inter-frame interval. ACK This indicates the time taken to transmit the ACK, where k represents the number of time slots in the TDMA stage. N This represents the total number of nodes participating in time slot allocation.

[0024] Furthermore, the calculation of the unit time slot length includes:

[0025]

[0026] Subject to β i +...+β N =1

[0027] β i ,...β N ∈[0,1]

[0028] Among them, T s β represents the unit time slot length. i This indicates that the device node d is assigned to it.i The weights, T p T represents the length of the protection time interval in the time slot, T0 represents the maximum error that synchronization may produce, and T trans T represents the average propagation time between wireless links. SIFS T represents the shortest inter-frame interval. ACK This indicates the time taken to transmit the ACK response message. This represents the amount of data in the device node's service flow, where v represents the data transmission rate. N This represents the total number of nodes participating in time slot allocation.

[0029] Furthermore, the location of the node's transmission time slot is calculated, including:

[0030]

[0031] Subject to

[0032] Among them, T i,p T represents the time when time-sensitive data from device node i arrives and cannot be sent, thus initiating a queuing process. i,t This represents the time slot location allocated to node i at the access point, where N represents the total number of nodes participating in the time slot allocation, and f i,j ∈{0,1} indicates whether time slot j belongs to terminal d. i f i,j =1 indicates that time slot j belongs to terminal d i Conversely, it does not belong to terminal d. i T td This indicates the total superframe length.

[0033] Preferably, S5 specifically includes:

[0034] Step 1: During the CSMA phase, the device node obtains the new time slot allocation result from the beacon frame and extracts the allocation information. Among them, the time-sensitive radio node in the device node reads the identity identifier corresponding to the current time slot number in its own time slot table at the start of each time slot to determine whether the current time slot belongs to the radio node.

[0035] Step 2: After compensating for time synchronization error deviation based on the allocation information, calculate the current superframe sequence number SP based on the compensated timestamp. id ;

[0036] Step 3: Calculate the slot number j in the current superframe based on the current local time and the superframe sequence number;

[0037] Step 4: Based on the calculated timeslot number, compare it with the new timeslot allocation result obtained in Step 1 to determine whether the current timeslot belongs to this node. If it belongs to this node, extract the data through the queue management module and wait to send it; otherwise, do not send the data and continue to wait for the timeslot belonging to this node.

[0038] Step 5: Check if there is any data to be sent that requires high reliability retransmission. If so, send it. If not, check if there is any data waiting to be sent in the high-priority time-sensitive queue. If so, send the data in that queue. Otherwise, poll and transmit non-time-sensitive data.

[0039] Furthermore, the current superframe sequence number SP is calculated based on the compensated timestamp. id ,include:

[0040]

[0041] in, Indicates rounding up, TSF cur T represents the current local timestamp of the device node. offset T represents the synchronization error currently measured using synchronization technology. sp This indicates the duration of a superframe, which is composed of the CSMA duration and the TDMA duration.

[0042] Furthermore, based on the current local time and the superframe sequence number, the slot number j in the current superframe is calculated, including:

[0043]

[0044] Among them, TSF cur T represents the current local timestamp of the device node. offset T represents the synchronization error currently measured using synchronization technology. sp T represents the duration of a superframe. s This indicates the time slot length, and % indicates the modulo operation.

[0045] The beneficial effects of this invention are as follows: This invention effectively guarantees the transmission of time-sensitive data in the network through time division and time slot retransmission mechanism; by using the queue information returned by the device nodes, the controller statistically analyzes the transmission data of each device node in the network, performs mathematical modeling, makes high-level use of network resources, minimizes the end-to-end latency of time-sensitive data, and meets the real-time and reliability requirements of time-sensitive data by optimizing the end-to-end latency. Attached Figure Description

[0046] Figure 1 This is a flowchart of the time slot allocation method of the present invention;

[0047] Figure 2 This is a flowchart of the node network access time slot allocation process of the present invention;

[0048] Figure 3 This is a schematic diagram of the data queue scheduling of the present invention;

[0049] Figure 4 This is a schematic diagram of the time slot composition structure of the present invention;

[0050] Figure 5 This is a schematic diagram illustrating the time slot calculation principle of this invention. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] A time slot allocation method for multiple service flows coexisting, such as Figure 1 As shown, it includes:

[0053] S1: The site internally maps and manages queues for the generated data based on the data's destination MAC address and priority attribute.

[0054] S2: The device node STA continuously extracts the queue status from the current data queue and reports the extracted queue status to the access node AP through feedback frames during the CSMA phase by competing for channel resources;

[0055] The queue status includes the queuing time period and data volume information of time-sensitive business data and non-time-sensitive business data with different time sensitivity and different generation periods;

[0056] The feedback frame consists of four main parts: header, feedback information, padding field and FCS field. The feedback information includes frame identification code, frame length information, device node association identification code and data queue length information.

[0057] S3: After receiving the queue information in the feedback frame, the access node (AP) performs mathematical modeling on periodic and non-periodic data services to divide them into time slots and obtain time slot allocation information.

[0058] S4: The controller of the access node AP adds the time slot allocation information to the extended field of the beacon frame periodically sent by the access node AP, obtains the new time slot allocation result, and sends it to the device node;

[0059] S5: The device node receives the new time slot allocation result issued by the controller in the access node (AP), updates the local time slot allocation table according to the new time slot allocation result, and sends the data in the time slot allocated in the updated local time slot allocation table.

[0060] Inside a spacecraft, there are data with different latency requirements that need to be transmitted; therefore, an adaptive hierarchical round-robin scheduling algorithm is used to allocate time reasonably to avoid data with different time sensitivity competing for network transmission resources equally. Figure 2 This is the node network access time slot allocation diagram of the present invention, such as... Figure 2 As shown, the specific methods for device nodes to feed back relevant data information so that the controller can make reasonable time slot allocation include:

[0061] 101. Time is divided into periodic phases, known as superframes, which include CSMA and TDMA phases. Device nodes transmit data using a hybrid CSMA / TDMA access method. Time is divided into periodic superframes, which are further subdivided into two parts: the CSMA phase and the TDMA phase. The CSMA phase marks the beginning of a superframe, responsible for providing network clock synchronization, broadcasting superframe information, terminal access, prioritizing time slot requests for the next superframe, and transmitting non-time-sensitive data; while the TDMA phase is primarily responsible for ensuring the transmission of time-sensitive data. Each phase contains at least one time slot. The TDMA and CSMA phases consist of k time slots T. s composition.

[0062] To utilize network resources more efficiently, carrier sense / collision detection (CSMA / CA) is disabled during the TDMA data transmission phase to improve network utilization. To ensure data transmission reliability, a retransmission mechanism needs to be designed for the existing scenario. For general sensor data, which is time-sensitive, secondary retransmission is usually unnecessary. However, for applications requiring high reliability, retransmission is necessary to ensure reliability. Therefore, this invention adopts an out-of-slot retransmission method based on the existing network. If a data packet fails to transmit within a time slot, the device node can be configured to retransmit the lost data packet on the next available link.

[0063] Data mapping and queue management are performed within the site. The device node (STA) collects the enqueue time period and data volume information of the current data queue and reports the queue status to the access point (AP) in the corresponding time slot.

[0064] Data generated within the site needs to be categorized. Figure 3 This is a schematic diagram of the data queue scheduling of the present invention, as shown below. Figure 3As shown, after a connection is established between a device station (STA) and an access point (AP), the data generated by the STA must first be queued in the station's internal waiting queue. In latency-sensitive wireless networks, the terminal's data buffer queue uses the existing IEEE 802.11 protocol and is managed according to service type priority. The device station manages multiple internal queues Q1, Q2, ..., Q... N Unified queue management is implemented. Hash values ​​are generated using relevant control information from data packets (destination MAC address, priority attribute), and mapped to the corresponding TID queues via a hash table. Queue scheduling and management are performed using a weighted fair queuing algorithm based on time slot allocation.

[0065] 102. Device nodes upload their queue-related information to the AP via feedback frames. The designed feedback frame consists of four main parts: header, feedback information, padding fields, and FCS field. The feedback information includes the frame identifier, frame length information, device node association identifier, and data queue length information. If a device node fails to send a feedback frame (i.e., does not receive an ACK packet from the AP on time), it will resend the feedback frame in the next feedback phase. If a time-sensitive service arrives at the device node's data queue, the node will request the corresponding time slot during the CSMA phase of the next superframe cycle.

[0066] 103. The access node (AP) receives and compiles relevant information from all nodes participating in time slot allocation, performs linear regression prediction, and calculates the total superframe length, node network entry time slot length, unit time slot length, and corresponding node transmission time slot position in the next time slot table. The access node then sends the predicted time slot allocation information to the device nodes via beacon frames.

[0067] After the controller obtains the queue information of each terminal in the current network, it performs mathematical modeling to divide the arrival time of periodic and non-periodic data services into time slots:

[0068] (1) For time-sensitive services with uncertainty, the arrival batches of services follow a Poisson process. The number of data frames arriving in the data queue within the time period t is a composite Poisson process, and its characteristic function is... as follows:

[0069]

[0070] f i C (x)=exp[exp(ixC i,j )]

[0071] in, This indicates that a Poisson distribution is used to process non-periodic data services, f i C(x) represents the number of data frames C included in the service. i,j Characteristic function, C i,j Indicates the number of data frames included in the service, δ i Let t represent the arrival rate of the event, t represent the time period for calculation, x represent the number of times business data arrives within the time period t, i represent business i, and exp represent the expected operation.

[0072] Since time slot allocation is dynamically and flexibly adjusted, based on the above modeling, it can be planned in a unified manner as a special "periodic data" that appears only once in the current superframe during each allocation.

[0073] (2) For periodic deterministic flow, where each data stream d is determined by its cycle time T c and payload P d Define, and the upper bound of the end-to-end latency requirement is U. s Assume there are N device nodes in the network, denoted as d1, d2, ..., dn. N Let each terminal d i Transmitted to terminal d n The corresponding T-time-sensitive data stream is represented as d i,n Its transmission period is denoted as T. i Each period T i The amount of data sent by the internal terminal is represented by P. i d Therefore, minimizing the end-to-end latency of time-sensitive data streams is the key problem that needs to be solved.

[0074] First determine the TDMA cycle length T. td As long as in T td If the time-sensitive data transmission within the frame meets the constraints, then there will be no problem with time-sensitive data transmission during the superframe cycle; that is, the scheduling strategy only needs to focus on one superframe cycle. Here, T... td Set to the least common multiple of the cycle times of all periodic time-sensitive data streams.

[0075] T td =[T i ,...,T N ]+k(2·|T0|+T trans +T SIFS +T ACK )

[0076] Among them, T td T represents the total superframe length. i T represents the time-sensitive data stream cycle period, T0 represents the maximum error that synchronization may produce, and T trans T represents the average propagation time between wireless links.SIFS T represents the shortest inter-frame interval. ACK This indicates the time taken to transmit the ACK, where k represents the number of time slots in the TDMA stage. N This represents the total number of nodes participating in time slot allocation.

[0077] The time slot length is:

[0078]

[0079] stβ i +...+β N =1

[0080] β i ,...β N ∈[0,1]

[0081] T p =2·|T o |+T trans

[0082] Where, β i This indicates that the device node d is assigned to it. i The weights, T p T is the length of the protection time interval in the time slot. o T represents the maximum error that synchronization might produce. trans Let T be the average propagation time between wireless links. To ensure the validity of the above formula, in this embodiment, T is... s It needs to be divisible by H, and the number of time slots in each terminal's transmission cycle must be an integer, that is:

[0083]

[0084]

[0085] Let f i,j ∈{0,1} indicates whether time slot j belongs to terminal d. i f i,j =1 indicates that time slot j belongs to terminal d i Conversely, it does not belong to terminal d. i .make To represent the transmission scheduling strategy in a TDMA superframe, the following must be satisfied:

[0086]

[0087]

[0088] Allocating radio time slot resources to all device nodes means allocating one or more radio time slots to each device node. Figure 4A schematic diagram of the time slot composition structure of the present invention is given, as follows: Figure 4 As shown, to ensure reliability, not all time slots are used for transmitting actual data. Based on the above constraints, C... i,m For each device node in the network, d i Given the number of time slots allocated within the m-th superframe, where j ranges from the number of time slots per sub-cycle; then:

[0089]

[0090]

[0091] The available transmission time slot locations for device node i are:

[0092]

[0093] Subject to

[0094] Among them, T i,p T represents the time when time-sensitive data from device node i arrives and cannot be sent, thus initiating a queuing process. i,t This represents the time slot location allocated to node i at the access point, where N represents the total number of nodes participating in the time slot allocation, and f i,j ∈{0,1} indicates whether time slot j belongs to terminal d. i f i,j =1 indicates that time slot j belongs to terminal d i Conversely, it does not belong to terminal d. i T td This indicates the total superframe length.

[0095] 104. Data is sent within the time slot allocated to the device node according to the time slot allocation result issued by the controller in the access point (AP).

[0096] Each superframe requires the distribution of time slot allocation results to ensure the lowest possible overall network latency. The time slot allocation process for the (m+1)th superframe begins with the CSMA phase of the previous (m-th) superframe, involving the collection of queue information and time slot allocation for network nodes. Once a device node receives the controller's time slot allocation result, it extracts the relevant information, updates its local time slot allocation table, and then transmits data within its assigned time slot according to the new table and the system's internal high-precision timer. Figure 5 This is a schematic diagram illustrating the principle of time slot calculation, such as... Figure 5 As shown, the superframe number SP at the current time is calculated. id for:

[0097]

[0098] in, Indicates rounding up, TSF cur T represents the current local timestamp of the device node. offset T represents the synchronization error currently measured using synchronization technology. sp This indicates the duration of a superframe, which is composed of the CSMA duration and the TDMA duration.

[0099] Then, based on the superframe sequence number S id and local time TSF cur The slot number of the current time in superframe j is calculated as follows:

[0100]

[0101] Among them, TSF cur T represents the current local timestamp of the device node. offset T represents the synchronization error currently measured using synchronization technology. sp T represents the duration of a superframe. s This indicates the time slot length, and % indicates the modulo operation.

[0102] Then, by comparing the obtained time slot allocation result table, if the current time slot j is allocated to the current node, the current node can send data; otherwise, it waits for the next time slot to arrive.

[0103] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for time slot allocation in multi-service flow coexistence, characterized by, The application comprises: S1: data mapping and queue management of generated data according to the data's own destination MAC address and priority attribute within the station; S2: the device node STA continuously extracts queue status from the current data queue and reports the extracted queue status to the access node AP through a feedback frame in the CSMA stage by competing for channel resources; The queue status includes the time-sensitive data and non-time-sensitive data of different time-sensitive degrees and different generation cycles, and the entry time period and data volume information; The feedback frame consists of four parts: header, feedback information, padding field and FCS field, and the feedback information includes frame identification code, frame length information, device node association identification code and data queue length information; S3: after the access node AP receives the queue information in the feedback frame, the periodic and non-periodic data services are mathematically modeled and time-slotted to obtain time slot allocation information; The mathematical modeling and time-slotting of the periodic and non-periodic data services comprise: The non-periodic data service is processed by Poisson distribution, which is regarded as a special periodic data that appears only once in the current superframe and periodic data for planning; The periodic data service calculates the time slot allocation information in the next time slot allocation table: total superframe length, unit time slot length and node transmission time slot position; The non-periodic data service is processed by Poisson distribution, which comprises: wherein, represents a Poisson distribution process for non-periodic data traffic, f i C (x) represents a characteristic function of the number of data frames C i,j of the traffic, f i C (x) = exp[exp(ixC i,j )], C i,j represents the number of data frames of the traffic, δ i represents the arrival rate as an event, t represents a time period of calculation, x represents the number of times of occurrence of traffic data arrival within the time period t, i represents traffic i, and exp represents an operation of taking an expectation. The total superframe length is calculated, which comprises: T td = [T i ,...,T N ]+k(2·|T0|+T trans +T SIFS +T ACK ) where T td denotes the total superframe length, T i denotes the cycle period of time-sensitive data flow, T0denotes the maximum error that can be generated by synchronization, T trans denotes the average propagation time between wireless links, T SIFS denotes the minimum interframe space, T ACK denotes the time spent on transmitting ACK, k denotes the number of time slots in the TDMA stage, N denotes the number of all nodes participating in time slot allocation; The unit time slot length is calculated, which comprises: subject to β i +...+β N =1 β i ,...β N ∈[0,1] where T s denotes the unit time slot length, β i denotes the weight assigned to the device node d i , T p denotes the guard time interval length in the time slot, denotes the data volume of the device node service flow, and v denotes the data transmission rate. The node transmission time slot position is calculated, which comprises: subject to where T i,p denotes the time when the time-sensitive data of device node i arrives at the start of the queue that cannot be transmitted, i,t denotes the time slot position allocated by the access point for node i, f i,j ∈{0, 1} denotes whether time slot j belongs to terminal d i , f i,j = 1 denotes that time slot j belongs to terminal d i , and vice versa i ; S4: the controller of the access node AP adds the time slot allocation information to the extension field of the beacon frame periodically sent by the access node AP to obtain a new time slot allocation result and send it to the device node; S5: the device node receives the new time slot allocation result issued by the controller in the access node AP, updates the local time slot allocation table according to the new time slot allocation result, and sends the data according to the time slot allocated in the updated local time slot allocation table.

2. The method of claim 1, wherein, The S1 specifically comprises: The destination MAC address and priority attribute of the generated data within the station are extracted, a hash value is generated, and the wireless data frame is stored in the corresponding buffer queue according to the generated hash value: if the hash value of the wireless data frame is a specific numerical set, the data packet is time-sensitive data, and then it is stored in the time-sensitive list; otherwise, the data packet is non-time-sensitive data, and then it is stored in the non-time-sensitive list and waits to be sent.

3. The method of claim 1, wherein, The S5 specifically comprises: Step 1: the device node obtains the new time slot allocation result in the beacon frame in the CSMA stage and extracts the allocation information, wherein the time-sensitive wireless node in the device node reads the identity corresponding to the current time slot number in the time slot table at the start time of each time slot to determine whether the current time slot belongs to the wireless node; Step 2: After compensating the time synchronization error deviation according to the allocation information, the superframe sequence number SP currently in is calculated according to the compensated time stamp id ; Step 3: the time slot number j in the current superframe is calculated according to the current local time and the superframe sequence number; Step 4: According to the calculated time slot number, compare the new time slot allocation result obtained in step 1 to determine whether the current time slot belongs to the node. If it belongs to the node, extract the data waiting to be sent through the queue management module, otherwise do not send data and continue to wait for the time slot belonging to the node; Step 5: Check if there is retransmission data with high reliability that needs to be sent. If there is, send it. If not, check if there is data in the high priority time-sensitive queue waiting to be sent. If there is, send the data in the queue. Otherwise, poll the non-time-sensitive data for transmission.

4. The method of claim 3, wherein, calculating a super frame number SP currently in according to the compensated time stamp id comprising: wherein, represents the up-rounding, TSF cur represents the current local time stamp of the device node, T offset represents the current synchronization error measured by the synchronization technique, T sp represents the time length of the superframe, which is composed of the CSMA time length and the TDMA time length.

5. The method of claim 3, wherein the time slot allocation method is characterized by, According to the current local time and the superframe sequence number, the time slot number j in the current superframe is calculated, including: where TSF cur represents the current local time stamp of the device node, T offset represents the current synchronization error measured by the synchronization technique, T sp represents the time length of the superframe, T s represents the time slot length, % represents the modulo operation.

Images