Network traffic scheduling method and device based on multiple links, equipment and storage medium

Abstract

The invention discloses a multi-link-based network traffic scheduling method and device, equipment and a storage medium. The method comprises the following steps: receiving data packets sent by a plurality of terminals, and determining a user role corresponding to each terminal; for any data packet, performing deep packet detection on the data packet sent by the terminal to obtain application information of the data packet, and dynamically determining the priority of the data packet based on the application information, the corresponding user role and the network state information; and performing MLO link selection and wireless network resource selection for each data packet based on the priority of each data packet, and dynamically adjusting the distribution of the MLO link and the wireless network resource based on the feedback information of each current terminal. Therefore, the problems of single link selection strategy, lack of dynamic adaptive capacity and the like in the prior art are solved, the MLO links and the wireless network resources are dynamically allocated, the allocation precision, flexibility and reliability of the MLO links and the wireless network resources are improved, and the utilization rate of the network resources is improved through the DPI technology and multi-link dynamic scheduling.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a multi-link-based network traffic scheduling method, device, equipment and storage medium. Background Art

[0002] With the popularity of wireless local area networks (WLANs), Wi-Fi has become the main means of supporting various network applications. However, different applications have very different requirements for network performance. For example, video conferencing and online games require low latency, while video streaming and file downloading focus more on bandwidth requirements.

[0003] And in WiFi7 technology, MLO technology is proposed. MLO (Multi-Link Operation) is a key technical feature in Wi-Fi 7. It allows devices to use multiple wireless bands for communication at the same time, thereby improving the efficiency, reliability and throughput of the network.

[0004] However, although MLO allows devices to communicate simultaneously on multiple frequency bands, the current link selection strategy may rely too much on pre-set rules or algorithms, which may result in insufficient consideration of real-time network conditions or the needs of specific applications in some cases, resulting in a single link selection strategy for multi-link operations in terms of traffic management and link selection. In addition, MLO technology is designed with little consideration for the specific needs of upper-layer applications, which means that even if there are multiple types of traffic (such as video streaming, voice calls, file downloads, etc.), MLO may treat them equally, rather than allocating resources based on the characteristics of the application, and therefore lacks application awareness. Summary of the invention

[0005] The main purpose of this application is to provide a multi-link-based network traffic scheduling method, device, equipment and storage medium to achieve dynamic allocation of MLO links and wireless network resources, and improve the allocation accuracy, flexibility and reliability of MLO links and wireless network resources.

[0006] To achieve the above objectives, the present application provides a multi-link based network traffic scheduling method, comprising:

[0007] Receiving data packets sent by multiple terminals, and determining a user role corresponding to each of the terminals;

[0008] For any data packet, perform deep packet inspection on the data packet sent by the terminal to obtain application information of the data packet, and dynamically determine the priority of the data packet based on the application information, the corresponding user role and the network status information;

[0009] MLO link selection and wireless network resource allocation are performed for each data packet based on the priority of each data packet, and the MLO link selection and wireless network resource allocation are dynamically adjusted based on the current feedback information of each terminal.

[0010] Optionally, the application information includes application type and application traffic characteristics; performing deep packet inspection on the data packet sent by the terminal to obtain application information of the data packet includes: performing feature extraction on the data packet to obtain the protocol type, port number and load characteristics of the data packet; determining the application type and application traffic characteristics corresponding to the data packet based on the protocol type, the port number and the load characteristics.

[0011] Optionally, dynamically determining the priority of the data packet based on the application information, the corresponding user role and the network status information includes: determining an initial priority of the data packet based on the application type of the data packet, application traffic characteristics and the corresponding user role; dynamically adjusting the initial priority of the data packet according to a preset strategy and based on the current network status information to obtain the priority of the data packet; wherein the preset strategy includes lowering the priority of the data packet whose user role is an ordinary user when network congestion is determined based on the network status information.

[0012] Optionally, the performing MLO link selection and allocating wireless network resources for each of the data packets based on the priority of each of the data packets includes: obtaining current status information of each link; performing MLO link selection for each of the data packets based on the priority of each of the data packets and the current status information of each link; allocating each of the data packets to a preset WMM queue based on the priority of each of the data packets, and allocating the wireless network resources for each of the data packets based on the priority of each of the data packets and the corresponding terminal type.

[0013] Optionally, obtaining the current status information of each link includes: using Link Telemetry to obtain the status information of each link; the status information of each link includes at least bandwidth utilization, packet loss rate, delay, signal strength, link quality index and SLA priority parameter.

[0014] Optionally, the wireless network resources include at least time resources, frequency resources and space resources; allocating the wireless network resources to each data packet based on the priority of each data packet and the corresponding terminal type includes: selecting a corresponding transmission mode for each data packet based on the corresponding terminal type of each data packet; adjusting the time resources, frequency resources and space resources allocated to each data packet based on the priority of each data packet, current status information of each link and user needs.

[0015] Optionally, the feedback information includes at least delay and packet loss rate; the dynamically adjusting the MLO link selection and the allocation of the wireless network resources based on the current feedback information of each of the terminals includes: dynamically adjusting the MLO link selection and allocated wireless network resources for the data packets corresponding to each of the terminals based on the delay and packet loss rate of each of the terminals.

[0016] In addition, to achieve the above-mentioned purpose, the present application also provides a multi-link based network traffic scheduling device, including: an acquisition module, used to receive data packets sent by multiple terminals and determine the user role corresponding to each of the terminals; a priority determination module, used to perform deep packet inspection on the data packets sent by the terminal for any data packet, obtain the application information of the data packet, and dynamically determine the priority of the data packet based on the application information, the corresponding user role and the network status information; a resource allocation module, used to perform MLO link selection and allocate wireless network resources for each of the data packets based on the priority of each of the data packets, and dynamically adjust the MLO link selection and the allocation of the wireless network resources based on the feedback information of each of the current terminals.

[0017] The present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the multi-link-based network traffic scheduling method as described above is implemented.

[0018] The present application also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the multi-link-based network traffic scheduling method as described in any one of the above is implemented.

[0019] The multi-link-based network traffic scheduling method of the present application identifies the application of the data packets sent by each terminal by using deep packet inspection technology, and further dynamically sets the priority of each data packet according to the application information corresponding to each data packet, the user role corresponding to each terminal and the current network status information, and finally performs MLO link selection and dynamic allocation of wireless network resources according to the priority of each data packet, so that the allocation process takes into account the influence of applications, users and network status on resource allocation, solves the problems of single link selection strategy and lack of dynamic adaptability in the prior art, realizes priority scheduling of data packets for key users, and improves the accuracy, flexibility and reliability of network traffic scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a scenario example of a multi-link-based network traffic scheduling method according to an embodiment of the present application;

[0021] Figure 2It is one of the flow charts of the multi-link-based network traffic scheduling method according to an embodiment of the present application;

[0022] Figure 3 This is the second flowchart of the multi-link-based network traffic scheduling method according to an embodiment of the present application;

[0023] Figure 4 This is the third flowchart of the multi-link-based network traffic scheduling method according to an embodiment of the present application;

[0024] Figure 5 This is the fourth flowchart of the multi-link-based network traffic scheduling method according to an embodiment of the present application;

[0025] Figure 6 is a data packet scheduling diagram of a specific example of the present application;

[0026] Figure 7 is a schematic diagram of a multi-link-based network traffic scheduling device according to an embodiment of the present application;

[0027] Figure 8 An example of a physical structure diagram of an electronic device is provided;

[0028] In the figure, 110 is a wireless access point; 120 is a terminal device; 700 is a multi-link-based network traffic scheduling device; 710 is an acquisition module; 720 is a priority determination module; 730 is a resource allocation module; 810 is a processor; 820 is a communication interface; 830 is a memory; and 840 is a communication bus.

[0029] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0031] With the rapid popularization of wireless local area networks (WLANs), Wi-Fi has become the main means of supporting various network applications. From daily home entertainment to enterprise-level communications, the wide application of Wi-Fi in different scenarios has put forward higher requirements for its performance. However, different applications have very different requirements for network performance. For example, video conferencing and online games require extremely low latency to ensure real-time interactive experience; while video streaming and file downloading pay more attention to bandwidth requirements to ensure smooth content transmission.

[0032] In order to meet these diverse needs, the IEEE 802.11ax standard (i.e. Wi-Fi 6) introduces two key technologies - OFDMA (Orthogonal Frequency Division Multiple Access) and MU-MIMO (Multi-User Multiple Input Multiple Output) to improve the overall performance of the network. Among them, OFDMA technology divides the available spectrum into multiple subcarriers and allows multiple users to use these subcarriers for communication at the same time, effectively reducing latency and improving spectrum efficiency. MU-MIMO technology uses multiple antenna arrays to create independent data channels, supports high-throughput transmission, and significantly improves the system's capacity and service quality.

[0033] Although Wi-Fi 6 has greatly improved the user experience, it still has certain limitations when facing more complex application scenarios. Therefore, MLO (Multi-Link Operation) technology is proposed in Wi-Fi 7 technology.

[0034] MLO technology allows a single Wi-Fi device to establish connections with access points (APs) on multiple frequency bands at the same time, including but not limited to 2.4GHz, 5GHz, and 6GHz. MLO technology can also effectively utilize the bandwidth of each frequency band by transmitting data in parallel on different frequency bands, thereby significantly improving the overall data transmission rate. MLO technology can also achieve load balancing between different frequency bands, reduce congestion in a specific frequency band, and improve the stability of the entire network. Based on the above advantages of MLO technology, functions such as multi-link aggregation, intelligent scheduling, and seamless switching can be realized.

[0035] However, in the existing multi-link operation technology, the following problems still exist in terms of traffic management and link selection:

[0036] 1. Single link selection strategy: Currently, multi-link selection is mostly based on static parameters and lacks a dynamic adaptation mechanism.

[0037] 2. Lack of application awareness: Unable to accurately allocate links based on the traffic characteristics of different applications (such as real-time and bandwidth requirements).

[0038] Based on this, the embodiments of the present application provide a multi-link-based network traffic scheduling method, device, equipment and storage medium. By combining the MLO characteristics of WiFi 7, and combining multiple factors such as application type, application characteristics, user role, current network status, etc., MLO links and wireless network resources are allocated to the data packets of each terminal, and dynamic adjustments are made to improve service quality, optimize user experience, and enhance the allocation accuracy, flexibility and reliability of MLO links and wireless network resources.

[0039] For ease of understanding, this specification provides a scenario example of a multi-link network traffic scheduling method. The scenario example is applied in Figure 1 In the application environment shown, the scenario may include a wireless access point 110 (AP) and a variety of terminal devices 120 , and each terminal device 120 is connected to the wireless access point 110 .

[0040] It should be noted that the terminal device 120 may include (but is not limited to): mobile terminals such as mobile phones, laptop computers, PADs (tablet computers), etc., and fixed terminals such as digital TVs, desktop computers, etc.

[0041] In this scenario example, each terminal device 120 can respectively conduct activities such as video conferencing, file downloading, web browsing, etc. When each terminal device 120 conducts the above activities, each terminal device 120 needs corresponding wireless network resources. At this time, after receiving the data packets sent by each terminal device 120, the wireless access point 110 can obtain the user role corresponding to each terminal device 120, and perform deep packet inspection (DPI) on the data packets of each terminal device 120 to obtain the application information corresponding to each data packet. The wireless access point 110 can determine the priority of processing and resource scheduling of each data packet based on different user roles, application information and current network status information, so as to schedule MLO links and wireless network resources for each terminal device 120 according to the priority; at the same time, the wireless access point 110 obtains the feedback information provided by each terminal device 120 in real time, so as to dynamically adjust the MLO link selection and wireless network resource scheduling according to the feedback information of each terminal device 120.

[0042] Therefore, when performing MLO link selection and wireless network resource scheduling, not only a single static factor is considered, but also factors such as application characteristics, user roles, network status, etc., and the MLO link selection and wireless network resource allocation strategy are dynamically adjusted according to the feedback from each terminal device 120, thereby improving the user experience, as well as improving the allocation accuracy, flexibility and reliability of MLO links and wireless network resources.

[0043] With reference to the scenario example of the multi-link network traffic scheduling method in the aforementioned embodiment, the multi-link network traffic scheduling method in the embodiment of the present application is introduced in detail below.

[0044] Figure 2 is one of the flow charts of the multi-link based network traffic scheduling method of the embodiment of the present application. The multi-link based network traffic scheduling method can be executed by the above-mentioned wireless access point, such as Figure 2 As shown, the multi-link based network traffic scheduling method may include the following steps:

[0045] Step 210: Receive data packets sent by multiple terminals, and determine the user role corresponding to each terminal.

[0046] It should be noted that the terminal in the embodiment of the present application may be a terminal device (such as a mobile phone, a computer, etc.) or a terminal application, and the terminal is not specifically limited here.

[0047] In actual application, when each terminal is connected to a wireless access point and performs network activities (such as video conferencing, gaming, file downloading, etc.), the wireless access point receives data packets sent by each terminal through the antenna and schedules these data packets to meet the network needs of each terminal.

[0048] After the wireless access point receives the data packets sent by each terminal, the wireless access point can first distinguish the users of each terminal and determine the user roles corresponding to each terminal. It should be noted that the user roles in this embodiment may include (but are not limited to) key users and ordinary users, and may also include student users and parent users, etc. This embodiment distinguishes the user roles corresponding to each terminal so that MLO link selection and wireless network resource allocation can be performed based on the user roles in the future, so as to ensure that each time resource allocation is performed, resource allocation can be prioritized for key users, and better slice resources can be allocated to key users.

[0049] Specifically, the wireless access point can determine the user role of each terminal by: each data packet contains a source MAC address, which is identity information that uniquely identifies the terminal (such as user name, department, whether it is a VIP user, etc.). Therefore, the user role of the corresponding terminal can be determined by the source MAC address of the data packet.

[0050] Step 220: For any data packet, perform deep packet inspection on the data packet sent by the terminal to obtain application information of the data packet, and dynamically determine the priority of the data packet based on the application information, the corresponding user role and the network status information.

[0051] After the wireless access point receives each data packet and determines the user role of the terminal corresponding to each data packet, the wireless access point can determine the priority of each data packet. The higher the priority of the data packet, the higher the requirements for delay, bandwidth, etc. for the processing of the data packet, and the more important the corresponding user role, at this time, the wireless access point will allocate a better MLO link and wireless network resources to the data packet; conversely, the lower the priority of the data packet, the lower the requirements for delay, bandwidth, etc. for the processing of the data packet, or the corresponding user role is not important, at this time, the wireless access point will allocate fewer MLO links and wireless network resources to the data packet.

[0052] It is understandable that different applications have different traffic characteristics (such as real-time and bandwidth requirements). Therefore, different MLO links and wireless network resources can be allocated to different applications, thereby improving the accuracy and flexibility of resource allocation while ensuring the quality of service (QoS).

[0053] For example, for commercial or industrial applications, these applications usually require high bandwidth, so the priority of these applications can be set to high priority; for applications such as video conferencing, low latency is usually required, so the priority of these applications can be set to medium priority; for applications such as streaming services, virtual reality, games, and training, these applications usually require high bandwidth and low latency, so the priority of these applications can be set to high priority; for IoT (Internet of Things) related applications, these applications usually need to optimize energy consumption and transmission efficiency, so the priority of these applications can be set to medium priority; for applications such as video downloading and file transfer, these applications often do not require low latency and very high bandwidth requirements, so the priority of these applications can be set to low priority.

[0054] Therefore, different priorities are set for different application types according to the traffic characteristics of the applications, so that MLO links and wireless network resources can be dynamically allocated according to the priorities. This can not only meet the traffic requirements of most applications, but also provide protection for high-priority low-latency or high-throughput applications and ensure the quality of service (QoS).

[0055] In some embodiments, the application information includes application type and application traffic characteristics; Figure 3 This is the second flowchart of the multi-link network traffic scheduling method according to an embodiment of the present application. Figure 3 As shown, in step 220, deep packet inspection is performed on the data packet sent by the terminal to obtain application information of the data packet, which may include the following steps:

[0056] Step 310: Extract features from the data packet to obtain the protocol type, port number, and payload features of the data packet.

[0057] Step 320: Determine the application type and application traffic characteristics corresponding to the data packet based on the protocol type, port number, and payload characteristics.

[0058] It should be noted that application types may include (but not limited to) commercial applications, industrial applications, streaming services, virtual reality, games, training, IoT device communications and downloads, etc.; application traffic characteristics may include (but not limited to) low latency, high bandwidth, high transmission efficiency, etc., and application types and application traffic characteristics are not specifically limited here. Different application types correspond to different application traffic characteristics, and the application traffic characteristics corresponding to each application type can be determined based on theoretical knowledge or experience.

[0059] In this embodiment, deep packet inspection (DPI) technology can be used to implement analysis and application identification of each data packet. It should be noted that DPI is an advanced network traffic monitoring and control technology that can work in the application layer of the open system interconnection (OSI) reference model. Different from the traditional shallow packet detection that only checks the packet header information (such as source IP address, destination IP address, port number, etc.), DPI can go deep into the payload part of the data packet, that is, the data content of the application layer, for detailed analysis and processing.

[0060] Taking a data packet as an example, specifically, this embodiment uses DPI technology to analyze the characteristic information of the data packet and extract the protocol type, port number and load characteristics of the data packet. Among them, the protocol type may include the currently existing communication protocol types, such as TCP protocol or UDP protocol; the load characteristics may include string characteristics, state machine characteristics, etc. As an example, for a TCP data packet, the possible application type can be preliminarily determined by checking its destination port. As another example, many applications send specific strings as identifiers during the communication process; for example, the handshake process of the BitTorrent protocol contains a fixed string such as "BitTorrent protocol". By matching these characteristic strings, the corresponding application type can be identified with high precision.

[0061] In this embodiment, the feature extraction method can be: header information analysis, DPI usually extracts basic information such as source / destination IP address, port number, protocol type, etc. from the IP data packet header. This information can be used to preliminarily filter and classify traffic. The feature extraction method can also be: payload content analysis, for non-encrypted traffic, DPI can directly parse the payload part of the data packet to find specific application layer protocol features, such as HTTP request line, DNS query string, etc., which helps to identify specific applications and services.

[0062] The classification models used can be: support vector machine (SVM), random forests, neural networks, etc. When training the model, a labeled sample data set is used so that it can learn to distinguish different types of traffic. The advantage of this method is that it can automatically adapt to emerging applications and has a high accuracy rate. The classification models used can also be: convolutional neural network (CNN), recurrent neural network (RNN) and its variant long short-term memory network (LSTM) to handle complex traffic patterns. This type of model is particularly suitable for processing long-term dependencies or scenarios that require understanding of context, such as video streaming, voice communication, etc.

[0063] Since there is a corresponding relationship between application type and application traffic characteristics, after using DPI technology to obtain the protocol type, port number and load characteristics of the data packet, and identifying the application type of the data packet based on the protocol type, port number and load characteristics of the data packet, the application traffic characteristics corresponding to the data packet can be directly obtained.

[0064] Figure 4 This is the third flow chart of the multi-link-based network traffic scheduling method of the embodiment of the present application. Figure 4 As shown, in some implementations, dynamically determining the priority of the data packet based on the application information, the corresponding user role, and the network status information in step 220 may include the following steps:

[0065] Step 410: Determine the initial priority of the data packet based on the application type of the data packet, the application traffic characteristics, and the corresponding user role.

[0066] Step 420: Dynamically adjust the initial priority of the data packet according to the preset strategy and based on the current network status information to obtain the priority of the data packet; wherein the preset strategy includes lowering the priority of the data packet whose user role is ordinary user when the network is congested based on the network status information.

[0067] It is understandable that different MLO links and wireless network resources can be allocated for different application types and application traffic characteristics. For example, a low-latency, high-stability MLO link can be allocated to a data packet of the application type video conferencing, thereby ensuring that the network is stable, unobstructed, and has low latency when the user is conducting a video conference. For another example, a low-energy transmission mode can be allocated to IoT devices to meet the requirements of device operating time control, cost control, and application scenarios. In addition, different MLO links and wireless network resources can also be allocated to different user roles. For key users, high-bandwidth links can be allocated to the corresponding data packets first.

[0068] Based on this, the embodiment of the present application jointly determines the initial priority of the data packet according to the application type, application traffic characteristics and user role corresponding to the data packet. Specifically, for application types (such as real-time communication) with high requirements for application traffic characteristics (such as low latency and high bandwidth), the priority of the data packet of this application type can be set to high priority; for application types (such as media streaming) with relatively low requirements for application traffic characteristics, the priority of the data packet of this application type can be set to medium priority; for application types (such as file download) with relatively low requirements for application traffic characteristics, the priority of the data packet of this application type can be set to low priority.

[0069] Furthermore, for a data packet corresponding to a user role of a key user, if judged only based on application traffic characteristics, the priority of the data packet is low priority, but because the user role is a key user, the priority of the data packet can be increased to medium priority.

[0070] As an example, originally for the file download application type, the priority of its related data packets should be low priority, but if the terminal with the user role as a key user sends a data packet related to the file download application to the wireless access point, the priority of the data packet is medium priority.

[0071] In this embodiment, after the initial priority is determined for each data packet, the priority of each data packet can be dynamically adjusted according to the real-time network status to ensure that the processing of data packets with higher traffic requirements is not affected, thereby ensuring the service quality of key users or key applications.

[0072] Specifically, the wireless access point can monitor the current network status in real time through the built-in management and monitoring functions to obtain the current network status information. Further, when the wireless access point determines that the current network status is good according to the current network status information, the subsequent MLO link selection and wireless network resource allocation can be performed according to the initial priority of each data packet. When the wireless access point determines that the current network status is congested according to the current network status information, the priority of the corresponding data packet of the ordinary user can be reduced.

[0073] For example, if the wireless access point determines that the current network state is congested according to the current network state information, the high-priority data corresponding to the ordinary user may be processed with a lower priority, so that the priority of the data packet is reduced to a low priority.

[0074] Therefore, the priority of each data packet can be dynamically adjusted according to the current network status to ensure the service quality of key users or key applications.

[0075] Step 230: Perform MLO link selection and allocate wireless network resources for each data packet based on the priority of each data packet, and dynamically adjust the MLO link selection and allocation of wireless network resources based on the feedback information of each terminal.

[0076] After the wireless access point determines the priority of each data packet, it can select the MLO link and allocate wireless network resources according to the priority of each data packet. After allocation, the wireless access point can also dynamically adjust the MLO link selection and wireless network resource allocation based on the feedback information of each terminal. For example, when the wireless access point detects that a user is watching a video on a terminal device and the video quality decreases, the wireless access point can increase the space resources and frequency resources allocated to the terminal to ensure that it obtains a more stable connection.

[0077] Figure 5 This is a fourth flow chart of the multi-link-based network traffic scheduling method according to an embodiment of the present application. Figure 5 As shown, in some implementations, performing MLO link selection and allocating wireless network resources for each data packet based on the priority of each data packet in step 230 may include the following steps:

[0078] Step 510: Obtain the current status information of each link.

[0079] Step 520: Based on the priority of each data packet and the current status information of each link, perform MLO link selection for each data packet.

[0080] Step 530: Allocate each data packet to each preset WMM queue based on the priority of each data packet, and allocate wireless network resources to each data packet based on the priority of each data packet and the corresponding terminal type.

[0081] Specifically, the wireless access point can monitor the status information of each link (eg, 6 GHz, 5 GHz, 2.4 GHz) in real time, so as to determine which links are better links according to the status information of each link, thereby allocating the better links to data packets with higher priority.

[0082] In some implementations, Link Telemetry can be used to obtain the status information of each link; the status information of each link includes at least broadband utilization, packet loss rate, delay, signal strength, link quality index and SLA (Service Level Agreement) priority parameter. Broadband utilization, packet loss rate, delay, signal strength, link quality index and SLA priority parameter can reflect the status of the link; for example, if the broadband utilization of the link is high, excessive broadband utilization may cause network congestion, reduce data transmission speed and even cause packet loss, so it can be determined that this link is not a better link; for another example, if the packet loss rate of the link is high, the high packet loss rate will cause problems such as slow application response and reduced video call quality, and in severe cases, it will cause connection interruption, so it can also be determined that this link is not a better link; for another example, if the signal strength of the link is weak, the weak signal may cause unstable data transmission, increase the number of retransmissions, and then increase the delay and packet loss rate. In extreme cases, if the signal is too weak, an effective connection cannot be established at all, so the link is not a better link.

[0083] It should be noted that link quality indicators may include jitter, bit error rate, signal-to-noise ratio, availability, etc. SLA priority parameters may include response time, resolution time, bandwidth guarantee, priority scheduling, availability guarantee, etc.

[0084] After the wireless access point determines the status of each link, it can match the corresponding link for each data packet according to the priority of each data packet and the current status of each link. For example, a high-priority data packet can be matched with the link with the best status, and a low-priority data packet can be matched with a link with a poor status.

[0085] As an example, if the links include 6GHz, 5GHz, and 2.4GHz, and a terminal with a user role of a key user sends a data packet related to file downloading, because its user role is a key user, the priority of the data packet can be set to medium priority. The originally selected link is 5GHz, but because the link status of 5GHz is not good, the 6GHz link can be allocated to the data packet of the key user at this time.

[0086] As another example, if the links include 6GHz, 5GHz, and 2.4GHz, and a terminal with a user role of an ordinary user sends a data packet related to a video conference, since the user role is an ordinary user, the priority of the data packet can be set to medium priority, and the originally selected link is 5GHz, but since the link status of 5GHz is not good, the 2.4GHz link can be allocated to the data packet of the ordinary user at this time.

[0087] In this embodiment, the MLO multi-link modes may include: same-frequency band MLO mode, different-frequency band MLO mode, link management mode, load balancing mode and redundant transmission mode, and these modes can be selected according to actual needs.

[0088] Therefore, by combining DPI technology with MLO technology and monitoring the link status in real time, the MLO link selection of each data packet can be dynamically adjusted and optimized, thereby improving the service quality in a multi-link environment.

[0089] Furthermore, after the wireless access point allocates the MLO link for each data packet, each data packet can be allocated to a different preset WMM queue according to the priority. It should be noted that the preset WMM queue may include (but not limited to) Background queue, BE (Best Effort) queue, Video queue and Voice queue.

[0090] As an example, according to the priority, the data stream of the video conference can be mapped to the Video queue or the Voice queue in the preset WMM queue, while the data streams of file downloading and web browsing can be mapped to the BE queue and the Background queue respectively.

[0091] After allocating each data packet to different preset WMM queues, the data packets in each queue can be prioritized. Each queue includes a LO (low priority) send queue and a HI (high priority) send queue, and the data packets can be allocated to different send queues according to the priority of each data packet.

[0092] Furthermore, the corresponding transmission mode can be selected for each data packet according to the traffic demand, priority and corresponding terminal type of each data packet, and time resources, frequency resources and space resources can be dynamically allocated to ensure the service level agreement (SLA) for each high-priority data packet.

[0093] In some embodiments, wireless network resources include at least time resources, frequency resources, and space resources; allocating wireless network resources to each data packet based on the priority of each data packet and the corresponding terminal type may include: selecting a corresponding transmission mode for each data packet based on the corresponding terminal type of each data packet; adjusting the time resources, frequency resources, and space resources allocated to each data packet based on the priority of each data packet, the current status information of each link, and user needs.

[0094] In this embodiment, the transmission mode (i.e., air interface slicing technology) may include (but not limited to): OFDMA mode, which is suitable for the low-latency requirements of multi-user concurrency; MU-MIMO mode, which is suitable for high-bandwidth scenarios; SU-MIMO mode, which is suitable for single-user high-performance transmission.

[0095] Specifically, air interface slicing technology is a series of key technologies introduced in the field of wireless communications to improve spectrum efficiency, enhance multi-user support capabilities and optimize resource allocation. ATF (Admission Traffic Filtering), OFDMA (Orthogonal Frequency Division Multiple Access), TWT (Target Wake Time) and MU-MIMO (Multi-User Multiple-Input Multiple-Output) are four important mechanisms, each of which has different working principles and technical characteristics.

[0096] ATF is a traffic management mechanism that is mainly used to control and filter data flows entering the network. It can determine which data flows are allowed to access the network, ensuring that only data flows that meet specific conditions or priorities can obtain services. It can also limit the rate and adjust the delay of the passing data flows to ensure QoS (Quality of Service). It can also reserve a certain amount of bandwidth and other resources for critical applications to ensure that their performance is not affected by other low-priority traffic. It is suitable for environments that require strict control and management of network traffic, such as enterprise-level Wi-Fi networks, Internet of Things (IoT) platforms, etc.

[0097] OFDMA is an orthogonal frequency division multiple access technology, which divides the available spectrum into multiple subcarriers and allows multiple users to communicate using these subcarriers at the same time. It can realize the ability of multiple users to transmit concurrently in the same time slot by allocating different subcarriers to different users. The number of subcarriers can also be dynamically allocated according to the actual needs of each user, thereby improving spectrum utilization. Since the subcarriers maintain an orthogonal relationship, mutual interference between users is reduced. It is widely used in 4G LTE, 5G NR and Wi-Fi 6 / 6E standards, and is particularly suitable for efficient communication in high-density user scenarios.

[0098] TWT is an optimization protocol for the sleep cycle of terminal devices, which aims to extend battery life without affecting user experience. It allows the terminal to negotiate with the AP to determine the specific time of the next wake-up, avoiding unnecessary frequent wake-ups. Between two wake-ups, the terminal can be completely powered off, greatly reducing power consumption. It can also reduce the possibility of competing for channels at the same time and improve transmission efficiency by coordinating the wake-up times of different terminals. It is very suitable for mobile devices, smart home sensors and other applications that rely on battery power and have infrequent data transmission.

[0099] MU-MIMO is a multi-user multiple-input multiple-output technology that enables an AP to send independent data streams to multiple terminals at the same time. It can use multiple antenna arrays to create multiple parallel data channels in the spatial dimension, increasing the total throughput. It is particularly suitable for large public places, offices and other environments that need to support a large number of users at the same time, such as airports, conference centers, stadiums, etc.

[0100] Therefore, the appropriate transmission mode can be selected for the data packet based on the terminal type (IE characteristics). For example, for laptops participating in a Zoom video conference, if they support MU-MIMO, the AP will use multiple antenna arrays to create multiple parallel data channels in the spatial dimension, so that independent data streams can be sent to multiple users at the same time, thereby improving the overall throughput and service quality.

[0101] Furthermore, the time resources, frequency resources and space resources allocated to each data packet can be adjusted based on the priority of each data packet, the current status information of each link and user needs, ensuring that high-priority data packets are scheduled first. By combining air interface slicing technology with priority and using intelligent algorithms to achieve refined resource allocation for low latency and high bandwidth requirements, resource utilization is optimized.

[0102] In some embodiments, the feedback information includes at least delay and packet loss rate; dynamically adjusting the MLO link selection and allocation of wireless network resources based on the current feedback information of each terminal may include: dynamically adjusting the MLO link selection and allocated wireless network resources for the data packets corresponding to each terminal based on the delay and packet loss rate of each terminal.

[0103] In this embodiment, the selection of MLO links and the allocation of wireless network resources can also be dynamically adjusted based on the feedback of the application traffic performance of each terminal (such as delay, packet loss rate). Specifically, the wireless access point can continuously monitor the link status and the actual needs of each user, and dynamically adjust the resource allocation strategy based on this information. For example, when the AP detects that the download speed of a terminal has slowed down, but the download priority is low, no special adjustment can be made temporarily to continue to ensure the scheduling of high-priority data packets.

[0104] The following further introduces the multi-link-based network traffic scheduling method of an embodiment of the present application through a specific example and a specific application scenario.

[0105] Figure 6 is a data packet scheduling diagram of a specific example of this application. Figure 6 As shown, in this example, user roles are divided into ordinary users / guests and VIP users, priorities are divided into low priority, medium priority and high priority, MLO links include 2.4GHz, 5GHz, and 6GHz, and preset WMM queues include Background queue, BE (Best Effort) queue, Video queue and Voice queue.

[0106] This example is an enterprise scenario. Assume that in an enterprise environment, multiple laptops and other wireless devices are connected to the same wireless access point (AP). Employees are using Zoom for video conferencing, while other devices are also downloading files, browsing the web, and other activities. In addition, the user who is conducting the video conference has a VIP user role, and the rest are ordinary users.

[0107] The wireless access point receives data packets sent by each device. Since the ongoing Zoom video conference requires low latency and stable data transmission, and the user role is a VIP user, the priority of the data packets corresponding to the Zoom video conference is set to high priority; file downloads require a large bandwidth but do not require high latency, so the priority of the data packets corresponding to file downloads is set to medium priority; web browsing requires less traffic and has lower requirements for latency and bandwidth, so the priority of the data packets corresponding to web browsing is set to low priority.

[0108] Furthermore, based on the priority of each data packet and the current status information of each link, a 6 GHz link is allocated to the data packet corresponding to the Zoom video conference, a 5 GHz link is allocated to the data packet corresponding to the file download, and a 2.4 GHz link is allocated to the data packet corresponding to the web browsing.

[0109] Furthermore, according to the priority of each data packet, the Zoom data stream is mapped to the Video queue or Voice queue in the WMM queue, while the data streams for file downloading and web browsing are mapped to the BE queue and Background queue respectively.

[0110] In this example, for devices that concurrently request connections, such as smart watches or sensors that are browsing the web, the AP uses OFDMA to allocate smaller resource units (RUs) to these devices, allowing them to share the spectrum in the same time slot and improve spectrum utilization. For laptops participating in Zoom video conferences, if they support MU-MIMO, the AP will use multiple antenna arrays to create multiple parallel data channels in the spatial dimension, allowing independent data streams to be sent to multiple users at the same time, thereby improving total throughput and service quality.

[0111] In this example, the AP continuously monitors the link status and actual needs of each user, and dynamically adjusts the resource allocation strategy based on this information. For example, when it detects that the video quality of a participant has degraded, the AP will immediately increase the spatial or frequency resources of the user to ensure a more stable connection.

[0112] In every scheduling decision, the AP always gives priority to high-priority applications (such as Zoom). This means that even if the network is heavily loaded, the system will try its best to ensure that Zoom's data stream is processed first and maintain the quality of the meeting.

[0113] Therefore, by combining the MLO feature of WiFi 7, link performance evaluation is combined with dynamic scheduling to improve the reliability and flexibility of the system. By combining DPI, user roles and application characteristics, service level assurance is enhanced and user experience is optimized. By introducing a performance feedback loop, it is ensured that the scheduling algorithm can be dynamically adjusted according to the actual operating status to improve network efficiency. Multi-link load balancing is achieved through MLO, and combined with OFDMA, MU-MIMO and other technologies, resource utilization is further optimized.

[0114] Based on the above embodiments, the embodiments of the present application also provide a multi-link based network traffic scheduling device. Figure 7 is a schematic diagram of a multi-link network traffic scheduling device according to an embodiment of the present application. Figure 7 As shown, the multi-link-based network traffic scheduling device 700 may include a connection acquisition module 710, a priority determination module 720, and a resource allocation module 730 in sequence.

[0115] Among them, the acquisition module 710 is used to receive data packets sent by multiple terminals and determine the user role corresponding to each terminal; the priority determination module 720 is used to perform deep packet inspection on the data packets sent by the terminal for any data packet, obtain the application information of the data packet, and dynamically determine the priority of the data packet based on the application information, the corresponding user role and the network status information; the resource allocation module 730 is used to select MLO links and allocate wireless network resources for each data packet based on the priority of each data packet, and dynamically adjust the MLO link selection and allocation of wireless network resources based on the feedback information of each terminal.

[0116] Therefore, by using deep packet inspection technology to identify the application of the data packets sent by each terminal, and further dynamically setting the priority of each data packet according to the application information corresponding to each data packet, the user role corresponding to each terminal and the current network status information, finally, MLO link selection and dynamic allocation of wireless network resources are performed according to the priority of each data packet. In this way, the allocation process takes into account the impact of applications, users and network status on resource allocation, realizes priority scheduling of data packets for key users, and improves the accuracy, flexibility and reliability of network traffic scheduling.

[0117] In some embodiments, the application information includes application type and application traffic characteristics; the priority determination module 720 is specifically used to: extract features from the data packet to obtain the protocol type, port number, and load characteristics of the data packet; determine the application type and application traffic characteristics corresponding to the data packet based on the protocol type, port number, and load characteristics.

[0118] In some embodiments, the priority determination module 720 is also specifically used to: determine the initial priority of the data packet based on the application type of the data packet, application traffic characteristics and the corresponding user role; dynamically adjust the initial priority of the data packet according to a preset strategy and based on the current network status information to obtain the priority of the data packet; wherein the preset strategy includes lowering the priority of the data packet whose user role is an ordinary user when network congestion is determined based on the network status information.

[0119] In some implementations, the resource allocation module 730 is specifically used to: obtain the current status information of each link; select an MLO link for each data packet based on the priority of each data packet and the current status information of each link; allocate each data packet to each preset WMM queue based on the priority of each data packet, and allocate wireless network resources to each data packet based on the priority of each data packet and the corresponding terminal type.

[0120] In some implementations, the resource allocation module 730 is further specifically used to: obtain the status information of each link using Link Telemetry; the status information of each link includes at least bandwidth utilization, packet loss rate, delay and signal strength.

[0121] In some embodiments, wireless network resources include at least time resources, frequency resources, and space resources; the resource allocation module 730 is also specifically used to: select a corresponding transmission mode for each data packet based on the corresponding terminal type of each data packet; adjust the time resources, frequency resources, and space resources allocated to each data packet based on the priority of each data packet, the current status information of each link, and user needs.

[0122] In some implementations, the feedback information includes at least delay and packet loss rate; the resource allocation module 730 is also specifically used to dynamically adjust the MLO link selection and allocated wireless network resources for the data packets corresponding to each terminal based on the delay and packet loss rate of each terminal.

[0123] It should be noted that for details not disclosed in the multi-link-based network traffic scheduling device of this embodiment, please refer to the details disclosed in the embodiment of the multi-link-based network traffic scheduling method in the embodiments of this specification, and will not be repeated here.

[0124] Based on the above embodiments, an embodiment of the present application further provides an electronic device. Figure 8 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 8As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication interface 820 and the memory 830 communicate with each other through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute a multi-link-based network traffic scheduling method, the method comprising: receiving data packets sent by multiple terminals and determining the user role corresponding to each terminal; for any data packet, performing deep packet inspection on the data packet sent by the terminal to obtain application information of the data packet, and dynamically determining the priority of the data packet based on the application information, the corresponding user role and the network status information; performing MLO link selection and allocating wireless network resources for each data packet based on the priority of each data packet, and dynamically adjusting the MLO link selection and allocation of wireless network resources based on the feedback information of each current terminal.

[0125] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0126] On the basis of the above embodiments, on the other hand, the present invention also provides a computer program product, which includes a computer program, and the computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the multi-link-based network traffic scheduling method provided by the above methods, the method including: receiving data packets sent by multiple terminals, and determining the user role corresponding to each terminal; for any data packet, performing deep packet inspection on the data packet sent by the terminal to obtain application information of the data packet, and dynamically determining the priority of the data packet based on the application information, the corresponding user role and the network status information; performing MLO link selection and allocating wireless network resources for each data packet based on the priority of each data packet, and dynamically adjusting the MLO link selection and allocation of wireless network resources based on the feedback information of each terminal.

[0127] On the basis of the above embodiments, on the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the multi-link-based network traffic scheduling method provided by the above methods, the method comprising: receiving data packets sent by multiple terminals, and determining the user role corresponding to each terminal; for any data packet, performing deep packet inspection on the data packet sent by the terminal to obtain application information of the data packet, and dynamically determining the priority of the data packet based on the application information, the corresponding user role and the network status information; performing MLO link selection and allocating wireless network resources for each data packet based on the priority of each data packet, and dynamically adjusting the MLO link selection and allocation of wireless network resources based on the feedback information of each current terminal.

[0128] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0129] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-link based network traffic scheduling method, characterized in that: include: Receiving data packets sent by multiple terminals, and determining a user role corresponding to each of the terminals; For any data packet, perform deep packet inspection on the data packet sent by the terminal to obtain application information of the data packet, and dynamically determine the priority of the data packet based on the application information, the corresponding user role and the network status information; MLO link selection and wireless network resource allocation are performed for each data packet based on the priority of each data packet, and the MLO link selection and wireless network resource allocation are dynamically adjusted based on the current feedback information of each terminal.

2. The multi-link based network traffic scheduling method according to claim 1, characterized in that: The application information includes application type and application traffic characteristics; the deep packet inspection is performed on the data packet sent by the terminal to obtain the application information of the data packet, including: Extracting features of the data packet to obtain the protocol type, port number and load features of the data packet; The application type and application traffic characteristics corresponding to the data packet are determined based on the protocol type, the port number, and the load characteristics.

3. The multi-link based network traffic scheduling method according to claim 2, characterized in that: The dynamically determining the priority of the data packet based on the application information, the corresponding user role and the network status information includes: Determining an initial priority of the data packet based on an application type of the data packet, application traffic characteristics, and a corresponding user role; The initial priority of the data packet is dynamically adjusted according to a preset strategy and based on the current network status information to obtain the priority of the data packet; wherein the preset strategy includes lowering the priority of the data packet whose user role is an ordinary user when network congestion is determined based on the network status information.

4. The multi-link based network traffic scheduling method according to claim 1, characterized in that: The performing MLO link selection and allocating wireless network resources for each of the data packets based on the priority of each of the data packets includes: Get the status information of each current link; Based on the priority of each data packet and the current status information of each link, assigning the MLO link selection to each data packet; Allocate each of the data packets to each preset WMM queue based on the priority of each of the data packets, and allocate the wireless network resources to each of the data packets based on the priority of each of the data packets and the corresponding terminal type.

5. The multi-link based network traffic scheduling method according to claim 4, characterized in that: The obtaining of the current status information of each link includes: Link Telemetry is used to obtain the status information of each link; the status information of each link includes at least bandwidth utilization, packet loss rate, delay, signal strength, link quality index and SLA priority parameter.

6. The multi-link based network traffic scheduling method according to claim 4, characterized in that: The wireless network resources include at least time resources, frequency resources and space resources; the wireless network resources are allocated to each data packet based on the priority of each data packet and the corresponding terminal type, including Selecting a corresponding transmission mode for each of the data packets based on the corresponding terminal type of each of the data packets; Based on the priority of each data packet, the current status information of each link and user demand, the time resources, the frequency resources and the space resources allocated to each data packet are adjusted.

7. The multi-link based network traffic scheduling method according to any one of claims 1 to 6, characterized in that: The feedback information includes at least a delay and a packet loss rate; the dynamically adjusting the MLO link selection and the allocation of the wireless network resources based on the current feedback information of each terminal includes: Based on the delay and packet loss rate of each terminal, the MLO link selection and allocated wireless network resources of the data packets corresponding to each terminal are dynamically adjusted.

8. A multi-link based network traffic scheduling device, characterized in that: include: An acquisition module, used for receiving data packets sent by multiple terminals and determining a user role corresponding to each of the terminals; A priority determination module, for performing deep packet inspection on a data packet sent by the terminal for any data packet, obtaining application information of the data packet, and dynamically determining the priority of the data packet based on the application information, a corresponding user role and network status information; A resource allocation module is used to select an MLO link and allocate wireless network resources for each data packet based on the priority of each data packet, and dynamically adjust the MLO link selection and allocation of the wireless network resources based on the current feedback information of each terminal.

9. An electronic device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the multi-link-based network traffic scheduling method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the multi-link-based network traffic scheduling method according to any one of claims 1 to 7 is implemented.

Images