Wireless communication link optimization system and method

Abstract

The invention discloses a wireless communication link optimization system and method, and relates to the field of wireless communication. According to the method, the defect that the communication quality is evaluated only by depending on a single or few performance indexes in the prior art is overcome, and a comprehensive evaluation model which considers multiple factors and can accurately reflect the task specificity in the cloud-edge collaborative environment is constructed; the problem of poor adaptability of static QoS configuration to network environment change in the prior art is also overcome; meanwhile, the limitation that QoS parameters are difficult to preset and adjust in the prior art is improved, a set of mechanism capable of flexibly adjusting the QoS parameters is designed, the QoS strategy can be periodically generated and issued, and manual intervention is not needed.

Description

Technical Field

[0001] The present invention relates to the field of wireless communications, and in particular to a wireless communications link optimization system and method. Background Art

[0002] The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.

[0003] With the rapid development of cloud computing and edge computing technologies, cloud-edge collaboration has become an important part of modern network architecture. Cloud-edge collaboration aims to improve system performance, reduce latency, improve reliability, and save bandwidth by reasonably allocating computing, storage, and network resources between cloud and edge devices. However, wireless communication link optimization in a cloud-edge collaboration environment has become one of the key technologies to achieve these goals.

[0004] In the traditional cloud computing model, all data processing and storage tasks are concentrated in remote cloud data centers. Although this centralized model has advantages in terms of resource utilization and maintenance costs, the delay problem of data transmission is more prominent in application scenarios with high real-time requirements. With the rapid increase in the number of IoT devices and the diversification of application scenarios, the traditional centralized cloud computing model is difficult to meet the needs of low latency, high bandwidth and high reliability.

[0005] Edge computing effectively alleviates the drawbacks of centralized cloud computing by deploying computing and storage resources at the edge of the network close to the data source. However, edge computing alone cannot fully utilize the powerful resource integration capabilities of cloud computing. Therefore, cloud-edge collaboration came into being, which combines the advantages of cloud computing and edge computing, and achieves optimal resource allocation by dynamically adjusting the distribution of computing tasks in the cloud and edge.

[0006] In the cloud-edge collaborative system, the wireless communication link is the key channel connecting the cloud and edge nodes. Due to the variability and uncertainty of the wireless communication environment, the fluctuation of the link quality will directly affect the overall performance of the system. In order to ensure the efficient operation of the cloud-edge collaborative system, it is necessary to optimize the wireless communication link, reasonably allocate link resources for different cloud-edge collaborative operation and maintenance tasks, and ensure the reliability of data transmission.

[0007] It is crucial to schedule operation and maintenance tasks in the cloud-edge collaborative architecture, which is large and complex and often requires decentralized monitoring and centralized management. First, through task scheduling, resource allocation can be rationalized and load balanced, and resource usage of cloud and edge devices can be optimized to avoid resource waste. Secondly, reasonable task scheduling can meet the needs of real-time applications, significantly reduce data transmission time and improve response speed by processing data nearby. Scheduling tasks that are sensitive to latency to the edge and scheduling tasks that require high computing power but are not sensitive to latency to the cloud can achieve better performance. In addition, through task redundancy and fault isolation, the reliability and fault tolerance of the system are enhanced to prevent single point failures from affecting the entire system. By preprocessing data at the edge node and reducing the amount of data transmitted to the cloud, bandwidth costs can be saved, while data filtering and aggregation further reduce bandwidth consumption. In addition, through dynamic scheduling, task allocation can be adjusted according to actual conditions, the flexibility and scalability of the system can be improved, and it is easier to add new edge nodes or cloud resources. Finally, optimizing task scheduling can reduce unnecessary computing and transmission, reduce energy consumption, achieve green computing, balance the energy consumption of each node, and extend the service life of the equipment.

[0008] In the cloud-edge collaboration system, communication quality assessment and QoS (Quality of Service) management are key links to ensure the efficient execution of key tasks and improve user experience. In the prior art, although there are some communication quality assessment and QoS optimization methods for cloud-edge collaboration, there are still some limitations and problems to be solved in practical applications.

[0009] Existing Solution 1: Traditional Static QoS Configuration

[0010] In existing cloud-edge collaborative systems, a common practice is to pre-set fixed QoS parameters for various tasks. This static configuration method usually has the following steps: roughly classify transmission tasks into several categories, and then set fixed bandwidth allocation ratios, priorities, maximum delay thresholds and other QoS parameters for each category; then configure the corresponding QoS policies on each node in the cloud-edge network to achieve differentiated services for different categories of tasks.

[0011] However, precisely because this solution adopts the static, pre-set QoS parameter implementation method of QoS parameter setting and network equipment configuration, the following problems exist: the network environment, task load, equipment status, etc. are often in dynamic change, and the statically configured QoS parameters are difficult to reflect the actual needs in real time and accurately, which may lead to waste of resources or substandard service quality of key tasks; in the face of emergencies such as network congestion, equipment failure, emergency task insertion, etc., static configuration is difficult to respond quickly and cannot flexibly adjust QoS policies to cope with changes; frequent monitoring and manual adjustments are required, which increases operation and maintenance costs.

[0012] Existing Solution 2: Communication Quality Evaluation Based on Simple Indicators

[0013] Another type of existing technology focuses on monitoring and statistical analysis of basic performance indicators (such as bandwidth utilization, latency, packet loss rate, etc.) during cloud-edge collaborative communication. This type of method usually includes the following steps: deploy monitoring tools at cloud edge nodes to collect various performance indicator data during communication in real time; perform statistics and calculations on the collected data to generate various performance reports, such as average latency, maximum packet loss rate, etc.; identify whether there are communication quality problems based on preset thresholds or experience judgments, and try to locate the cause of the fault; based on the analysis results, make local adjustments to network equipment parameters, task scheduling strategies, etc., in an attempt to improve communication quality.

[0014] Although this method can provide a certain degree of communication quality evaluation, it has the following limitations: it only focuses on a single or a few performance indicators, ignores the comprehensive impact of task specificity and network dynamics on communication quality in the cloud-edge collaborative environment, and the evaluation results may not be sufficient to fully reflect the actual service quality. It relies on periodic or post-analysis, and the response to sudden communication quality problems is not timely enough. Analysis based on simple indicators cannot achieve a detailed analysis of the communication quality of the cloud-edge collaborative network. Summary of the invention

[0015] The purpose of the present invention is to: in response to the problems existing in the prior art, provide a wireless communication link optimization system and method, make up for the deficiency of the prior art that only relies on a single or a few performance indicators to evaluate the communication quality, and construct a comprehensive evaluation model that considers multiple factors and can accurately reflect the task specificity in the cloud-edge collaborative environment; it also overcomes the problem that the static QoS configuration in the prior art has poor adaptability to changes in the network environment; at the same time, it improves the limitations of the prior art that QoS parameters are pre-set and difficult to adjust, and designs a mechanism that can flexibly adjust QoS parameters, so that QoS policies can be periodically generated and issued without manual intervention.

[0016] The technical solution of the present invention is as follows:

[0017] A wireless communication link optimization system, comprising:

[0018] A link quality testing module, wherein the link quality testing module obtains real-time inter-node communication link parameters of each edge node;

[0019] Data aggregation module, all data aggregation modules aggregate the inter-node communication link parameters of all edge nodes and store them in the database for the QoS policy update module to update the QoS parameters;

[0020] A task classification module, which classifies transmission tasks according to task characteristics and specifies link parameters that each transmission task focuses on, so that the QoS policy update module updates QoS parameters;

[0021] A QoS policy update module, which combines the processing results of the data aggregation module and the task classification module and allocates a suitable communication link according to task characteristics and QoS parameters;

[0022] The link allocation scheme issuing module issues the allocation result of the QoS policy updating module to each edge node, and the edge node can allocate a suitable communication link to the transmission task according to the updated policy.

[0023] Furthermore, the inter-node communication link parameters include:

[0024] Throughput, latency, signal-to-noise ratio, packet loss rate, bit error rate, and link distortion.

[0025] Furthermore, the task classification module first determines the types of node transmission tasks in cloud-edge collaboration, and analyzes the requirements of each task for performance indicators of the communication link.

[0026] Furthermore, the types of node transmission tasks include:

[0027] Sensor data, control data, status data, network management data, and log data;

[0028] The performance indicators of the communication link include:

[0029] Data volume, real-time, high throughput, low latency, high signal-to-noise ratio, low packet loss rate, low bit error rate, and low link distortion.

[0030] Furthermore, the QoS policy update module has a built-in optimization model, which calculates the mean square error between the communication link performance required for each type of task and the actual performance of the communication link, and allocates a suitable communication link according to the communication link allocation strategy.

[0031] Furthermore, the optimization model is expressed as follows:

[0032]

[0033] in:

[0034] For the task set, T is the number of tasks;

[0035] is the set of communication links, L is the number of communication links;

[0036] x ij is a binary decision variable, indicating whether task i is assigned to communication link j. If so, then x ij =1, otherwise x ij =0;

[0037] represents the mean square error between the kth performance requirement of task i and the kth performance indicator on the real-time communication link j; d i =(d i1 ,d i2 ,...,d i6 ) is the performance requirement vector for each task, corresponding to the six indicators of throughput, delay, signal-to-noise ratio, packet loss rate, bit error rate and link distortion; c i =(c j1 ,c j2 ,...,c j6 ) is the real-time performance indicator vector of each communication link, which also corresponds to the six indicators of throughput, delay, signal-to-noise ratio, packet loss rate, bit error rate and link distortion; The smaller it is, the higher the match between task i and communication link j is, and the smaller the performance surplus is.

[0038] Furthermore, the communication link allocation strategy includes:

[0039] Compression: When a large amount of image or video information needs to be transmitted but there is no link that can meet the requirements, the cloud server requires the edge node to compress or block the data to be transmitted locally, and then transmit the compressed or block data through the communication link. The cloud server restores the compressed image or splices the block data to restore the original data;

[0040] Cache: When the existing communication link cannot meet the task transmission requirements, the cloud server requires the edge node to put the data to be sent in the sending cache and wait for the next QoS policy update of the cloud server before sending the data according to the new link allocation plan;

[0041] Preemption: For particularly urgent information, the cloud server sends an interrupt to the communication link that meets the real-time requirements but is occupied, allowing the communication link to serve more urgent tasks first. The previously interrupted task will regain the right to use the communication link after the urgent task is transmitted.

[0042] Furthermore, the present invention also proposes a wireless communication link optimization method, based on the above-mentioned wireless communication link optimization system, comprising:

[0043] Step S1: Obtain the real-time inter-node communication link parameters of each edge node;

[0044] Step S2: Summarize the inter-node communication link parameters of all edge nodes and store them in a database;

[0045] Step S3: classify the transmission tasks according to the task characteristics, and specify the link parameters that each transmission task focuses on;

[0046] Step S4: combining the processing results of step S2 and step S3, and allocating a suitable communication link according to the task characteristics and QoS parameters;

[0047] Step S5: The allocation result is distributed to each edge node, and the edge node can allocate a suitable communication link to the transmission task according to the updated strategy.

[0048] Furthermore, the step S3 includes:

[0049] Step S31: Determine the type of node transmission task in cloud-edge collaboration;

[0050] Step S32: Analyze the performance indicator requirements of each task for the communication link.

[0051] Furthermore, the step S4 comprises:

[0052] Step S41: Calculate the mean square error between the communication link performance required for each type of task and the actual performance of the communication link;

[0053] Step S42: Allocate appropriate communication links according to the communication link allocation strategy.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] 1. The present invention improves the accuracy of communication quality evaluation: the link quality test module not only takes into account traditional performance indicators, but also incorporates packet loss rate estimation and link distortion estimation, making the evaluation results more comprehensive and accurate, and can truly reflect the actual communication service quality.

[0056] 2. The present invention improves the adaptive capability and flexibility of the system: the present invention can regularly collect communication link quality data, automatically generate and adjust QoS policies, and avoid the limitations of traditional static configuration when facing network changes.

[0057] 3. The present invention improves the service guarantee of key tasks: by dynamically adjusting the QoS parameters and link allocation of transmission tasks, the present invention can ensure that key tasks still receive high-quality services when network resources are tight or emergencies occur.

[0058] 4. The present invention optimizes resource utilization: real-time evaluation and dynamic policy adjustment can avoid resource waste and unreasonable allocation, and improve the overall resource utilization efficiency, network performance and task execution efficiency of the cloud-edge collaborative system. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a system program flowchart proposed in Example 2;

[0060] Figure 2 This is a schematic diagram of the system structure proposed in Example 2. DETAILED DESCRIPTION

[0061] It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0062] The features and performance of the present invention are further described in detail below in conjunction with the embodiments.

[0063] Embodiment 1

[0064] A wireless communication link optimization system can be used for cloud-edge collaborative operation and maintenance task scheduling, including:

[0065] Link quality test module, the link quality test module obtains the real-time inter-node communication link parameters of each edge node; it should be noted that in cloud-edge collaboration, the communication link refers to the data transmission channel between the cloud service (such as a data center or a cloud server) and the edge device (such as an IoT device, an edge server, etc.), through which data collection, transmission, processing and control instructions are issued to ensure efficient, reliable and real-time communication between the cloud and edge devices to support the collaborative work of distributed computing and intelligent applications; the link quality test module can measure the performance indicators in the communication link;

[0066] Data aggregation module, all data aggregation modules aggregate the inter-node communication link parameters of all edge nodes and store them in the database for the QoS policy update module to update the QoS parameters;

[0067] A task classification module, which classifies transmission tasks according to task characteristics and specifies link parameters that each transmission task focuses on, so that the QoS policy update module updates QoS parameters;

[0068] A QoS policy update module, which combines the processing results of the data aggregation module and the task classification module and allocates a suitable communication link according to task characteristics and QoS parameters;

[0069] The link allocation scheme issuing module issues the allocation result of the QoS policy updating module to each edge node, and the edge node can allocate a suitable communication link to the transmission task according to the updated policy.

[0070] In this embodiment, specifically, the inter-node communication link parameters include:

[0071] Throughput rate, delay, signal-to-noise ratio, packet loss rate, bit error rate and link distortion;

[0072] That is, in addition to basic communication parameters such as throughput, latency, and signal-to-noise ratio, each edge node will also send a certain amount of data packets to other edge nodes, and record the number of data packets sent and received, as well as any transmission errors or packet losses, in order to estimate the overall packet loss rate. In addition, specific signal waveforms are transmitted between edge nodes, and the receiving end compares the received waveform with the original waveform sent, such as cross-correlation or mean square error, to evaluate link distortion.

[0073] Specifically, for the nth channel (i.e., the nth communication link), the data format stored in the database is as follows:

[0074] Table 1 Data format in the database

[0075] Throughput Latency Signal-to-Noise Ratio Packet loss rate Bit Error Rate Link Distortion Channel n T L SNR PLR BER MSE

[0076] In this embodiment, specifically, the task classification module first determines the type of node transmission tasks in the cloud-edge collaboration, and analyzes the requirements of each task for the performance indicators of the communication link;

[0077] Among them, throughput T refers to the amount of data successfully transmitted per unit time in the communication link; high throughput can ensure the rapid transmission of large amounts of data in the cloud-edge collaborative system, improve the efficiency of data processing and response, and support complex applications and big data analysis;

[0078]

[0079] Where T is the throughput, in bits per second (bps), t is the transmission time, and D is the amount of data successfully transmitted during the transmission time, in bits (bit);

[0080] Latency L refers to the time required for data to be transmitted from the source node to the destination node. Low latency is crucial for cloud-edge collaborative applications with high real-time requirements (such as industrial control and autonomous driving). Too high latency may cause the system to respond untimely, affecting the accuracy and safety of operations.

[0081] L=T r -T t

[0082] Where L is the total delay, and the node calculation method is the timestamp T when the data is received. r The timestamp T when the data is sent t The difference is in seconds (s);

[0083] The signal-to-noise ratio (SNR) refers to the ratio of signal strength to noise strength, usually expressed in decibels (dB). A higher signal-to-noise ratio means better signal quality and lower bit error rate in the communication link, thereby improving the data transmission reliability and stability of the cloud-edge collaborative system.

[0084]

[0085] Where SNR is the signal-to-noise ratio, that is, the signal strength P s With the noise intensity P n The ratio of, in decibels (dB);

[0086] The packet loss rate (PLR) refers to the ratio of data packets lost during data transmission to the total number of data packets sent. A low packet loss rate is crucial for cloud-edge collaborative systems, especially in data transmission scenarios that require high reliability. A high packet loss rate will lead to incomplete data and increased retransmission overhead, affecting system performance.

[0087]

[0088] Where PLR ​​is the packet loss rate, that is, sending N total In the case of data packets, the number of data packets that the receiver failed to receive is N lost And the total number of packets N total The ratio of

[0089] The bit error rate (BER) refers to the ratio of the number of erroneous bits to the total number of bits sent during the transmission process. A lower bit error rate means higher accuracy of data transmission, reducing data errors and retransmission requirements, and improving the overall efficiency and reliability of the cloud-edge collaborative system.

[0090]

[0091] BER is the bit error rate, that is, sending N total bits, the number of bits that the receiver recognizes incorrectly is N error And the total number of bits N total The ratio of

[0092] Link distortion refers to the phenomenon that signals are distorted due to various interferences and attenuation during transmission. Lower link distortion can ensure accurate data transmission in the cloud-edge collaborative system and avoid misjudgment and system performance degradation caused by information distortion, especially in application scenarios with high requirements for data integrity.

[0093] The sender sends a standard sine wave, and the receiver receives this sine wave. Since the channel must be distorted, there must be a deviation between the received sine wave and the standard sine wave of the sender. The mean square error (MSE) is used to measure this deviation:

[0094]

[0095] Where MSE is the mean square error between the transmitted signal and the received signal, N is the number of sampling points for the original signal, and x i and i are the values ​​of the original signal and the received signal at the i-th sampling point respectively.

[0096] In this embodiment, specifically, the types of the node transmission tasks include:

[0097] Sensor data, control data, status data, network management data, and log data;

[0098] Among them, sensor data: generally refers to the raw data collected by the sensors carried by the nodes, usually including: temperature, humidity, audio, image, video, pressure, light intensity, sound, vibration intensity, gas concentration, etc. The demand of sensor data for communication links is medium to high throughput because of the frequent data transmission; the latency demand is medium to high, depending on the real-time requirements of the application; the signal-to-noise ratio needs to be kept medium to ensure data accuracy; the packet loss rate and bit error rate need to be low to avoid data loss and errors; link distortion also needs to be minimized to maintain data accuracy;

[0099] Control data: used to manage and control the operation of sensor nodes or networks, usually including configuration parameters, control instructions (such as sensor start / stop, mode switching, etc.), calibration commands, etc. The communication requirements of control data are mainly low throughput, because the amount of data is usually not large but needs to be transmitted quickly; low latency requirements to ensure timely execution of commands; high signal-to-noise ratio requirements to ensure accurate transmission of commands; low packet loss rate and bit error rate requirements to avoid system out of control or command errors; link distortion also needs to be kept at a low level to prevent command misunderstanding;

[0100] Status data: reflects the current health status of the sensors on the node, usually including: battery power of the node, signal strength, online / offline status of the node, memory usage, etc. Status data requires low throughput and medium latency for the communication link, because the data volume is not large and needs to be timely but not urgent; the signal-to-noise ratio and bit error rate require medium to ensure that the data reflects the true situation of the node but allow a certain error; the packet loss rate requires low, occasional packet loss can be tolerated but frequent packet loss will affect monitoring; the link distortion requires medium, distortion will affect status judgment but is not particularly sensitive;

[0101] Network management data: used to maintain network topology and ensure normal data transmission, usually including: routing update information, topology information, error reporting and correction information, network node joining / exit notifications, etc. The requirements of network management data for communication links include medium throughput, especially when the topology changes; low latency to ensure timely routing updates and error reporting; high signal-to-noise ratio to ensure the accuracy of management data; low packet loss rate and bit error rate to avoid network instability and management information errors; link distortion must also be minimized to maintain the effectiveness of network management;

[0102] Log data: Log data related to node network operation and maintenance. These data are used to ensure the normal operation and optimization of the network. They usually include: fault detection and diagnosis data, maintenance logs (including node fault records, repair records, etc.), performance monitoring data (such as latency, throughput, etc.), firmware update data, security and access control logs, etc. The requirements of the communication link for operation and maintenance data are medium throughput, especially when it comes to system logs and performance monitoring data; medium latency, most data does not require real-time but fault detection data requires low latency; medium signal-to-noise ratio and bit error rate to ensure data accuracy but allow a certain error; medium packet loss rate requirements, occasional packet loss can be tolerated but frequent packet loss will affect operation and maintenance efficiency; link distortion must also be kept at medium, distortion will affect the accuracy of operation and maintenance data but is not particularly sensitive;

[0103] The performance indicators of the communication link include:

[0104] Data volume, real-time performance, high throughput, low latency, high signal-to-noise ratio, low packet loss rate, low bit error rate, and low link distortion;

[0105] It should be noted that the requirements of various tasks for link performance indicators are as follows:

[0106] Table 2 Requirements of various tasks for link performance indicators

[0107] Performance Indicators Sensor Data Controlling Data Status data Network management data Log data Data volume medium Lower Lower medium Higher Real-time medium Higher Lower Higher medium High throughput medium Lower Lower medium Higher Low latency medium Higher medium medium medium High signal-to-noise ratio medium Higher medium medium medium Low packet loss rate Higher Higher Lower medium medium Low bit error rate Higher Higher medium medium medium Low link distortion Higher Higher medium medium Higher

[0108] It should be noted that the classification method of tasks can refer to the following table:

[0109] Table 3 Transmission task division method

[0110]

[0111]

[0112] In this embodiment, specifically, the QoS policy update module has a built-in optimization model, which calculates the mean square error between the communication link performance required for each type of task and the actual performance of the communication link, and allocates a suitable communication link according to the communication link allocation strategy.

[0113] In this embodiment, specifically, the optimization model is expressed as follows:

[0114]

[0115] in:

[0116] For the task set, T is the number of tasks;

[0117] is the set of communication links, L is the number of communication links;

[0118] x ij is a binary decision variable, indicating whether task i is assigned to communication link j. If so, then x ij =1, otherwise x ij =0;

[0119] represents the mean square error between the kth performance requirement of task i and the kth performance indicator on the real-time communication link j; d i =(d i1 ,d i2 ,...,d i6 ) is the performance requirement vector for each task, corresponding to the six indicators of throughput, delay, signal-to-noise ratio, packet loss rate, bit error rate and link distortion; c i =(c j1 ,c j2 ,...,c j6) is the real-time performance indicator vector of each communication link, which also corresponds to the six indicators of throughput, delay, signal-to-noise ratio, packet loss rate, bit error rate and link distortion; The smaller it is, the higher the match between task i and communication link j is, and the smaller the performance surplus is. By calculating the mean square error of the performance between the task and the link, the link that best matches the task is measured, and resource utilization is maximized while meeting the transmission requirements. For example, if the performance requirement of a task throughput is 1Mbps, and the performance indicators of the throughput of the two links are 1.2Mbps and 2Mbps respectively, it can be concluded that the link with a throughput of 1.2Mbps has a smaller mean square error than the task's 1Mbps throughput requirement, which means that the link with a throughput of 1.2Mbps is more suitable for the task. In actual calculations, it is necessary to comprehensively consider six indicators: throughput, latency, signal-to-noise ratio, packet loss rate, bit error rate, and link distortion to more accurately measure the match between tasks and links.

[0120] The first constraint Indicates that each task must be assigned to a link. The second constraint Indicates that each link can only carry one task.

[0121] In this embodiment, specifically, when considering the link allocation strategy, it should also be noted that there may be a situation where a task is not allocated to a suitable link. For example, a node needs to transmit a high-resolution image, but the throughput of the existing link cannot meet its needs. In this case, the cloud server needs to specify a "special strategy" separately. For the above situation, the node is required to compress the image locally until the existing link can meet the transmission of the compressed image before transmission. This type of allocation strategy can be divided into the following types:

[0122] Compression: When a large amount of information such as images or videos needs to be transmitted but there is no link that can meet the requirements, the cloud server will require the node to compress or block the data to be transmitted locally, and then transmit the compressed or block data through the link. The cloud server will restore the compressed image or splice the block data to restore the original data.

[0123] Cache: When the existing link cannot meet the task transmission requirements, for example, high link occupancy leads to low throughput, but the transmission task is not particularly urgent, the cloud server will require the node to place the data to be sent in the sending cache and wait for the next cloud server QoS policy update before sending the data according to the new link allocation plan.

[0124] Preemption: For particularly urgent information, such as control instructions with high real-time requirements, it must be transmitted to the cloud server within the specified time. At this time, the cloud server will send an interrupt to the link that meets the real-time requirements but is occupied, so that the link can be used for more urgent tasks first. The previously interrupted task will regain the right to use the link after the urgent task is transmitted.

[0125] In this embodiment, specifically, it also includes: a log recording module, which saves each step of the system's evaluation operation and the final evaluation result in a log, so as to facilitate checking the cause of the failure when a system failure occurs.

[0126] This embodiment further proposes a wireless communication link optimization method, based on the wireless communication link optimization system described above, including:

[0127] Step S1: Obtain the real-time inter-node communication link parameters of each edge node;

[0128] Step S2: Summarize the inter-node communication link parameters of all edge nodes and store them in a database;

[0129] Step S3: classify the transmission tasks according to the task characteristics, and specify the link parameters that each transmission task focuses on;

[0130] Step S4: combining the processing results of step S2 and step S3, and allocating a suitable communication link according to the task characteristics and QoS parameters;

[0131] Step S5: The allocation result is distributed to each edge node, and the edge node can allocate a suitable communication link to the transmission task according to the updated strategy.

[0132] In this embodiment, specifically, step S3 includes:

[0133] Step S31: Determine the type of node transmission task in cloud-edge collaboration;

[0134] Step S32: Analyze the performance indicator requirements of each task for the communication link.

[0135] In this embodiment, specifically, step S4 includes:

[0136] Step S41: Calculate the mean square error between the communication link performance required for each type of task and the actual performance of the communication link;

[0137] Step S42: Allocate appropriate communication links according to the communication link allocation strategy.

[0138] Embodiment 2

[0139] Figure 1The system flowchart of this embodiment is shown, in which the link quality test module first tests the link quality. In addition to the basic communication parameters such as throughput, delay, signal-to-noise ratio, etc., each node will also send a certain amount of data packets to other nodes, and record the number of data packets sent and received and any transmission errors or packet loss, so as to estimate the overall packet loss rate. In addition, a specific signal waveform will be transmitted between nodes. The receiving end will compare the received waveform with the original waveform sent, such as cross-correlation or mean square error to evaluate the link distortion. After the statistics are completed, the data will be sent to the data aggregation module. After the data is processed, it will be sent to the QoS policy update module. On the other side, the classification module of the transmission task will also send the tasks classified and graded to the QoS policy update module. The module integrates the information of the two, assigns the corresponding communication links to the tasks of different categories and grades, and integrates them into a new QoS policy. Then, the final QoS policy is sent to the link allocation plan sending module for updating the plan to the edge node to complete the QoS policy update. This update process is carried out periodically. If you want to stop the update and issuance of the QoS policy, you need to send an "end" instruction to the cloud server that manages the update and issuance of the QoS policy.

[0140] Figure 2 The schematic diagram of the system structure of this embodiment shows that it can be generally divided into a cloud server, a master node and an edge node (i.e., the edge node in the figure). The cloud server is responsible for grading the tasks, receiving the link quality information sent by the master node, and comprehensively calculating the new link allocation scheme (QoS strategy). The master node is responsible for collecting link quality information from the edge nodes, and retrieving the new QoS policy from the cloud server and sending it to each edge node. The edge node is responsible for collecting link quality information and providing priority communication links for the assigned tasks after updating the QoS policy. During the entire cloud-edge collaboration process, the communication between nodes is wireless transmission.

[0141] The above-mentioned embodiments only express the specific implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the protection scope of the present application. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the technical solution concept of the present application, and these all belong to the protection scope of the present application.

[0142] This background section is provided to generally present the context of the invention, and the work of the presently named inventors, the work to the extent described in this background section, and aspects of this section that did not constitute prior art at the time of application are neither explicitly nor implicitly admitted to be prior art to the present invention.

Claims

1. A wireless communication link optimization system, characterized in that: include: A link quality testing module, wherein the link quality testing module obtains real-time inter-node communication link parameters of each edge node; Data aggregation module, all data aggregation modules aggregate the inter-node communication link parameters of all edge nodes and store them in the database for the QoS policy update module to update the QoS parameters; A task classification module, which classifies transmission tasks according to task characteristics and specifies link parameters that each transmission task focuses on, so that the QoS policy update module updates QoS parameters; A QoS policy update module, which combines the processing results of the data aggregation module and the task classification module and allocates a suitable communication link according to task characteristics and QoS parameters; The link allocation scheme issuing module issues the allocation result of the QoS policy updating module to each edge node, and the edge node can allocate a suitable communication link to the transmission task according to the updated policy.

2. A wireless communication link optimization system according to claim 1, characterized in that: The inter-node communication link parameters include: Throughput, latency, signal-to-noise ratio, packet loss rate, bit error rate, and link distortion.

3. A wireless communication link optimization system according to claim 1, characterized in that: The task classification module first determines the types of node transmission tasks in cloud-edge collaboration, and analyzes the requirements of each task for the performance indicators of the communication link.

4. A wireless communication link optimization system according to claim 3, characterized in that: The types of node transmission tasks include: Sensor data, control data, status data, network management data, and log data; The performance indicators of the communication link include: Data volume, real-time, high throughput, low latency, high signal-to-noise ratio, low packet loss rate, low bit error rate, and low link distortion.

5. A wireless communication link optimization system according to claim 1, characterized in that: The QoS policy update module has a built-in optimization model, which calculates the mean square error between the communication link performance required for each type of task and the actual performance of the communication link, and allocates a suitable communication link according to the communication link allocation strategy.

6. A wireless communication link optimization system according to claim 5, characterized in that: The optimization model is expressed as follows: in: For the task set, T is the number of tasks; is the set of communication links, L is the number of communication links; x ij is a binary decision variable, indicating whether task i is assigned to communication link j. If so, then x ij =1, otherwise x ij =0; represents the mean square error between the kth performance requirement of task i and the kth performance indicator on the real-time communication link j; d i =(d i1 ,d i2 ,...,d i6 ) is the performance requirement vector for each task, corresponding to the six indicators of throughput, delay, signal-to-noise ratio, packet loss rate, bit error rate and link distortion; c i =(c j1 ,c j2 ,...,c j6 ) is the real-time performance indicator vector of each communication link, which also corresponds to the six indicators of throughput, delay, signal-to-noise ratio, packet loss rate, bit error rate and link distortion; The smaller it is, the higher the match between task i and communication link j is, and the smaller the performance excess is.

7. A wireless communication link optimization system according to claim 6, characterized in that: The communication link allocation strategy includes: Compression: When a large amount of image or video information needs to be transmitted but there is no link that can meet the requirements, the cloud server requires the edge node to compress or block the data to be transmitted locally, and then transmit the compressed or block data through the communication link. The cloud server restores the compressed image or splices the block data to restore the original data; Cache: When the existing communication link cannot meet the task transmission requirements, the cloud server requires the edge node to put the data to be sent in the sending cache and wait for the next QoS policy update of the cloud server before sending the data according to the new link allocation plan; Preemption: For particularly urgent information, the cloud server sends an interrupt to the communication link that meets the real-time requirements but is occupied, allowing the communication link to serve more urgent tasks first. The previously interrupted task will regain the right to use the communication link after the urgent task is transmitted.

8. A wireless communication link optimization method, characterized in that: A wireless communication link optimization system according to any one of claims 1 to 7, comprising: Step S1: Obtain the real-time inter-node communication link parameters of each edge node; Step S2: Summarize the inter-node communication link parameters of all edge nodes and store them in a database; Step S3: classify the transmission tasks according to the task characteristics, and specify the link parameters that each transmission task focuses on; Step S4: combining the processing results of step S2 and step S3, and allocating a suitable communication link according to the task characteristics and QoS parameters; Step S5: The allocation result is distributed to each edge node, and the edge node can allocate a suitable communication link to the transmission task according to the updated strategy.

9. A wireless communication link optimization method according to claim 8, characterized in that: The step S3 comprises: Step S31: Determine the type of node transmission task in cloud-edge collaboration; Step S32: Analyze the performance indicator requirements of each task for the communication link.

10. A wireless communication link optimization method according to claim 8, characterized in that: The step S4 comprises: Step S41: Calculate the mean square error between the communication link performance required for each type of task and the actual performance of the communication link; Step S42: Allocate appropriate communication links according to the communication link allocation strategy.

Images