Smart campus system based on SDN

Abstract

The invention provides a smart campus system based on an SDN, and belongs to the technical field of informatization construction of campus teaching, scientific research and management, the smart campus system comprises an application layer, a management layer and a resource layer, the resource layer is used for collecting data information and analyzing enhanced data to obtain useful information, and the useful information is input to the management layer; the resource layer comprises a resource data collection module, an intelligent decision-making module, a real-time performance monitoring module, an SDN controller and a reinforcement learning module. The resource data collection module, the resource utilization analysis module and the intelligent decision-making module carry out processing and analysis of a deep Q network through the reinforcement learning module and input the processed and analyzed data to the SDN controller. After the scheme is adopted, the intelligent campus system has the following advantages that teaching, management and scientific research of a school are integrated through the latest concept based on the SDN technology, the transmission technology and the education technology, an organic intelligent campus system is formed, and all elements of a traditional campus can be subjected to digital intelligent management.

Description

Technical Field

[0001] The present invention relates to the technical field of information construction of teaching, scientific research and management in a campus, and specifically refers to a smart campus system based on SDN. Background Art

[0002] The main problems in the current school education informatization construction are: the application system and technical solutions are not unified. Due to historical reasons, with the establishment of various information application systems in schools, these subsystems and information resource libraries have adopted different database systems and application development technologies; system management is decentralized, and the user management and authorization mechanisms of each subsystem are not unified; information is decentralized, the data of each application system is inconsistent, and there is a lack of or even no data sharing relationship and exchange path, forming an "information island"; the comprehensive service capability of the information system is weak, it is difficult to implement comprehensive business processes and comprehensive queries, and the decision-making support capability is weak. As the school's management, it is unable to obtain an overall data view, and thus it is impossible to conduct a comprehensive analysis and utilization of the data; the allocation of teaching resources and technical forces between schools is unbalanced, duplication of construction, and lack of sharing mechanism; educational technology does not provide effective support for improving teaching quality and school management efficiency. The smart campus is based on the construction concept of a digital campus and uses information technology to change the way school managers, teachers, students, parents and campus resources interact with each other. However, the existing smart campuses do not pay enough attention to education and teaching and scientific research, and the understanding of smart campuses is not deep and accurate enough. The school's various teaching activities, scientific research projects, system management and campus resources and application systems have not been fully and effectively integrated, thus insufficient effective support for improving teaching quality and campus management efficiency.

[0003] For example, the Chinese patent with application number 201020218043.8 and titled “Campus Security System” only involves campus security, warnings, etc. through the specific setting of circuits, and does not involve other aspects of the campus; another example is the Chinese patent with application number 200520055375.4 and titled “A Campus Network Management System”, which manages the network within the campus through a central server and user terminals, but also does not involve the macro, overall campus system design. Summary of the invention

[0004] The present invention aims to solve the above technical problems and provide a smart campus system based on SDN.

[0005] In order to solve the above technical problems, the technical solution provided by the present invention is:

[0006] A smart campus system based on SDN, including:

[0007] Application layer, the application layer includes a campus management module responsible for managing various campus resources and operating status, a teaching management module responsible for teaching management, course arrangement, online learning, and performance analysis, a security monitoring module responsible for ensuring campus safety, an intelligent service module that provides intelligent services to improve campus life, an information promotion module that provides various services to the public, displays campus style or conveys campus dynamics, and a cloud management module for managing students, teachers, classes, and grades within the school;

[0008] Management layer, which includes a leading controller for network structure configuration, control management and data information, and an SDN network control module for distributed data storage, analysis, processing and connection;

[0009] Resource layer, the resource layer is used to collect data and analyze the reinforcement data to obtain useful information and input it to the management layer. The resource layer includes a resource data collection module, a resource utilization analysis module, an intelligent decision-making module, a real-time performance monitoring module, an SDN controller and a reinforcement learning module. The resource data collection module, the resource utilization analysis module and the intelligent decision-making module process and analyze the deep Q network through the reinforcement learning module and input it to the SDN controller;

[0010] The application layer is connected to the cloud platform, and the cloud platform includes a cloud SDN controller component. The cloud SDN controller component serves as a network management component of the cloud platform and is responsible for managing the network resource configuration of the cloud platform according to the cloud application load and resource usage on the cloud platform;

[0011] The cloud SDN controller component is connected to the intelligent decision module, and the information sent to the intelligent decision module includes network status, performance data, configuration changes, event notifications, resource reports and policy execution results to support the intelligent decision module in network optimization and management;

[0012] The SDN controller manages and configures the network infrastructure, and sends local application load conditions, network status, performance data, configuration changes, resource utilization, security events, traffic statistics, and basic equipment status information to the receiving intelligent decision module;

[0013] The intelligent decision-making module makes optimization decisions in real time according to the information sent by the cloud SDN controller component, the network and the SDN controller, and sends network configuration instructions, optimization strategies, resource allocation, policy updates, monitoring instructions, fault responses and compliance requirements to the cloud SDN controller component. The information sent to the SDN controller mainly includes network status analysis results, network micro-segmentation strategies, optimization strategy recommendations, resource scheduling instructions, policy update requests and performance adjustment requirements;

[0014] The application layer includes local applications and cloud applications. Local applications run on local physical servers, and cloud applications run on cloud platforms, which together constitute the application services of the smart campus. Local applications are connected to the SDN controller.

[0015] Preferably, the SDN network control module includes a northbound interface, a westbound interface, an eastbound interface and a southbound interface, the northbound interface is connected to the application layer, the westbound interface is connected to the leading controller, the southbound interface is connected to the resource layer, and the westbound interface is connected to the external network.

[0016] Preferably, the external network is interactively connected to a cloud platform management module, and the cloud platform management module is connected to a teacher terminal and a parent terminal to achieve instant communication between home and school.

[0017] Preferably, the resource data collection module collects campus teacher and student data and SDN network resource usage data.

[0018] Preferably, the resource utilization analysis module analyzes resource data and resource utilization efficiency through the big data technology of the reinforcement learning module, and provides resource optimization suggestions based on the resource analysis results to improve resource utilization efficiency and network performance.

[0019] Preferably, the intelligent decision-making module provides comprehensive SDN network analysis and optimization suggestions through the big data technology of the reinforcement learning module, providing intelligent support for the dominant controller.

[0020] Preferably, the real-time performance monitoring module monitors the performance of the SDN network in real time and combines with the security monitoring module to predict the development trend and plan and provide a reference solution.

[0021] Preferably, the campus management module includes dormitory management, classroom management, library management, laboratory management, canteen management, stadium management, school bus management and affairs repair management; the teaching management module includes student performance management, student attendance management, teacher schedule management, electronic homework, comprehensive teacher-student evaluation, and interactive classroom; the security monitoring module includes video monitoring and access control; the intelligent service module includes campus card management, payment management and one-card pass; the information promotion module includes campus announcement management, broadcast management, campus culture display, school portal and creative forum.

[0022] Preferably, the dominant controller has multiple leaf switches and multiple root switches that are interactively connected for data transmission. The multiple leaf switches are connected to servers, storage modules, and routers, and are respectively connected to the dominant controller. One root switch is connected to multiple leaf switches.

[0023] Preferably, the process formula (1) of iteratively updating the Q value of the deep Q network is:

[0024]

[0025] The above formula (1) converges to an optimal Q value, the target Q value y t It can be expressed by formula (2):

[0026]

[0027] Define the loss function Y(θ t ) back propagates to update the deep Q network parameters θ t , as shown in formula (3):

[0028] Y(θ t )=[y t -Q(S t ,a|θ t+1 )] 2 (3).

[0029] After adopting the above scheme, the present invention has the following advantages:

[0030] The present invention integrates the school's teaching, management and scientific research into one by using the latest concepts based on SDN technology, transmission technology and educational technology to form an organic smart campus system, which can perform digital smart management on all elements of the traditional campus.

[0031] The smart campus system comprehensively and effectively integrates the school's various teaching activities, scientific research projects, system management, campus resources and various application systems, improving the teaching quality and campus management efficiency, realizing a campus model of intelligent service and management as well as interactive uploading and synchronous management of resources, reducing the pressure on school servers, ensuring the interoperability and security of resources at all levels and types of schools, allowing teaching resources to be shared safely and quickly, and evaluating teachers and students through relevant data obtained through the smart campus system, providing a good home-school interaction experience.

[0032] The above summary is for illustrative purposes only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments and features described above, further aspects, embodiments and features of the present invention will be readily apparent by reference to the accompanying drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0034] Figure 1 is a system diagram of the present invention;

[0035] Figure 2 It is a connection diagram of the SDN network control module of the present invention;

[0036] Figure 3 It is a connection diagram of the main controller of the present invention;

[0037] Figure 4 is a schematic diagram of a deep Q network of the present invention;

[0038] Figure 5 It is a schematic diagram of the processing method of the deep Q network of the present invention;

[0039] Figure 6 It is a system diagram of the cloud SDN controller components of the present invention. DETAILED DESCRIPTION

[0040] Specific embodiments of the present invention will now be mentioned in detail. Although the present invention is described in conjunction with these specific embodiments, it should be appreciated that it is not intended to limit the present invention to these specific embodiments. On the contrary, these embodiments are intended to cover substitutions, changes or equivalent embodiments that may be included in the spirit and scope of the invention defined by the claims. In the following description, a large number of specific details are set forth in order to provide a comprehensive understanding of the present invention. The present invention can be implemented without some or all of these specific details. In other cases, in order not to make the present invention unnecessarily obscure, well-known process operations are not described in detail.

[0041] When used in conjunction with "including," "methods comprising," or similar language in this specification and the appended claims, the singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0042] The present invention is further described in detail below in conjunction with the full text.

[0043] Combined with Figure 1-Figure 5 , a SDN-based smart campus system, including:

[0044] Application layer, the application layer includes a campus management module responsible for managing various campus resources and operating status, a teaching management module responsible for teaching management, course arrangement, online learning, and performance analysis, a security monitoring module responsible for ensuring campus safety, an intelligent service module that provides intelligent services to improve campus life, an information promotion module that provides various services to the public, displays campus style or conveys campus dynamics, and a cloud management module for managing students, teachers, classes, and grades within the school;

[0045] Management layer, which includes a leading controller for network structure configuration, control management and data information, and an SDN network control module for distributed data storage, analysis, processing and connection;

[0046] Resource layer, the resource layer is used to collect data and analyze the reinforcement data to obtain useful information and input it into the management layer. The resource layer includes a resource data collection module, a resource utilization analysis module, an intelligent decision-making module, a real-time performance monitoring module, an SDN controller and a reinforcement learning module. The resource data collection module, the resource utilization analysis module and the intelligent decision-making module process and analyze the deep Q network through the reinforcement learning module and input it into the SDN controller.

[0047] The application layer is connected to the cloud platform, and the cloud platform includes a cloud SDN controller component. The cloud SDN controller component serves as a network management component of the cloud platform and is responsible for managing the network resource configuration of the cloud platform according to the cloud application load and resource usage on the cloud platform;

[0048] The cloud SDN controller component is connected to the intelligent decision module, and the information sent to the intelligent decision module includes network status, performance data, configuration changes, event notifications, resource reports and policy execution results to support the intelligent decision module in network optimization and management;

[0049] The SDN controller manages and configures the network infrastructure, and sends local application load conditions, network status, performance data, configuration changes, resource utilization, security events, traffic statistics, and basic equipment status information to the receiving intelligent decision module;

[0050] The intelligent decision-making module makes optimization decisions in real time according to the information sent by the cloud SDN controller component, the network and the SDN controller, and sends network configuration instructions, optimization strategies, resource allocation, policy updates, monitoring instructions, fault responses and compliance requirements to the cloud SDN controller component. The information sent to the SDN controller mainly includes network status analysis results, network micro-segmentation strategies, optimization strategy recommendations, resource scheduling instructions, policy update requests and performance adjustment requirements;

[0051] The application layer includes local applications and cloud applications. Local applications run on local physical servers, and cloud applications run on cloud platforms, which together constitute the application services of the smart campus. Local applications are connected to the SDN controller.

[0052] Furthermore, the intelligent decision module is responsible for receiving data from the cloud SDN controller component and the SDN controller, and performing intelligent analysis on the collected data.

[0053] Its main functions include:

[0054] 1. Adaptive learning and optimized decision-making: Using deep learning and reinforcement learning algorithms, the intelligent decision-making module can achieve adaptive learning, optimize the decision-making process, and improve the accuracy and efficiency of decision-making. By continuously learning network behavior patterns, it can automatically adjust algorithm parameters to adapt to changes in the network environment and formulate network micro-segmentation strategies;

[0055] 2. Network performance evaluation and automatic adjustment: Combining statistical analysis and machine learning algorithms, the module evaluates network performance in real time. Based on the evaluation results, the module can automatically adjust network configuration, including path selection, bandwidth allocation, quality of service (QoS) settings, etc., to optimize network performance and meet resource usage and user needs;

[0056] 3. Dynamic resource scheduling and pre-allocation: The intelligent decision-making module dynamically schedules computing resources, storage resources, and network resources, predicts future resource requirements through advanced prediction models, implements resource pre-allocation, and significantly reduces resource allocation delays;

[0057] 4. Real-time monitoring and security response: The module monitors network activities in real time and uses anomaly detection algorithms to identify potential security threats. Once abnormal behavior or security incidents are detected, the corresponding security response mechanism is immediately triggered, such as isolating the attack source and updating the security policy.

[0058] Furthermore, the connection interface between the cloud SDN controller component and the intelligent decision-making module, the connection interface between the SDN controller and the intelligent decision-making module, and the connection between the SDN controller and the local application adopt a northbound interface, using RESTful API or gRPC protocol; the SDN controller and the network infrastructure adopt a southbound interface, using OpenFlow or other southbound protocols for communication between the SDN controller and the network infrastructure (such as switches and routers) to realize the configuration and management of network equipment.

[0059] Furthermore, the application layer includes a campus management module responsible for managing various campus resources and operating status, a teaching management module responsible for academic management, course scheduling, online learning, grade analysis, student course selection, and teacher course scheduling, a security monitoring module responsible for ensuring campus safety, an intelligent service module that provides intelligent services to improve campus life, an information promotion module that provides various services to the public, showcases the campus style or conveys campus dynamics, and is used to manage students, teachers, classes, and grades within the school.

[0060] The existing OpenStack Neutron provides SDN network management functions, but they are not network management components of the cloud platform, but exist as an independent component of the cloud platform. Using the SDN controller as a cloud platform component can provide more comprehensive and flexible network management functions, such as:

[0061] (1) Dynamically adjust network configuration based on cloud platform requirements: Dynamically adjust network configuration, such as bandwidth allocation and security policies, based on the requirements of cloud platform applications to achieve flexible allocation and efficient use of network resources.

[0062] (2) Automated network management based on cloud platform automation tools: Utilize cloud platform automation tools, such as Heat and TOSCA, to automatically configure, manage, and optimize the network and improve network management efficiency.

[0063] (3) Manage network services based on the cloud platform service catalog: Use the cloud platform's service catalog to manage network services, such as firewalls, load balancers, VPNs, etc., and automatically allocate and manage service instances.

[0064] The SDN network control module includes a northbound interface, a westbound interface, an eastbound interface and a southbound interface.

[0065] The northbound interface is connected to the application layer. This interface is the interface between the SDN control layer and the application layer. Through this interface, applications in the top layer of the network can exchange information with each other to realize a programmable control network architecture. Unlike the southbound interface, the northbound interface has no standardized protocol. In the process of information exchange, the form of exchange is mainly determined by the network and the target application of the exchange.

[0066] The westbound interface is connected to the leading controller and serves as an information exchange channel between the SDN control layer and other network areas. The interface leads the routing decision of the controller through the exchange of network status information and realizes the seamless interaction of multiple network flows. To achieve smooth information interaction, the westbound interface usually adopts a standardized inter-domain routing protocol.

[0067] The southbound interface is connected to the resource layer, which is the network interface between the resource layer and the management layer, through which the execution of the external network can be controlled. Its functions are implemented by executing a standardized instruction set defined by hardware, such as the IETF For CES protocol and the OpenFlow protocol.

[0068] The eastbound interface is connected to the external network, and the interface can realize communication with other control layers in the non-SDN network. The teacher terminal or the parent terminal uses a smartphone to understand the student's life and learning status, which simplifies the process of teaching, management and home-school interaction, making the operation more convenient and efficient.

[0069] It integrates mobile phones, computer network integration, comprehensive services, etc. It provides a new way of communication between home and school, and provides specialized services for home-school interaction for students' parents, so that family education and campus education can be synchronized, jointly forming an educational synergy and promoting the healthy growth of minors.

[0070] The SDN-based smart campus system proposed in the present invention integrates the intelligent and automated management advantages of SDN technology and cloud computing technology, realizes efficient management of network resources through the cloud SDN controller component, optimizes cloud resource allocation, and improves network performance and reliability. At the same time, the integration of intelligent decision-making modules greatly improves the intelligence and efficiency of decision-making, supports cross-cloud service management, and enhances the security and compliance of the system, creating a flexible, efficient and secure network environment for the campus.

[0071] The external network interactively connects to the cloud platform management module, and the cloud platform management module is connected to the teacher terminal and the parent terminal to achieve instant communication between home and school.

[0072] The resource data collection module is used to collect campus teacher and student data and SDN network resource usage data.

[0073] The resource utilization analysis module analyzes resource data and resource utilization efficiency through the big data technology of the reinforcement learning module, and provides resource optimization suggestions based on the resource analysis results to improve resource utilization efficiency and network performance.

[0074] The intelligent decision-making module provides comprehensive SDN network analysis and optimization suggestions through the big data technology of the reinforcement learning module, providing intelligent support for the leading controller.

[0075] The real-time performance monitoring module monitors the performance of the SDN network in real time and combines with the security monitoring module to predict development trends and plan and provide reference solutions.

[0076] The campus management module includes dormitory management, classroom management, library management, laboratory management, canteen management, stadium management, school bus management, affairs repair management; the teaching management module includes student performance management, student attendance management, teacher schedule management, electronic homework, comprehensive teacher-student evaluation, and interactive classroom; the security monitoring module includes video monitoring and access control; the intelligent service module includes campus card management, payment management and one-card pass; the information promotion module includes campus announcement management, broadcast management, campus culture display, school portal and creative forum.

[0077] The dominant controller has multiple leaf switches and multiple root switches that are interactively connected for data transmission. The multiple leaf switches are connected to servers, storage modules, and routers, and are respectively connected to the dominant controller. One root switch is connected to multiple leaf switches.

[0078] The process formula (1) of updating the Q value by value iteration of the deep Q network is:

[0079]

[0080] The above formula (1) converges to an optimal Q value, the target Q value y t It can be expressed by formula (2):

[0081]

[0082] Define the loss function Y(θ t ) back propagates to update the deep Q network parameters θ t , as shown in formula (3):

[0083] Y(θ t )=[y t -Q(S t ,a|θ t+1 )] 2 (3).

[0084] The campus management system is the core component of the smart campus and is responsible for managing the school's resources and operating status. It provides a basis for school decision-making through data collection, analysis and processing, helping schools to achieve effective allocation and full utilization of resources. The campus management system provides a basis for school decision-making through data collection, analysis and processing, helping schools to achieve refined management.

[0085] The security monitoring module is an important part of the smart campus, including video surveillance, access control, etc. to ensure campus safety, responsible for monitoring the various operating conditions of the school, and providing timely information support for school management. Through various sensors and cameras installed on the campus, the Internet of Things technology can realize real-time monitoring of the campus environment. When any abnormal situation is detected, such as illegal intrusion, fire, leakage, etc., the system will automatically trigger an alarm and quickly pass the information to the management personnel through the Internet of Things device so that timely action can be taken. In addition to basic monitoring functions, the Internet of Things platform can also perform data analysis to help school managers better understand the campus security situation. This includes analyzing behavioral patterns in surveillance videos and using big data technology to predict potential security risks so as to respond in advance. The smart campus system supported by the Internet of Things technology is not limited to traditional video surveillance, but also includes energy management, fire management and other aspects to form a comprehensive security protection system. This comprehensive security management platform can effectively improve the overall security level of the campus

[0086] The smart campus system is a comprehensive information platform that optimizes campus resource allocation and manages the education process by integrating modern information technologies such as the Internet of Things, big data, and cloud computing, thereby providing teachers and students with an efficient, convenient, and safe learning and working environment.

[0087] The information promotion module provides various services to the public, showcases the campus style, and conveys campus dynamics. Internally, it integrates various business application systems and data resources on the smart campus to provide unified services for various users.

[0088] The cloud management module is mainly used to manage the relevant information of students, teachers, classes, and grades within the school, and thus supports the functions of teacher and student information recording and statistics.

[0089] The teaching management module is a platform that integrates academic management, course scheduling, online learning, performance analysis and other functions, allowing teachers to conduct teaching activities efficiently and conveniently, while also supporting students' independent learning.

[0090] In order to achieve instant communication between home and school, it is convenient for teachers and parents to understand and grasp the students' life and learning situation anytime and anywhere. The smart campus system also has student performance management, student attendance management and teacher schedule management. It can combine objective data and subjective evaluation to assess students' moral, intellectual, physical, aesthetic and labor qualities, and obtain the results of students' comprehensive quality scores, and rank them according to the scores. It assesses teachers' records, classroom teaching, student progress, and teacher honors, and obtains the results of teachers' teaching level scores. Teachers can publish the number and general scope of social practice each semester on the teacher's terminal. After completing the social practice, students can complete the social practice assessment by uploading photos and quotations.

[0091] By analyzing students' learning behaviors, grades, and feedback, big data technology can help educational institutions better understand students' specific needs and thus provide more personalized teaching content and resources.

[0092] Big data analysis can reveal which educational resources are most popular and most effective. This analysis helps schools to rationally allocate limited educational resources, such as teachers, teaching materials, and facilities, to improve resource utilization efficiency.

[0093] The smart campus management system uses the data support provided by big data analysis to help school management formulate more scientific and reasonable education policies, thereby optimizing the allocation of educational resources.

[0094] By analyzing large amounts of data, we can discover and solve the inequality in the allocation of educational resources and ensure that every student has access to high-quality educational resources.

[0095] The smart campus solution integrates data from different departments and sources to achieve data fusion and status visualization, thereby improving the transparency and efficiency of educational resource allocation.

[0096] Through the Internet of Things technology, smart classrooms can connect various devices, such as smart blackboards, smart podiums, etc., to achieve intelligent interconnection and control between devices.

[0097] One-card management: Through online balance inquiry, online recharge and other functions, students and faculty can conveniently carry out daily consumption, such as dining and shopping, without carrying cash or multiple cards.

[0098] The teacher terminal or parent terminal uses devices such as smartphones and tablets as well as intelligent sensing technology (such as biometrics) to simplify the process of teaching, management and home-school interaction, making operations more convenient and efficient.

[0099] The smart campus system can also automate logistical services, such as canteen management and cleaning services, to reduce manual intervention and improve service efficiency.

[0100] In the SDN multi-controller network operation scenario, due to the time-varying nature of traffic, the state of the SDN network is random and the number is close to infinite. Traditional reinforcement learning algorithms such as Q learning cannot directly handle this. Deep Q-Network (DQN) can handle infinite input states and effectively generate control behaviors of the target system, so it can handle complex network states in SDN networks well. Figure 3 As shown, the input is the system state S t , is an m×n two-dimensional matrix, which outputs the Q values ​​of m×n actions through the convolution layer, Flatten layer, fully connected layer and output layer.

[0101] DQN introduces the approximate function Q(S t ,a|θ t )(a∈A) represents the input system state S at time t t The output obtained is t Denotes the parameters of DQN at time t. The process of DQN fitting the action value function and updating the Q value through value iteration can be expressed by the formula.

[0102]

[0103] The above formula (1) can converge to an optimal Q value, the target Q value y t It can be expressed by formula (2).

[0104]

[0105] The processing method of deep Q network is as follows Figure 3 As shown in, it shows the process of DQN fitting the action value function to obtain the Q value. Figure 4 In the SDN data center network, enter the system status S t , after DQN processing, the Q values ​​of all migration actions (a∈A) in this system state are obtained.

[0106] In order to optimize the neural network parameters and make the value obtained by the approximate function infinitely close to the Q value, the loss function Y(θ t ) back propagates to update the deep Q network parameters θ t , as shown in formula (3).

[0107] Y(θ t )=[y t -Q(S t ,a|θ t+1 )] 2 (3)

[0108] The smart campus system also supports mobile access, through WeChat mini-programs and other forms, allowing teachers and students to access the system anytime and anywhere without downloading and installing additional applications.

[0109] Embodiment 1:

[0110] During the peak period of student course selection, the network management process of the smart campus system works as follows:

[0111] First, the intelligent decision-making module predicts the upcoming course selection traffic peak by analyzing historical data and real-time monitoring information, and generates a network resource demand report based on the prediction results. The intelligent decision-making module then sends this report to the cloud SDN controller component through the internal communication protocol.

[0112] After receiving the resource demand report from the intelligent decision-making module, the cloud SDN controller component feeds back the analysis results and resource optimization suggestions to the intelligent decision-making module. The intelligent decision-making module formulates a specific network configuration strategy based on the received information.

[0113] Next, the intelligent decision module sends a resource pre-allocation instruction directly to the SDN controller through the northbound interface. The instruction includes parameters such as pre-allocated bandwidth and number of server connections. Based on these parameters, the SDN controller sends flow table entries to the network infrastructure through the southbound interface to ensure the priority and availability of resources required by the course selection system.

[0114] At the same time, the intelligent decision module formulates network micro-segmentation strategies based on the recommendations of the cloud SDN controller component and sends these strategies to the SDN controller through the northbound interface. The SDN controller executes these strategies and creates corresponding network partitions to isolate different services (such as course query, course selection, database access) and improve security.

[0115] When the course selection starts, the intelligent decision module continuously monitors the network status and sends real-time traffic data, server load and other information to the cloud SDN controller component for analysis. The cloud SDN controller component sends the analysis results and suggestions back to the intelligent decision module, which dynamically adjusts the micro-segmentation strategy accordingly.

[0116] If it is necessary to adjust resource quotas or optimize traffic paths, the intelligent decision module directly sends adjustment instructions to the SDN controller. After receiving the adjustment instructions, the SDN controller updates the flow table entries in real time, adjusts network routing and resource allocation, and ensures the efficient operation of the course selection system.

[0117] During the entire process, if the intelligent decision-making module detects abnormal behavior or security threats, it will immediately send an alarm message to the SDN controller and trigger the corresponding security response mechanism, such as isolating the attack source or updating the security policy to protect the security of the course selection system.

[0118] Through this adjustment, the intelligent decision-making module becomes the core component that directly controls the SDN controller, while the cloud SDN controller component plays the role of analysis and suggestion, assisting the intelligent decision-making module to make more accurate network management decisions.

[0119] Embodiment 2:

[0120] In the smart classroom management application, the network management process of the smart campus system is as follows:

[0121] First, the intelligent decision module collects environmental data inside and outside the classroom, equipment status information, and teacher and student behavior patterns through connected sensors and monitoring devices. Based on this data, the intelligent decision module predicts the resource usage demand in the classroom and generates a resource optimization report. The intelligent decision module then sends this report to the cloud SDN controller component through an internal communication protocol.

[0122] After receiving the report from the intelligent decision module, the cloud SDN controller component analyzes the data and provides optimization suggestions on network resource allocation, classroom equipment control, and energy management. Based on these suggestions, the intelligent decision module formulates specific network configuration and classroom management strategies.

[0123] Then, the intelligent decision module sends classroom network configuration instructions directly to the SDN controller through the northbound interface, including wireless access point (AP) power adjustment, network bandwidth allocation, and virtual local area network (VLAN) settings. Based on these instructions, the SDN controller sends corresponding configurations to the network infrastructure through the southbound interface to ensure the reasonable allocation of network resources in the classroom.

[0124] During class, the intelligent decision module dynamically adjusts the classroom's environmental control systems, such as lighting, temperature, and projection equipment, based on the analysis results of the cloud SDN controller component. For example, when it detects that the classroom usage rate has dropped, the intelligent decision module will instruct the SDN controller to reduce the power of some APs to save energy.

[0125] At the same time, the intelligent decision-making module will adjust the network service quality (QoS) in real time through the SDN controller according to real-time activities in the classroom, such as video playback or online interaction, to ensure that key applications have sufficient network resources.

[0126] If a technical failure or security incident occurs in the classroom, the intelligent decision-making module will immediately receive an alarm from the cloud SDN controller component and send instructions to the SDN controller, such as resetting the faulty device, restricting suspicious network traffic, or changing network access rights, to protect the technical environment and the safety of teachers and students in the classroom.

[0127] Through this intelligent classroom management, the intelligent decision-making module becomes the core component that directly controls the SDN controller, while the cloud SDN controller component provides data analysis and policy recommendations, jointly ensuring the stability of the network environment and efficient use of resources in the smart classroom, and providing teachers and students with a comfortable and safe teaching environment.

[0128] The technical solution of the smart campus system of the present invention may involve the use of software, but the software used is the most commonly used software by technical personnel in this field, and is not within the scope of protection of the claims of this patent application.

[0129] The present invention and its embodiments are described above, and such description is not restrictive. What is shown in the full text is only one of the embodiments of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A smart campus system based on SDN, characterized in that: include: Application layer, the application layer includes a campus management module responsible for managing various campus resources and operating status, a teaching management module responsible for teaching management, course arrangement, online learning, and performance analysis, a security monitoring module responsible for ensuring campus safety, an intelligent service module that provides intelligent services to improve campus life, an information promotion module that provides various services to the public, displays campus style or conveys campus dynamics, and a cloud management module for managing students, teachers, classes, and grades within the school; Management layer, which includes a leading controller for network structure configuration, control management and data information, and an SDN network control module for distributed data storage, analysis, processing and connection; Resource layer, the resource layer is used to collect data and analyze the reinforcement data to obtain useful information and input it to the management layer. The resource layer includes a resource data collection module, a resource utilization analysis module, an intelligent decision-making module, a real-time performance monitoring module, an SDN controller and a reinforcement learning module. The resource data collection module, the resource utilization analysis module and the intelligent decision-making module process and analyze the deep Q network through the reinforcement learning module and input it to the SDN controller; The application layer is connected to the cloud platform, and the cloud platform includes a cloud SDN controller component. The cloud SDN controller component serves as a network management component of the cloud platform and is responsible for managing the network resource configuration of the cloud platform according to the cloud application load and resource usage on the cloud platform; The cloud SDN controller component is connected to the intelligent decision module, and the information sent to the intelligent decision module includes network status, performance data, configuration changes, event notifications, resource reports and policy execution results to support the intelligent decision module in network optimization and management; The SDN controller manages and configures the network infrastructure, and sends local application load conditions, network status, performance data, configuration changes, resource utilization, security events, traffic statistics, and basic equipment status information to the receiving intelligent decision module; The intelligent decision-making module makes optimization decisions in real time according to the information sent by the cloud SDN controller component, the network and the SDN controller, and sends network configuration instructions, optimization strategies, resource allocation, policy updates, monitoring instructions, fault responses and compliance requirements to the cloud SDN controller component. The information sent to the SDN controller mainly includes network status analysis results, network micro-segmentation strategies, optimization strategy recommendations, resource scheduling instructions, policy update requests and performance adjustment requirements; The application layer includes local applications and cloud applications. Local applications run on local physical servers, and cloud applications run on cloud platforms, which together constitute the application services of the smart campus. Local applications are connected to the SDN controller.

2. According to claim 1, a smart campus system based on SDN is characterized in that: The SDN network control module includes a northbound interface, a westbound interface, an eastbound interface and a southbound interface, wherein the northbound interface is connected to the application layer, the westbound interface is connected to the leading controller, the southbound interface is connected to the resource layer, and the eastbound interface is connected to the external network.

3. According to claim 2, a smart campus system based on SDN is characterized in that: The external network interactively connects to the cloud platform management module, and the cloud platform management module is connected to the teacher terminal and the parent terminal to achieve instant communication between home and school.

4. According to the SDN-based smart campus system of claim 1, it is characterized by: The resource data collection module is used to collect campus teacher and student data and SDN network resource usage data.

5. The SDN-based smart campus system according to claim 1 is characterized in that: The resource utilization analysis module analyzes resource data and resource utilization efficiency through the big data technology of the reinforcement learning module, and provides resource optimization suggestions based on the resource analysis results to improve resource utilization efficiency and network performance.

6. The SDN-based smart campus system according to claim 1 is characterized by: The intelligent decision-making module provides comprehensive SDN network analysis and optimization suggestions through the big data technology of the reinforcement learning module, providing intelligent support for the leading controller.

7. The SDN-based smart campus system according to claim 1 is characterized by: The real-time performance monitoring module monitors the performance of the SDN network in real time and combines with the security monitoring module to predict development trends and plan and provide reference solutions.

8. The SDN-based smart campus system according to claim 1, characterized in that: The campus management module includes dormitory management, classroom management, library management, laboratory management, canteen management, stadium management, school bus management, affairs repair management; the teaching management module includes student performance management, student attendance management, teacher schedule management, electronic homework, comprehensive teacher-student evaluation, and interactive classroom; the security monitoring module includes video monitoring and access control; the intelligent service module includes campus card management, payment management and one-card pass; the information promotion module includes campus announcement management, broadcast management, campus culture display, school portal and creative forum.

9. The SDN-based smart campus system according to claim 1, characterized in that: The dominant controller has multiple leaf switches and multiple root switches that are interactively connected for data transmission. The multiple leaf switches are connected to servers, storage modules, and routers, and are respectively connected to the dominant controller. One root switch is connected to multiple leaf switches.

10. The SDN-based smart campus system according to claim 1, characterized in that: The process formula (1) of iterative updating of Q value of the deep Q network is: The above formula (1) converges to an optimal Q value, the target Q value y t It can be expressed by formula (2): Define the loss function Y(θ t ) back propagates to update the deep Q network parameters θ t , as shown in formula (3): Y(θ t )=[y t -Q(S t ,a|θ t+1 )] 2 (3)。

Images