Online task scheduling and English online cooperative learning based on 5G mobile communication network

The task density of the data processing platform is increasing, and effective online task scheduling directly determines the business flexibility of the data processing platform. This article starts with the remarkable dynamic characteristics of 5G cellular networks, creates an adaptive environment to optimize online task scheduling, and designs the workload characteristics of data processing and computing tasks. On this basis, based on the 5G mobile communication network programming model and the operating and functional principles of its supporting system, the actual structure and field of online task scheduling work templates have been developed and designed. In addition, this article is developing a technology-based, non-intrusive online task scheduling program that can perform detailed real-time detection of the actual implementation of online task scheduling. In this paper, 5G cellular network is used to further reduce the service cache location of online content, and collaborative English learning and deployment at the edge of the network closer to the end user can further reduce network delay, which is important for improving mobile network communication and improving the efficiency of network content distribution. This article creates a model for online collaborative learning of college English on a 5G cellular network and analyzes the data based on experiments with comparative models to improve their self-confidence and interpersonal skills, and these skills can help improve students' language skills.


Introduction
5G communication has brought much convenience to people's life (Kabalci 2019). However, it can be seen from the growing communication and network demands that the existing demands have exerted significant pressure on the network infrastructure (David et al. 2015). The original network infrastructure can no longer meet the actual needs of the current development, so it needs to be optimized (Israr et al. 2021). In order to provide mobile users with a better experience, it is essential to actively carry out and promote research on fifth-generation mobile technologies (Lee et al. 2019). Among them, the use of cache and computing resources in 5G mobile communication networks is seen as an important way to reduce network transmission delays and content redundancy and improve content distribution efficiency and network computing processing capabilities (Lee et al. 2020). 5G core network content caching is seen as an important way to improve content distribution efficiency and end-user service quality. In this article, 5G cellular network is applied to online task scheduling and English online cooperative learning (Li 2021). Through the collaborative learning of the 5G core network, students can use a wide range of network resources to request the information they need, which facilitates the input and output of language information. In the context of the 5G core network, students can realize instant communication through video, audio, QQ and other technologies that can be saved and played back. In order for peers to understand their intentions, they choose the easiest path to understand, which has a positive impact on students' language acquisition. In addition to English online cooperative learning, this article also uses 5G cellular network for online task scheduling. This paper makes a comprehensive and in-depth analysis of the characteristics of data processing technology and the execution process of computing task scheduling, and proposes a 5G core network system. The system solution optimizes online task scheduling on two levels of vertical processing and horizontal multitasking of a single computing series of tasks. Parallel scheduling of computing tasks ultimately achieves the goal of improving computing platform resource utilization and increasing platform business flexibility. Through group communication, students can reflect on their own misunderstandings and compare the similarities and differences between their opinions. Through in this paper, the results can be seen that the practice of cooperative learning in English online can be very good help bridge the gap between relatively unfamiliar students, can promote students' communication, and can improve the level of English in the process of communication, enhance interpersonal confidence and practical English ability in the process of communication with people, promote the students' comprehensive development.

Related work
Assuming that based on the significant dynamic characteristics of big data processing, in order to build an adaptive system to optimize resource allocation, a concept of large-scale computing tasks and the operation and load characteristics of computing tasks are proposed (Baresi et al. 2019). Taking data processing as the starting point, on the basis of the operation and functional principle of the emerging big data processing system programming model and its auxiliary system, this paper designs and constructs the structure and composition of the arithmetic problem outline in detail (Casado and Younas 2015). In addition, a non-intrusive dynamic test program based on technology is being developed to realize detailed real-time recognition of the actual execution of computing tasks and generate specific running profile values for computing tasks (Ullah et al. 2022;Kos et al. 2015). The literature provides the current overview of computing problems. Starting from the sequential execution level of vertical individual computing tasks, this paper proposes a self-optimization method of adaptive dynamic resource allocation, namely performance optimization, performance prediction, review and performance methods, and constructs the corresponding computing models in turn. This paper summarizes the operation of computing tasks and the prediction model of executing computing tasks. Based on the level of multi-task concurrent scheduling, an adaptive and resource-saving dynamic concurrent scheduling method is proposed, and a scheduler prototype is designed and developed. The scheduler innovatively solves the performance requirements of multiple users for various computing tasks in the process of multi-task scheduling. In the face of multiple computing tasks that appear dynamically and randomly, it uses the latest state matrix of resource allocation detection system resource utilization to accept the utility scoring model of computing tasks based on user quality requirements to execute computing tasks. It takes maximizing the overall value of computing task efficiency as the scheduling goal, and constantly updates the resource scheduling and computing task allocation on each processor node, so as to meet the win-win situation of multi-user computing platform tasks, meet the quality requirements, and improve the overall utilization of platform resources. The literature has shown that RDS scheduler can dynamically adjust the allocation of platform resources across multiple concurrent computing tasks, and its performance is better than Hadoop system (Lai et al. 2014). It can complete the time target relaxation and target compression of computing tasks. In contrast, the system can reduce the time needed to complete the computing task (Chen et al. 2013). In this paper, a white box analysis method is proposed to predict the performance of computing tasks based on the known work profiles and assumptions of computing tasks, and a black box method is proposed to perform comprehensive evaluation based on the training decision tree (She et al. 2017). This paper simulates, predicts and evaluates the performance of computing tasks. The computational performance optimization model uses subspace decomposition and recursive random search technology to search the solution space of huge multi-dimensional resource plan efficiently, and makes optimization comparison according to user optimization objectives and corresponding resource plan constraints (Zhang 2021). 3 Online task scheduling of 5G mobile communication network

5G mobile communication network
The 5G cellular network architecture is developed using new technologies such as SDN and NFV to support efficient data connections and services. The system architecture introduces many new features such as modular network functions, separation of control and forwarding functions, service-oriented and IT-oriented interfaces, and enhanced capabilities, and is open in compliance with the trend of 5G network flexibility, agility and openness. In particular, the control plane of the system architecture is based on the service-oriented architecture of the IT system, that is, the network functions and interface interactions of the control plane are implemented using service-based design ideas. Figure 1 depicts a 5G core network architecture that combines network slicing and distributed caching. The 5G network partition abstracts the resources of the physical network infrastructure and divides them into virtual networks to provide various network services and functions to the outside world. For example, for smart phone services, network operators can divide physical network infrastructure resources into cellular network segments and provide cellular data and voice services, which is a high-speed, low-latency segment. For network services, network operators can divide physical network infrastructure resources into car networking slices with high mobility and low latency characteristics.
The cost formula of nodes obtained by combining Fig. 1 is as follows: The price P that the entire network segment must pay to the network infrastructure provider can be expressed as formula (2): The power consumption cost of cached content is mainly related to the power consumption caused by cached and updated content. Therefore, in this article, a proportional model of energy consumption is established, namely formula (3): The energy cost of cached content can be expressed as formula (4): The power consumption cost of content response is mainly the power consumption generated by responding to and serving users' content requests. The energy consumption cost can be expressed as formula (5): Including the power consumption cost of content response per unit of cache capacity, therefore, the total cost of energy consumption can be written as formula (6): In order to avoid overloading the cache host, the number of content responses is limited, so the energy cost can be expressed as formula (7): The function of the total revenue of the network infrastructure provider can be expressed as formula (8): The total income can be rewritten as formula (9): Now, the task of maximizing total income is shown in formula (10): What happens to the revenue of the network infrastructure as the number of cache nodes increases? Fig. 2 illustrates this.
What happens to network infrastructure revenue as the number of network slices increases? Fig. 3 illustrates this.
In a cluster of small cell base stations, the requested part of the file-encoded data packet is jointly transmitted. Therefore, the joint file transfer rate of the file G requested by the user can be expressed as formula (11): The response rate of users requesting files can be expressed as formula (12): The average cooperative bit rate can be written as formula (13): The average response number can be expressed as formula (14): Obviously, caching files on a small cell base station consumes power, and the power consumption of caching usually depends on the number of cached files. Therefore, this article creates a power scaling model as the cache power model. The cache power model can be expressed as formula (15): It is more efficient to request the file content cached by the neighboring small cell base station than the macro-base station. Therefore, this paper proposes a cooperative caching and distribution mechanism to make full use of the file content cached in the small cell cluster to reduce content transmission and downloading. Among them, the total transmission energy consumption can be expressed as formula (16): If the file cannot be recovered after co-transmission between small cell base stations, the macro-cell base station must send the remaining encoded data packets to the local small cell base station. Therefore, the energy consumption model of bus transmission can be expressed as formula (17): The total energy consumption model can be expressed as formula (18):

Online task scheduling
For a long time, many scientists at home and abroad have made unremitting efforts in the simultaneous multi-tasking research and achieved many fruitful results. However, there is no multi-task concurrent scheduling algorithm that can be applied to all fields and achieve the best scheduling effect at the same time. For a given arithmetic problem and its related operation overview, the following formula can be used to obtain a performance prediction model for arithmetic problems, such as formula (19): Taking into account the cost statistics field of the current configuration file of the calculation task in the source cluster, it is estimated that the model cluster of the cost statistics field value of the virtual work configuration file of the calculation task after the estimated cluster resource is switched to the target is formula (20): In order to provide a concise formula and avoid using conditions as much as possible, this article has made the following definitions and initializations, such as formula (21): Read the input block in formula (22) and formula (23): Then, the time required to complete the operation, such as formula (24): The time required to complete the operation, such as formula (25): Each scheduling algorithm has its own capabilities and some limitations. Set configuration parameters for cluster properties, including the number of nodes in the cluster, the number of allocated and pruned task slots that each node can accommodate, available memory, and network settings for each task to run. They are the most important attributes of cluster resources that will affect the performance of computing tasks.
Cost status information is the collection and recording of the cost of performing calculations at various stages in the entire process of completing a calculation task. It reflects the execution time of each segment when the data flows through each calculation micro-stage in turn, as shown in Table 1.
Data flow statistics are calculated results obtained after statistical analysis of the data flow status information of each subtask recorded during the execution of the process. With the help of statistics, the data flow status information of each subtask can be further analyzed, and the overall data during the entire execution process can be summarized. The development and development of the stream. For example, according to each input record, average the total number of output records produced by each subproblem, and average the compression rate of each output Online task scheduling and English online cooperative learning based on 5G mobile communication… 7609 record of the sub-problem. This article assumes that these data attributes do not change when the same arithmetic problem is executed multiple times until the data distribution in the data set changes significantly. Table 2 lists all the data flow statistics fields in the execution configuration file. Statistical cost information is the calculation result obtained after statistical analysis of the cost status information of each data processing obtained during the execution of the calculation task. Cost statistics show the time required to complete the calculation task from a general perspective, such as from a distributed file system. The average time to read the record or the average time to execute the function for each input record. It is generally considered that this time is fixed and used for different executions of computing tasks, provided that the resources received from the cluster nodes do not change. Table 3 lists all the cost statistics fields included in the current summary.
At this point, run the calculation task again for two different parameter values, and use the working state detector to track and monitor the original data and collect and generate the work profile information. The time field value of each refinement level in Table 4 is before and after

English online cooperative learning form
Cooperative learning, also known as collaborative learning, public learning or group cooperative learning, is based on heterogeneous learning groups, which systematically promotes learning by the interaction of dynamic learning elements, and takes team performance as the evaluation standard. To move toward collaborative learning activities is one of the most widely used face-to-face learning organizations in the world. Cooperative learning has rich research foundation in terms of itself. Roseth et al. concluded that collaborative learning makes students ''more likely to succeed than independent learning, and more likely to build more positive relationships between students, thus achieving better results and more positive partnerships.'' Cooperative learning is a purposeful activity, which emphasizes the interdependence and interaction among group members. Its specific learning strategies include four aspects.
1. Select the topic, clarify the content and learning objectives; 2. The study group is divided in a meaningful way and the responsibilities of the group members are explained, usually according to the principle of ''homogeneity among groups and heterogeneity within groups.'' If  there are differences in gender, academic achievement, skills, personality, etc., the cooperative group should include 4-5 students; 3. Carry out cooperative learning activities such as thematic discussion, art performance, research, etc., and the team members study face to face, and then solve their own learning problems after division of labor, then convey the learning results to other members, and finally combine the learning results; 4. Submit the results and scores of the survey, each group shall show the learning results to the whole class.
Therefore, cooperative learning is a teaching method which takes the exploration and application of interpersonal relationship in face-to-face learning. At present, the application research of College English teaching mainly focuses on the same school in traditional classroom, and few of them are combined. How to make full use of the two teaching methods to provide convenient and flexible communication channels for students is still a challenge to be solved. This paper applies cooperative learning strategy to online learning in English classroom and combines online learning with face-to-face learning to improve the learning experience and student efficiency in this learning mode.

Characteristics of College English cooperative learning under the network environment
The development of network technology has broken the characteristics of traditional English classroom limited by time and space. Students can not only participate in online teaching at the same time, but also conduct real-time cooperation and dialogue in the form of online data exchange. If students miss the teaching time, they can also learn the content produced in this class by watching the class record and express their views in the form of comments. They can also communicate with other members in the message board with time difference, so as to promote the contact between students. With recording, the audience can hear the information over and over again. This kind of repetition is a prerequisite for learning the language from short-term understanding to long-term acquiring the ability to repeat practice through video and recording.

Design of College English Cooperative learning model based on 5G mobile communication network
The cooperative learning model based on 5G cellular network is based on Krashen's speech input hypothesis and Swain's speech output hypothesis. For Krashen's comprehensible input hypothesis, Swain thinks that the function of comprehensible reasoning is to provide the opportunity to detect real language communication and retrieval in context, which is the prerequisite for mastering a second language. Swain believes that text to speech also plays an important role in language acquisition. Students should be given enough opportunities to practice using the language they learn in class. It holds that understanding language input is only a part of language learning, and students should also be able to use the target language, that is, to carry out activities related to speech output. This paper combines the above theories with students' actual learning process and tries to build a coeducational college English model in a mixed environment, as shown in Fig. 4.

Experimental method and description
Statistical analysis of the two stages before and after the students and the data obtained were all analyzed by SPSS11.5. The comparison between the experimental class and the control class adopts independent sampling t test, and the results are shown in Table 5. Table 5 shows that there is no significant difference in the scores of interpersonal relationship between the experimental class and the control class, but the scores of self-confidence, learning interest and pre-exam skill assessment (P [ 0.05). This shows that the initial conditions of the four classes before the experiment are the same, so the experimental class and the control class are comparable.
The experimental results show that collaborative learning based on network environment can help overcome the loneliness of students and enhance their self-confidence and interpersonal ability. Cooperative learning in the network environment uses the network to communicate with each other. Although students can communicate through video, text and audio, they lack the instant and emotional intensity of face-to-face communication. At the same time, when learning together in class, it is difficult and fast to find information, which leads to the lack of depth and breadth of communication between students. Therefore, only combining the two learning environments can overcome the above shortcomings.

Conclusion
Collaborative learning in traditional classrooms means that students conduct group activities within a period of time to achieve common goals or tasks in the classroom. Learning is carried out in the form of group cooperation. Each group member has his own task and accomplishes the overall learning goal through the cooperation of several people. In traditional collaborative classroom learning, students' face-to-face communication, using physical movements, facial expressions and other forms to make language expression more accurate and vivid, allows students to quickly understand their own information. Team members can get emotional improvement in long-term cooperation, so as to make progress together. Collaborative classroom learning, although emotional face-to-face communication and instant communication, can be realized, but the communication content lacks depth and breadth. Students can only communicate on pre-prepared materials and cannot effectively deal with short-term issues such as obtaining new materials. Therefore, this article uses a 5G cellular network to help students learn English online. This article focuses on the 5G core network cache and network partition technology, and proposes an efficient 5G core network cache resource allocation mechanism. Simulation results demonstrate the effectiveness of the proposed method. Through the application of this system, students can use the Internet to collaborate on learning, search for content, and increase the depth and breadth of the topics discussed, and then conduct face-to-face communication with the prepared content in the classroom environment.   Funding This article is sponsored by a higher education reform program titled ''Innovative research of mixed teaching mode of language and literature courses based on output-oriented method'' (2019JSJG277).
Data availability Data will be made available on request.

Declaration
Conflict of interest The authors declare that they have no conflict of interests.
Ethical approval This article does not contain any studies with human participants performed by any of the authors.