Traditional network data fluctuation controllers often have large amount of computation and complex operation, which results in the shortcomings of the controllers themselves, such as short holding time of network non-fluctuating data and large communication error. Therefore, a more intelligent and practical network data fluctuation controller is designed. The designed controller constructs the network data fluctuation control model, gives the optimal control scheme of network data fluctuation, and extracts it using AT91SAM9260 microprocessor. AT91SAM9260 microprocessor inserts the optimal control scheme into the corresponding nodes of network initial data, aggregates the network with small data fluctuation and less interference parameters, and distributes the control to the opposite type of network. The experimental results show that the designed controller has long holding time for non-fluctuating data and minimal communication error. Compared with the traditional controller, its performance is more excellent and its application field is more extensive. With the continuous development of electronic technology, network has become an indispensable communication tool in people’s lives, and the amount of network data is also increasing. The data loss and communication errors caused by network data fluctuation have gradually led to some major security risks, and the network data fluctuation controller has thus emerged [1?3]. Traditional network data fluctuation controllers often have the disadvantages of large computation and complex operation, such as short holding time of non-fluctuating data and large communication error. Therefore, in recent years, the designed network data fluctuation controller has been developing towards a more intelligent and practical direction, which makes the performance of the controller more excellent and the application field more extensive. In the communication of network data [4?6], the data is usually transmitted to the corresponding user nodes in the state of data packets. Whether the communication work can be completed smoothly or not is closely related to the network bandwidth. If the packet capacity exceeds the maximum load that the network bandwidth can bear, the network data in the packet will fluctuate to varying degrees. At this time, the designed network data fluctuation controller needs to control the network data by modeling. The purpose is to divide the data packets beyond the network bandwidth load into several optimal data packets, so as to increase the holding time of the controller for non-volatile data and reduce the communication error. Figure 1 shows the structure of the network data fluctuation control model. As can be seen from Figure 1, the data packet with the best communication capacity is divided by the network data fluctuation control model, and the communication paths are different. At this time, the network data received by the designed network data fluctuation controller may be reversed in sequence. Therefore, a buffer device is added between the control model and the controller.
The buffer device will store and sort the user network data within the specified task time, and then transmit it to the designed network data wave.
The control is carried out in the dynamic controller. Through the analysis of the network data fluctuation control model in Formula 4, the optimal control scheme of the network data fluctuation controller can be given, which can reduce the amount of operation of the controller and simplify the steps of using the controller. By changing the parameters of the control model and making comparative analysis, the factors affecting the performance of the controller can be further found, which is convenient for the later maintenance of the controller.
According to the network data fluctuation control model, the overall design of the network data fluctuation controller is given, as shown in Figure 2. As shown in Figure 2, the designed network data fluctuation controller consists of input module, control module, storage module and visualization module. Controller aggregation control design can know the specific fluctuation of network data according to the fluctuation control model of network data. In order to prolong the holding time of the designed network data fluctuation controller for the non-fluctuating data, the controller will automatically select the aggregation control mode to control the controlled network when the network data fluctuation is small and the disturbance parameters are small. In aggregation control, the best communication capacity data packet in the controlled network will be transferred to the control module to form a mapping topology control structure, as shown in Figure 3. This structure will uniformly formulate control details and generally smooth the fluctuation of network data. With the help of the network data fluctuation control model, the data in the optimal communication capacity data packets in the aggregation control can interact freely, thermostatic element which makes the application field of the controller more extensive and conforms to the original design intention.
In order to reduce the communication error of the designed network data fluctuation controller, the controller will choose the distributed control mode to control the controlled network when the network data fluctuation is large and the interference parameters are large, as shown in Figure 4. As shown in Figure 4, distributed control corresponds to aggregation control, that is, the control module will start the next control work only after the control work of the former optimal communication capacity data packet is finished and the control results are transmitted to the next optimal communication capacity data packet. Distributed control has a loop-like transverse data flow topology, and the best communication capacity data packet can receive the control results of the best communication capacity data packet in front, and adjust the next control scheme according to the success of the control results.
The control effect of this control mode is very good, but the control efficiency is not as good as that of aggregate control. Therefore, in practical application, the designed network data fluctuation controller usually uses aggregate control and distributed control interactively, so that the performance of the controller is better and the application field is wider. In this experiment, OPNET software is used to simulate the controller, the network data fluctuation controller based on non-linear control and the network data fluctuation controller based on dwell time method.
OPNET software simulates the performance and parameters of the controlled network, and debugs the control model established by the three controllers, and outputs the simulation results of non-fluctuating data holding time and communication errors of the controllers. The fluctuation trajectory of network data given by OPNET software can directly reflect the holding time of non-fluctuating data by the controller, as shown in Fig. 5-7. The faster the trajectory of network data fluctuation is stabilized, the better the stability of the control effect of the controller is proved, and the longer the controller holds the non-fluctuating data.
From Fig. 5 to 7, it is obvious that the trajectory of network data fluctuation in the controller is the fastest. When the network fluctuation data is transmitted to the seventeenth network fluctuation node, there is no network data fluctuation in the controlled network, which proves that the controller has a long holding time for non-fluctuation data.
Table 1 shows the output results of communication errors of three kinds of controllers. It can be seen that the communication errors of the controllers in this paper are very small, not more than 0.15 cm, which proves that the application fields of the controllers are more extensive and can better achieve the original design purpose.
This paper designs a more intelligent and practical network data fluctuation controller, which consists of input module, control module, storage module and visualization module. The input module includes network data fluctuation control model and network initial data. Control module is the most important module in the controller, and its core is AT91SAM9260 microprocessor. Storage module includes analog-to-digital conversion circuit, storage device and communication interface. The visualization module includes synchronous dynamic random access memory (SDRAM) and visualization plug-in.
The experimental results show that the designed controller has a long holding time for non-fluctuating data and a minimal communication error.