System and method for managing file transmission between external network and internal network

Abstract

A system and method for managing file transfer between an external network and an internal network are provided. The method includes the following steps: calculating transfer parameters corresponding to the files; selecting an action based on the current state, a Q-Table, and an ε-greedy policy; providing the transfer parameters and the action to a first network device or a second network device through one-way control; and updating the Q-Table based on a Q-Learning formula using the action, the current state, the updated state, and the target reward.

Claims

1. A system for managing file transfer between an external network and an internal network, comprising: The intelligent traffic management reinforcement learning module includes an action unit, a Q-Table processing unit, a status unit, and an environmental control unit. The system includes an intermediate data exchange module, communicatively connected to the intelligent traffic management reinforcement learning module; a first network device, communicatively connected to the intermediate data exchange module, wherein the first network device corresponds to an external network; and a second network device, communicatively connected to the intermediate data exchange module, wherein the second network device corresponds to an internal network. The state unit calculates transmission parameters corresponding to the file. The action unit selects an action based on the current state, the Q-Table, and the ε-greedy policy. The environment control unit provides the transmission parameters and the action to either the first network device or the second network device via one-way control. The Q-Table processing unit updates the Q-Table based on the Q-Learning formula, utilizing the action, the current state, the updated state, and the target reward. The intermediate data exchange module includes a log unit, through which the environment control unit obtains file information corresponding to the file. The file corresponds to the current state. The file information includes the number of files, file size, and transmission time. The transmission parameters include the transmission window opening time and network bandwidth. The state is represented by a vector, which includes the transmission parameters and the file information.

2. The system as described in claim 1, wherein the intelligent traffic management reinforcement learning module further includes an environment perception unit, wherein the environment perception unit monitors the current network environment status; and the action unit adjusts the ε value of the ε-greedy policy according to the current network environment status.

3. The system as described in claim 1, wherein the environment control unit obtains file information corresponding to an update of the file via the log unit; and the status unit uses the updated file information to obtain an updated status.

4. The system as claimed in claim 1, wherein the intelligent traffic management reinforcement learning module further includes a reward unit, wherein the reward unit uses the current state and the updated state to obtain the target reward.

5. The system as described in claim 1 further includes an external network intermediary data exchange module and external network device resources, wherein the external network intermediary data exchange module includes an external network intermediary receiving / sending unit, an external network intermediary file storage unit, and an external network intermediary southbound API unit, wherein the external network device resources include an external network device receiving / sending unit, an external network device northbound API, MEC resources, server resources, and network device resources, wherein the file corresponds to the MEC resource, the server resource, or the network device resource, wherein the external network device receiving / sending unit transmits the file to the external network intermediary receiving / sending unit; the external network device northbound API transmits a file transmission request to the external network intermediary southbound API unit; after the external network intermediary receiving / sending unit receives the file, the external network intermediary file storage unit stores the file.

6. The system as described in claim 1 further includes an external network intermediary data exchange module and an internal network intermediary data exchange module, wherein the environment control unit provides the transmission parameters to the first network device through the one-way control; the first network device enables the connection between the external network intermediary data exchange module and the intermediary data exchange module based on the transmission parameters; when the first network device enables the connection between the external network intermediary data exchange module and the intermediary data exchange module, the second network device disconnects the connection between the intermediary data exchange module and the internal network intermediary data exchange module.

7. The system as described in claim 6, wherein the first network device disconnects the external network intermediate data exchange module and the connection between the intermediate data exchange modules based on the transmission parameters; the environment control unit provides the transmission parameters to the second network device through the one-way control; the second network device enables the connection between the intermediate data exchange module and the internal network intermediate data exchange module based on the transmission parameters; when the second network device enables the connection between the intermediate data exchange module and the internal network intermediate data exchange module, the first network device disconnects the external network intermediate data exchange module and the connection between the intermediate data exchange modules.

8. A method for managing file transfer between an external network and an internal network, suitable for a system including an intelligent traffic management reinforcement learning module, an intermediary data exchange module, a first network device, and a second network device, wherein the intelligent traffic management reinforcement learning module includes an action unit, a Q-Table processing unit, a state unit, and an environment control unit, wherein the first network device corresponds to the external network, and the second network device corresponds to the internal network, wherein the method includes the following steps: the state unit calculates transmission parameters corresponding to the file; the action unit selects an action based on the current state, the Q-Table, and an ε-greedy policy; the environment control unit provides the transmission parameters and the action to the first network device through one-way control. The method includes either a backup or a second network device; and the Q-Table processing unit updates the Q-Table based on the Q-Learning formula using the action, the current state, the updated state, and the target reward, wherein the intermediary data exchange module includes a log unit, and the method further includes the following steps: the environment control unit obtains file information corresponding to the file via the log unit; wherein the file corresponds to the current state; wherein the file information includes the number of files, the file size, and the transmission time; wherein the transmission parameters include the transmission window opening time and the network bandwidth; wherein the state is represented by a vector, wherein the vector includes the transmission parameters and the file information.

Images