TW202145807A - Mobile edge computing platform and packet offloading method based on mobile edge computing - Google Patents

Abstract

A mobile edge computing (MEC) platform and a packet offloading method based on MEC are provided. In the method, an uplink traffic packet from the mobile network is received. An endpoint identification for transmitting user data via tunneling protocol by the user equipment (UE) is obtained from the signaling packet. Whether to forward the uplink traffic packet, which is designed to be transmitted to the core network before, to an internal server is determined according to the endpoint identification and a destination network address of the uplink traffic packet. Accordingly, the efficiency of local offloading can be improved.

Description

Mobile edge computing platform and packet offloading method based on mobile edge computing

The present invention relates to a mobile edge computing (MEC) technology, and more particularly, to a mobile edge computing platform and a packet offloading method based on the mobile edge computing.

Providing the user local offload function in the mobile edge computing architecture can provide mobile user-specific services more quickly, and can also reduce the burden on the core network. Mobile edge computing usually uses multiple fields such as the source Internet Protocol (IP) address, destination IP address, source port (port), and destination port of the packet to confirm whether it needs local Uninstalled package. Excessive field analysis will increase the processing time of the packets, thereby affecting the performance.

In view of this, embodiments of the present invention provide a mobile edge computing platform and a packet offloading method based on mobile edge computing, which can check the outer layer information of the packet to confirm whether the packet needs to be offloaded locally, thereby improving processing efficiency.

The mobile edge computing platform of the embodiment of the present invention includes (but is not limited to) a communication transceiver, a storage and a processor. The communication transceiver receives uplink traffic packets from the mobile network. The memory records several modules. The processor is coupled to the communication transceiver and the storage, and accesses and executes those software modules recorded in the storage. Those software modules include a signal capture analysis module and a local offload judgment module. The signaling extraction and analysis module obtains the endpoint identification code of the user equipment that transmits the user data through the tunneling protocol from the signaling packet, and obtains the user information of the user equipment according to the endpoint identification code. The local offload judging module judges whether to forward the upstream traffic packets originally sent to the core network to the internal server according to the endpoint identification code and the destination network address of the upstream traffic packets.

The mobile edge computing-based packet offloading method according to the embodiment of the present invention includes (but is not limited to) the following steps: receiving an uplink traffic packet from a mobile network. Obtain the endpoint identification code of the user equipment for transmitting user data through the tunnel protocol from the signaling packet. Obtain the user information of the user equipment according to the endpoint identifier. According to the endpoint identification code and the destination network address of the upstream traffic packet, it is judged whether to forward the upstream traffic packet originally sent to the core network to the internal server.

Based on the above, the mobile edge computing platform and the packet unloading method based on the mobile edge computing of the embodiments of the present invention check the endpoint identification code corresponding to the packet to obtain more user information, and further combine the destination address to determine whether to forward the packet to the internal network to offload packets to the local. In this way, not only the packet processing time can be reduced, but also the flexible comparison conditions can be increased, thereby improving the transmission efficiency and the strength of information security protection.

In order to make the above-mentioned features and advantages of the present invention more obvious and easy to understand, the following embodiments are given and described in detail with the accompanying drawings as follows.

FIG. 1 is a block diagram of components of a communication architecture according to an embodiment of the present invention. Please refer to FIG. 1 , the communication architecture includes (but is not limited to) a mobile edge computing (MEC) platform 100 , one or more user equipment (UE) 10 , a mobile network 20 , and an internal server for MEC services 30. The core network 40 and the Internet (Internet) 50.

The user equipment 10 may be a networkable device such as a computer host, a smart phone, a tablet computer, a wearable device, or the like.

The mobile network 20 may be various generations of mobile phones such as Universal Mobile Telecommunications System Terrestrial Radio Access Network (UMTS Terrestrial Radio Access Network, UTRAN), Evolved Universal Terrestrial Radio Access Network (E-UTRAN), etc. Internet standard access network.

The internal server 30 refers to a server related to the internal application service that does not need to go through the core network 40 .

The core network 40 may be a core network of various generations of mobile network standards or a network entity thereof (eg, a mobility management entity (MME), a Serving Gateway (S-GW), or a storage gateway). Access and mobility management function (Access and mobility Management Function, AMF), etc.).

The MEC platform 100 includes (but is not limited to) a communication transceiver 110 , a storage 130 and a processor 150 . The MEC platform 100 may be a gateway device, a routing device, a host computer, a server and other devices between the mobile (access) network 20 and the core network 40 .

The communication transceiver 110 may be a network module, interface card, chip or circuit that supports wired communication such as Ethernet and optical fiber technology, or supports wireless communication such as Wi-Fi and Bluetooth.

The storage 130 may be any type of fixed or removable random access memory (Random Access Memory, RAM), read-only memory (Read-Only Memory, ROM), flash memory (Flash Memory), or similar components or a combination of the above elements. Storage 130 stores code, device configuration, buffered or persistent data, such as signaling capture analysis module 131 (including first and second signaling capture analysis modules 132, 133), net name system (DNS) analysis module 134, local uninstall condition setting module 135, and local uninstall determination module 136 and other software modules or code, user information, information about traffic packets, control signaling, local uninstall condition information, etc. , information and a comparison table, and the detailed operation of each software module will be described in subsequent embodiments.

It should be noted that the aforementioned software modules can be deployed on the same hardware platform, and the communication between the modules is based on a programmed interface. Alternatively, the aforementioned software modules can also be deployed on different hardware platforms, and the communication between the modules is based on an IP-based communication protocol.

The processor 150 is coupled to the communication transceiver 110 and the memory 130, and can be a CPU, a microprocessor, a microcontroller, a Digital Signal Processing (DSP) chip, a Field Programmable Gate Array (Field Programmable Gate) Array, FPGA) and other programmable units or integrated circuits (Integrated Circuit, IC). In this embodiment, the processor 150 is used to execute all operations of the MCE platform 100 , and can access and execute those software modules recorded in the storage 130 .

In order to facilitate the understanding of the operation process of the present invention, a number of embodiments will be given below for detailed description. Hereinafter, the method according to the embodiment of the present invention will be described with reference to each device, element and module in the communication architecture shown in FIG. 1 . Each process of the method can be adjusted according to the implementation situation, and is not limited to this.

FIG. 2 is a flowchart of a packet offloading method based on mobile edge computing according to an embodiment of the present invention. Referring to FIG. 2, the processor 150 receives the uplink traffic packet from the mobile network 20 through the communication transceiver 110 (step S210). Specifically, the uplink traffic packet is a packet transmitted by any user equipment 10 via the base station in the mobile network 20, and may include the data traffic (or user data/user plane) to be transmitted by the user equipment 10 (user-plane data). In some embodiments, the base station encapsulates the packets of the user equipment 10 into a tunneling protocol (eg, GPRS Tunneling Protocol (GTP) or other similar protocol) format (eg, GTP-U packets), enabling mobile edge The computing platform 100 receives the GTP-U packet carrying the data traffic.

It is worth noting that in the case where the mobile edge computing platform 100 is not provided, the uplink traffic packets will be transmitted to the Internet 50 via the core network 40 . In this embodiment of the present invention, the mobile edge computing platform 100 is configured to determine whether the uplink traffic packets need to be offloaded locally (ie, forwarded to the internal server 30 ).

The signaling extraction and analysis module 131 can acquire the endpoint identification code of the user equipment 10 for transmitting the user data via the S1 interface from the signaling packet (step S230 ). Specifically, the endpoint identifier is entrained in the header of the tunnel protocol-related packet and used to distinguish different connections in the tunnel. For example, the GTP-U packet includes a tunnel endpoint identifier (TEID). In addition, the mobile edge computing platform 100 can further identify which user equipment 10 the tunnel belongs to according to the specific endpoint identification code. That is, the user equipment 10 can be distinguished.

Next, the signaling extraction and analysis module 131 can obtain the user information corresponding to the user equipment 10 according to the endpoint identification code (step S250 ). Specifically, since the endpoint identification code can be used to identify the user equipment 10, the signaling capture analysis module 131 can further search for the user information corresponding to the user equipment 10 from the previously stored user data table of the entire network. The network-wide user data table is a collection of user information recorded by the telecommunications company (may be in the form of a database, a comparison table, etc.) of all user equipments 10 . The user information can be a door number, an access point name (APN), a base station network address (ie, the network address of the base station in the mobile network 20, and the user equipment 10 is currently receiving the base station service), user network address, or a combination thereof. The network address is, for example, an IP address, a media access control (Media Access Control, MAC) address, or an identification code in other networks. In addition, unlike the prior art, which needs to search for multiple fields in the packet to obtain user information, the embodiment of the present invention can search for the recorded user information based on the endpoint identification code, which can obviously improve the processing efficiency.

In one embodiment, the signaling capture analysis module 131 can first determine whether the endpoint identifier and its user information conform to the local uninstall permission. If the local offload permission is met, the endpoint identification code is provided to the local offload determination module 136 for the local offload determination module 136 to determine whether to forward the uplink traffic packet. In other embodiments, all mechanisms of the local uninstall determination may be performed by the local uninstall determination module 136 . It should be noted that, the setting of the local uninstall authority will be described in detail in subsequent embodiments.

It should be further noted that, the content in the user information may be changed according to actual needs, which is not limited in the embodiment of the present invention.

Next, the local offload determination module 136 may determine whether to forward the uplink traffic packet originally sent to the core network 40 to the internal server 30 according to the endpoint identification code and the destination network address of the uplink traffic packet (step S270 ). ). Specifically, the local uninstallation determination module 136 can obtain the local uninstallation condition information set in advance (and also modified afterwards) from the local uninstallation condition setting module 135 . This local offload condition information may record the network addresses (for destination network address comparison) that can/need to be offloaded locally (ie, sent to the internal server 30). The local offload determination module 136 compares whether the endpoint identification code and the destination network address corresponding to the uplink traffic packet have been recorded in the local offload condition information. If it is recorded in the local offload condition information, it means that the uplink traffic packet can be offloaded locally, and the local offload determination module 136 forwards the uplink traffic packet to the internal server 30 . On the other hand, if it has not been recorded in the local offload condition information, the local offload determination module 136 still sends the uplink traffic packet to the core network 40 (for example, the Serving Gateway (S-GW) therein. )).

In one embodiment, the network domain address corresponds to a specific time-to-live. The local uninstall determination module 136 can obtain the corresponding time-to-live (the time-to-live of the corresponding domain network address) according to the destination network address, and determine whether the time-to-live has expired, so as to confirm the DNS status. If the time-to-live has not expired, the local offload determination module 136 can remove the tunnel protocol header (even the outer IP header) in the upstream traffic packet, and forward the remaining packets to the internal server 30 . If it has expired, the upstream traffic packet can be sent to the core network 40 .

The establishment of local offload condition information, signaling analysis, DNS resolution, and local offload determination are further described below in sequence.

FIG. 3 is a flowchart of unloading condition setting according to an embodiment of the present invention. Referring to FIG. 3 , the local uninstall condition setting module 135 waits for the setting staff to set user rights for installation information, and the setting staff obtains the user authorization data and enters the user interface provided by the local uninstall condition setting module 135 to modify the user rights for installation information ( Including user information such as door number, APN, base station network address, website address, destination network address, and related to the aforementioned local uninstallation authority) (step S310). It should be noted that all fields are not required, and the level of detail determines the severity of the conditions. The local uninstall condition setting module 135 can store user information such as door number, APN, and base station network address in the local uninstall condition information. For example, Table (1) is an example illustrating local uninstall condition information: Table (1) door number APN Base station IP address 0988654321 CHT1 1.2.3.4 0987653210 CHT2 1.2.4.4 0986543210 AUO 3.2.1.2 The local uninstall condition setting module 135 may set the local uninstall condition information according to the newly added or deleted user information (step S320 ), and further update the local uninstall condition information to the local uninstall condition information of the local uninstall condition setting module 135 (step S330 ) ).

The local uninstall condition setting module 135 can set the local uninstall condition information according to the newly added or deleted website information (step S340 ), and further update the website information to the domain name system analysis module 134 (step S350 , for adding a specific website address) information), or send a deletion instruction to the domain name system analysis module 134 (step S355, for deleting the website information). The website information is, for example, domain names such as www.cht.com, www.chtlearn.com, etc.

The local uninstall condition setting module 135 can set the local uninstall condition information according to the added or deleted destination network address information (step S360 ), and further update the destination network address information to the local uninstall determination module 136 ( Step S370). In addition, the local uninstall condition setting module 135 can perform operations such as deleting, adding, and modifying the installation information provided by the user according to the user authorization data.

FIG. 4 is a flowchart of signaling extraction and analysis according to an embodiment of the present invention. Referring to FIG. 4 , the signaling extraction and analysis module 131 determines the packet type (step S410 ). If one or more pieces of first signaling are received from the core network 40 (the GTP-C packet of the S11 interface will be taken as an example, but not limited to this type) (step S421 ), the first signaling is retrieved The retrieval and analysis module 132 can obtain the corresponding door number, access point name, and user from the packets related to create session (eg, CreateSession Request and Response packets) in the first signaling. network address (steps S424 and S425). Next, the first signaling extraction and analysis module 132 can also obtain the corresponding endpoint identification code, gate number, Access point name, base station network address, and user network address, and store the user information in the network-wide user data table accordingly (steps S426 and S427). For example, Table (2) is an example of user information: Table (2) TEID User IP address door number APN Base station IP address 789456 10.30.1.2 0988654321 CHT1 1.2.3.4 567831 10.20.1.3 0987653210 CHT2 1.2.4.4 In addition, if the first signaling extraction and analysis module 132 receives packets related to delete session (eg, DeleteSession Request and Response packets) from the first signaling (step S422 ), it can automatically The corresponding user information is deleted from the user data table (step S423).

On the other hand, if one or more second signalings from the base station of the mobile network 20 are received (the following will take the S1-AP packet of the S1 interface as an example, but not limited to this type) (step S431), the second signaling acquisition and analysis module 133 can obtain the initial context setup (eg, InitialContext SetupRequest) packet (eg, the InitialContext SetupRequest packet) from the second signaling (step S432), and obtain the user from the packet The endpoint identification code of the device 10 (for example, TEID, and its value is, for example, 789456, 567891, etc., but not limited to this) (step S433), and then through this endpoint identification code to look up the network-wide user data table data corresponding to User information such as APN, door number, user network address, and base station network address (steps S451 and S452). Next, the signaling capture analysis module 131 can compare these combinations (ie, endpoint identifiers) with the local uninstall condition table information (step S453 ) to determine whether the endpoint identifier complies with the local uninstall permission. The signaling capture analysis module 131 can transmit the endpoint identification code to the local uninstall determination module 136 according to the comparison result (in the case of conforming to the local uninstall authority) (step S454 ). Otherwise (not in compliance), the local uninstall determination module 136 does not need to be updated. In addition, if the second signaling is received and the analysis module 133 obtains packets related to the user equipment context release (eg, UEContextRelease packets) from the second signaling (step S434 ), the local uninstall determination module 136 will be notified Delete the user information corresponding to the endpoint identification code (step S435).

FIG. 5 is a flowchart of Domain Name System (DNS) resolution according to an embodiment of the present invention. Referring to FIG. 5 , the left half of the figure is responsible for processing information from the local uninstall condition setting module 135 . If the domain name system analysis module 134 receives the newly added website information (ie, the domain name to be added) (step S514 ), the domain name and the (domain) website address are stored in the local uninstalled domain name. A lookup table (step S515). If the domain name system analysis module 134 receives the information of the website to be deleted (ie, the domain name to be deleted) (step S511 ), it searches the domain name to be deleted in the comparison table of domain name and website address. corresponding network address (step S512 ), and issue a delete instruction to delete the network address to the local uninstall determination module 136 (step S513 ). On the other hand, the right half of the figure is for processing downlink traffic packets from the core network 40 (GTP-U downlink packets are used as an example in the following, but not limited thereto). If the DNS resolution module 134 receives the GTP-U downlink packet to be transmitted to a certain user equipment 10 (step S521 ), it is determined whether the packet is used for querying a network domain in the Domain Name System (DNS). A response packet is received (step S522). If it is a response packet, the domain name system parsing module 134 determines whether the domain name contained in the response packet exists in the domain name table (recorded in the local uninstall condition information) that needs to be uninstalled locally (that is, whether it can be locally uninstalled or not). uninstall) (step S523). If it exists in the domain name table to be uninstalled locally, the domain name system resolution module 134 obtains and returns the corresponding domain network address and its lifetime (eg, TTL (Time To Live)) to the local uninstallation The judging module 136 (for the local uninstall judging module 136 to judge whether to forward the uplink traffic packet according to the network address and the survival time), and store the information such as the network address and the survival time in the network A comparison table between domain names and web addresses (steps S524 and S525). For example, Table (3) is an example of a comparison table of domain names and web addresses: Table (3) domain name IP address www.cht.com 10.20.1.1 www.chtlearn.com 10.20.2.2

FIG. 6 is a flowchart of local uninstall determination according to an embodiment of the present invention. Please refer to Figure 6, the left half of the figure is for processing the data transmitted from each module. If the command and the endpoint identification code thereof are received from the signaling capture analysis module 131 (step S611 ), the local uninstall determination module 136 determines whether the command is a new or deleted command (step S612 ). If the command is added, the local uninstall determination module 136 adds an endpoint identification code that can be uninstalled locally from the local uninstall condition information (step S613 ); if the instruction is deleted, the local uninstall determination module 136 adds the locally uninstalled condition information from the local uninstall condition information. The locally uninstallable endpoint identification code is deleted (step S614).

If the command and its destination network address are received from the local uninstall condition setting module 135 (step S621 ), the local uninstall determination module 136 determines whether the command is a new or deleted command (step S622 ). If the instruction is added, the local uninstall determination module 136 adds a destination network address that can be uninstalled locally from the local uninstall condition information (step S623); if the instruction is deleted, the local uninstall determination module 136 selects the local uninstall condition from the local uninstall condition information (step S623). The destination network address that can be downloaded locally is removed from the information (step S624).

If receiving the command from the domain name system resolution module 134 and its domain network address and time-to-live (step S631 ), the local uninstall determination module 136 determines whether the command is an addition or deletion command (step S632 ). If the instruction is newly added, the local uninstall determination module 136 adds the locally uninstallable domain network URL and its survival time from the local uninstall condition information (step S633); if the instruction is deleted, the local uninstall determination module 136 automatically From the local uninstall condition information, delete the domain network URLs that can be uninstalled locally and their survival time (step S634).

On the other hand, the right half of the figure is for processing the user's GTP-U uplink packet (ie, an example of the uplink traffic packet, but not limited thereto). After receiving the GTP-U uplink packet (step S651 ), the local offload determination module 136 extracts the endpoint identifier from the packet and looks up a table (eg, local offload condition information) to determine whether local offload is required (step S652 ). If local offloading is required, the local offloading judging module 136 compares whether the destination network address recorded in the inner layer of the packet is the network address that requires local offloading (step S653 ), and determines the destination network address Whether the time-to-live has expired (step S655). If not expired, the local offload determination module 136 can remove the outer IP and GTP fields to perform local offload (step S657 ), and forward the remaining packets to the corresponding local offload server (ie, the internal server 30 ) (step S658). On the other hand, if it has expired, the local uninstall determination module 136 may delete the corresponding network address information (step S656 ), and transmit the GTP-U uplink packet to the S-GW of the core network 40 (step S654 ) ) (ie, no local uninstall required).

For the convenience of readers to better understand the spirit of the invention of the embodiments of the present invention, another embodiment is described below:

The first step: the setting personnel set the user permissions for installation. The second step: the local uninstall condition setting module 135 sends the URL information in the setting to the domain name system resolution module 134, sends the destination IP information to the local uninstall determination module 136, and sends the remaining user IP addresses, User information such as door number, base station IP address, APN, etc., is stored in the local uninstall condition setting module 135 and is used for searching by the subsequent signaling extraction and analysis module 131 . Step 3: The signaling capture and analysis module 131 will automatically capture the signaling information of the mobile users in the entire network, and after analyzing the signaling, store the user IP addresses, door numbers, base station IPs, APNs and other users of the users in the entire network. Information to the entire network user data sheet. The fourth step: the communication transceiver 110 receives the user traffic packet in the MEC area. If the signaling capture and analysis module 131 determines that the packet is an uplink packet, it will be directed to the local offload determination module 136, and if it is a downlink packet, it will be directed to the network domain. Name system resolution module 134 . Step 5: If the domain name system resolution module 134 determines that the packet is a DNS response packet, and the domain name queried has a URL to be uninstalled locally, then the corresponding IP address and its TTL are sent to the local uninstall determination module. Group 136. Step 6: When the user generates a signaling procedure in the MEC area, the signaling capture and analysis module 131 captures the TEID in the user's signaling information. Step 7: The signaling capture and analysis module 131 uses the TEID to query the corresponding user information (including the user IP address, door number, base station IP address, and APN) in the user data table of the entire network, and uses these TEIDs. The user information is inquired into the local uninstall condition setting module 135 . If the local uninstall condition information is met, the signaling capture analysis module 131 transmits the TEID to the local uninstall determination module 136 . The eighth step: the local uninstall determination module 136 receives the user traffic packet in the MEC area, and grabs the TEID in the packet and looks up the table to determine whether local uninstall is required. If local offloading is required, the local offloading determining module 136 then grabs the destination IP address and determines whether it is the local offloading destination IP address, and determines that the TTL of the IP address has not expired. Finally, the local offload determination module 136 removes the GTP header from the qualified packets and sends them to the corresponding internal server 30 of the MEC service; the unqualified packets are sent to the core network 40 .

To sum up, the mobile edge computing platform and the packet offloading method based on the mobile edge computing according to the embodiments of the present invention include the following features:

In the embodiments of the present invention, various channels, including data from base stations, APNs, user numbers, browsing websites, etc., can be used as conditions for local unloading, thereby increasing the flexibility of comparison.

The embodiments of the present invention provide administrators with multiple local uninstall condition settings, which can be set in a small range to finely control the behavior of a single user, or set in a large range to control the behavior of users in an entire area, which can more flexibly manage mobile edge network users.

In the embodiment of the present invention, it is only necessary to look at the outer layer information of the packet to confirm whether it is a packet that needs to be unloaded locally, thereby increasing the processing efficiency of the packet.

In terms of commercial applications, the embodiments of the present invention provide enterprise users with the ability to manage the local uninstallation of company employees in a more subtle manner to meet the needs of the enterprise, and to improve information security protection and prevent internal data leakage.

Although the present invention has been disclosed above by the embodiments, it is not intended to limit the present invention. Anyone with ordinary knowledge in the technical field can make some changes and modifications without departing from the spirit and scope of the present invention. Therefore, The protection scope of the present invention shall be determined by the scope of the appended patent application.

10: User Equipment 20: Mobile Internet 30: Intranet 40: Core Network 50:Internet 100: Mobile Edge Computing Platform 110: Communication transceiver 130: Storage 131: Signal capture analysis module 132: The first signaling capture and analysis module 133: Second signaling capture and analysis module 134: Domain Name System Resolution Module 135: Local uninstall condition setting module 136: Uninstall judgment module locally S210~S270, S310~S370, S410~S454, S511~S525, S611~S658: Steps

FIG. 1 is a block diagram of components of a mobile edge computing platform according to an embodiment of the present invention. FIG. 2 is a flowchart of a packet offloading method based on mobile edge computing according to an embodiment of the present invention. FIG. 3 is a flowchart of unloading condition setting according to an embodiment of the present invention. FIG. 4 is a flowchart of signaling extraction and analysis according to an embodiment of the present invention. FIG. 5 is a flowchart of domain name system (Domain Name System, DNS) resolution according to an embodiment of the present invention. FIG. 6 is a flowchart of local uninstall determination according to an embodiment of the present invention.

S210~S270:

Claims (10)

A Mobile Edge Computing (MEC) platform, comprising: a communication transceiver for receiving an uplink communication packet from a mobile network; a memory for recording a plurality of software modules; and A processor, coupled to the communication transceiver and the storage, accesses and executes the software modules recorded in the storage, and the software modules include: a signaling extraction and analysis module, which obtains an endpoint identification code of a user equipment that transmits user data via a tunneling protocol from a signaling packet, and obtains the user information of the user equipment according to the endpoint identification code; and A local offload determination module determines whether to forward the uplink traffic packet originally sent to a core network to an internal server according to the endpoint identification code and a destination network address of the uplink traffic packet. The mobile edge computing platform of claim 1, wherein the user information includes a door number, an Access Point Name (APN), a base station network address, and a user network address, the The signaling capture analysis module determines whether the user information and the endpoint identification code conform to a local offload authority, and if the local offload permission is met, the local offload determination module determines whether to forward the uplink traffic packet. The mobile edge computing platform of claim 1, wherein the communication transceiver receives a downlink traffic packet from the core network, and the software modules further comprise: A domain name system (Domain Name System, DNS) parsing module determines whether the downlink traffic packet is a response packet for querying a domain, and obtains a domain network corresponding to the downlink traffic packet according to the domain address and its survival time, and the local offload judging module determines whether to forward the uplink traffic packet according to the network domain network address and the survival time. The mobile edge computing platform according to claim 3, wherein the local offload determination module obtains the corresponding survival time according to the destination network address, and determines whether the survival time has expired, and if the survival time has not expired, Then, the local offload determination module removes the header of the tunnel protocol in the uplink traffic packet and forwards the remaining packets to the internal server. The mobile edge computing platform of claim 2, wherein the communication transceiver receives at least one first signaling from the core network and at least one second signaling from a base station, and the signaling is captured and analyzed The module obtains the corresponding door number, the access point name, and the user network address from the packet about create session in the at least one first signaling, and obtains the corresponding address from the at least one first signaling The corresponding endpoint identification code, the gate number, the access point name, the base station network address, and the user network address are obtained from the packet about modifying the bearer, and stored in the a network-wide user data table, and the signaling extraction and analysis module obtains the endpoint identification code from a packet related to initial context setup in the at least one second signaling, and according to the endpoint identification The code is used to query the user information from the user data table of the whole network. A packet offloading method based on mobile edge computing, comprising: receiving an uplink traffic packet from a mobile network; Obtaining an endpoint identifier of a user equipment for transmitting user data via a tunneling protocol from the signaling packet; obtain user information of the user equipment according to the endpoint identifier; and Whether to forward the upstream traffic packet originally sent to a core network to an internal server is determined according to the endpoint identification code and a destination network address of the upstream traffic packet. The packet unloading method based on mobile edge computing as claimed in claim 6, wherein the user information includes a door number, an access point name, a base station network address, and a user network address, wherein it is determined whether to The steps of forwarding the upstream traffic packet originally sent to the core network to the internal server include: determining whether the endpoint identifier conforms to a local uninstall permission; and If the local offload permission is met, it is determined whether to forward the uplink traffic packet. The packet unloading method based on mobile edge computing as described in claim 6, further comprising: receiving a downlink traffic packet from one of the core networks; determining whether the downlink traffic packet is a response packet for querying a network domain; and Obtaining a network domain network address and its survival time corresponding to the downlink traffic packet according to the network domain, wherein the step of determining whether to forward the uplink traffic packet originally sent to the core network to the internal server includes the following steps: : It is judged whether to forward the uplink traffic packet according to the network domain network address and the survival time. The packet offloading method based on mobile edge computing according to claim 8, wherein the step of judging whether to forward the uplink traffic packet according to the network domain network address and the time-to-live includes: Obtain the corresponding time-to-live according to the destination network address; Determine whether the survival time has expired; If the time-to-live has not expired, remove the header of the tunnel protocol in the upstream traffic packet and forward the remaining packets to the internal server. The packet unloading method based on mobile edge computing as described in claim 7, further comprising: receiving at least one first signaling from the core network and at least one second signaling from a base station; Obtain the corresponding door number, the access point name, and the user network address from the packet about establishing the session in the at least one first signaling; Obtain the corresponding endpoint identification code, the gate number, the access point name, the base station network address, and the user network address from the packet about modifying the bearer in the at least one first signaling, and Accordingly, it is stored in a network-wide user data table; Obtaining the endpoint identifier from the packet in the at least one second signaling for initiating context establishment; and The user information is inquired from the network-wide user data table according to the endpoint identification code.

Images