CN111757401B

CN111757401B - Gateway selection system and method

Info

Publication number: CN111757401B
Application number: CN201910251671.8A
Authority: CN
Inventors: 龙思锐
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2019-03-29
Filing date: 2019-03-29
Publication date: 2021-12-14
Anticipated expiration: 2039-03-29
Also published as: CN111757401A; WO2020200136A1

Abstract

The embodiment of the application discloses a gateway selection system and a method, wherein the gateway selection system comprises an MME, a plurality of SGW-Cs and a plurality of SGW-Us, wherein: the first SGW-C is used for sending an identifier of the first SGW-U to the MME when detecting that the first SGW-U is abnormal, wherein the first SGW-C is one of the SGW-Cs, and the first SGW-U is the SGW-U which is in butt joint with the first SGW-C among the SGW-Us; the MME is used for receiving the identification of the first SGW-U from the first SGW-C, determining a first session influenced by the first SGW-U according to the identification of the first SGW-U, and reselecting a second SGW-C which does not fail for the first session; the second SGW-C is used to reselect the second SGW-U that has not failed for the first session. Based on the system, the second SGW-U that is not failing can be reselected for the first session.

Description

Gateway selection system and method

Technical Field

The present application relates to the field of communications technologies, and in particular, to a gateway selection system and method.

Background

The third generation partnership project (3 GPP) standards working group started research on EPS (evolved packet system) from R8. The whole EPS system can be divided into a radio access network and a core network. The core network includes a Mobility Management Entity (MME), a serving gateway (S-GW), a packet data gateway (PDN-GW), and the like. With the advent of Software Defined Network (SDN) technology, CU (control plane and user plane) separation is becoming the developing direction of mobile networks, and 3GPP has proposed LTE architecture research for CU separation in Release 13 stage. As shown in fig. 1, under the CU separation architecture, an original PGW is split into two functional entities, namely, a packet data Gateway Control plane function (PGW-C) and a packet data Gateway User plane function (PGW-U), and an original SGW is split into two functional entities, namely, a Serving Gateway Control plane function (SGW-C) and a Serving Gateway User plane function (SGW-U). The SGW-C and the SGW-U can be in a one-to-many relationship during deployment, and one SGW-C device may correspond to one or more SGW-U devices. Similarly, the PGW-C and the PGW-U may be in a one-to-many relationship, and one PGW-C device may correspond to one or more PGW-U devices.

In practical applications, the SGW-U in the core network may fail or restart. Service interruption can result after an SGW-U in the core network fails or reboots. Therefore, the SGW-U needs to be reselected after a functional entity in the core network fails or reboots. How to reselect the SGW-U is a problem to be solved urgently at present.

Disclosure of Invention

The embodiment of the application provides a gateway selection system and a method, which can reselect an SGW-U.

In a first aspect, an embodiment of the present application provides a gateway selection system, where the gateway selection system includes a mobility management entity MME, a plurality of serving gateway control plane functional entities SGW-C, and a plurality of serving gateway user plane functional entities SGW-U, where: the first SGW-C is used for sending an identifier of the first SGW-U to the MME when the first SGW-U is detected to be abnormal, wherein the first SGW-C is one SGW-C in the plurality of SGW-Cs, and the first SGW-U is the SGW-U which is in butt joint with the first SGW-C in the plurality of SGW-Us; an MME for receiving an identification of the first SGW-U from the first SGW-C; the MME is also used for determining a first session influenced by the first SGW-U according to the identification of the first SGW-U and reselecting a second SGW-C which does not fail for the first session; a second SGW-C to reselect a non-failed second SGW-U for the first session. Based on the system described in the first aspect, the MME can reselect a second SGW-C that is not failed for the first session, and the second SGW-C can reselect a second SGW-U that is not failed for the first session. The first session may then be subsequently serviced by the second SGW-C and the second SGW-U.

In a particular design, the MME is further configured to record, in a session, an identity of an SGW-U used for the session; the specific implementation manner of the MME determining the first session affected by the first SGW-U according to the identifier of the first SGW-U is as follows: the MME determines a session of the sessions having an identification of the first SGW-U as a first session affected by the first SGW-U. Based on this specific design, the MME can accurately determine the first session affected by the first SGW-U.

In a specific design, the MME reselects the second SGW-C that has not failed for the first session in the following manner: and when the MME receives the uplink signaling of the first session, the MME reselects the second SGW-C which does not fail for the first session. In the specific design, when the MME receives the uplink signaling of the first session, it may trigger reselection of the SGW-C for the corresponding session, which is beneficial to timely recovery of the emergency service through the specific design.

In a specific design, the gateway selection system further includes a plurality of packet data network gateway control plane function entities PGW-C and a plurality of packet data network gateway user plane function entities PGW-U, where: the first PGW-C is used for receiving an identification of a first SGW-U with an abnormality from a first PGW-U or the first SGW-C, the first PGW-C is a PGW-C in the multiple PGW-Cs and is in butt joint with the first SGW-C, and the first PGW-U is a PGW-U in the multiple PGW-Us and is in butt joint with the first SGW-U and the first PGW-C; the first PGW-C is further used for determining a second session influenced by the first SGW-U according to the identification of the first SGW-U; the first PGW-C is further configured to send a first message to the MME when receiving a downlink signaling of the second session, where the first message is used to trigger the MME to reselect an non-failed SGW-C for the second session; the MME is also used for receiving a first message from the first PGW-C and reselecting a third SGW-C which does not fail for the second session according to the first message; a third SGW-C to reselect a non-failed third SGW-U for the second session. Through the specific design, the session reselection SGW-U influenced by the abnormal SGW-U can be triggered in time through the downlink signaling sent by the network side, and the influenced service can be recovered in time.

In a specific design, the first SGW-C is further configured to send, to the first PGW-C, an identification of an SGW-U used for the session; the first PGW-C is further used for receiving the identification of the SGW-U used by the session from the first SGW-C and recording the identification of the SGW-U corresponding to the session in the session; the specific implementation manner of the first PGW-C determining, according to the identifier of the first SGW-U, the second session affected by the first SGW-U is as follows: the first PGW-C determines that a session of the sessions having an identification of the first SGW-U is a second session affected by the first SGW-U. Based on the specific design, the first PGW-C can accurately determine the first session affected by the first SGW-U.

In a specific design, the gateway selection system further includes a plurality of packet data network gateway control plane function entities PGW-C and a plurality of packet data network gateway user plane function entities PGW-U, where: the first SGW-C is further used for determining a second session influenced by the first SGW-U when the first SGW-U is detected to be abnormal; the first PGW-C is used for receiving a first signaling corresponding to a second session and sending a second signaling corresponding to the second session to the first SGW-C according to the first signaling, and the first PGW-C is a PGW-C which is in butt joint with the first SGW-C among the PGW-Cs; the first SGW-C is further configured to receive a second signaling from the first PGW-C, and send a response of the second signaling to the first PGW-C, where the response of the second signaling carries a cause value used for indicating that the context of the second session is not found; the first PGW-C is further configured to receive a response of the second signaling from the first SGW-C, and send a first message to the MME, where the first message is used to trigger the MME to reselect an non-failed SGW-C for the second session; the MME is also used for receiving a first message from the first PGW-C and reselecting a third SGW-C which does not fail for the second session according to the first message; a third SGW-C to reselect a non-failed third SGW-U for the second session. Through the specific design, the session reselection SGW-U influenced by the abnormal SGW-U can be triggered in time through the signaling sent by the network side, and the influenced service can be recovered in time.

In a specific design, the first SGW-C is further configured to record, in the session, an identification of an SGW-U used by the session; the first SGW-C determines that the second session affected by the first SGW-U is implemented as: the first SGW-C determines that a session of the sessions having the identity of the first SGW-U is a second session affected by the first SGW-U. Based on the specific design, the first SGW-C can accurately determine the first session affected by the first SGW-U.

In a specific design, the first SGW-C is further configured to delete the second session after receiving a request for creating a dedicated bearer corresponding to the second session. Since the SGW-C needs to be reselected for the second session, the first SGW-C keeps the second session as it is, and the first SGW-C needs to delete the second session.

In a specific design, the gateway selection system further includes a plurality of packet data network gateway control plane function entities PGW-C and a plurality of packet data network gateway user plane function entities PGW-U, where: the first PGW-U is used for determining a third session influenced by the first SGW-U when the first SGW-U is detected to be abnormal, and the first PGW-U is a PGW-U which is in butt joint with the first SGW-U in the PGW-Us; the first PGW-U is further configured to send a notification message to the first PGW-C when receiving the downlink data packet of the third session, where the notification message is used to notify the third session that the downlink data packet is received, and the first PGW-C is a PGW-C of the multiple PGW-cs, where the PGW-C is docked with the first PGW-U; the first PGW-C is further configured to receive a notification message from the first PGW-U and send a second message to the MME, where the second message is used to trigger the MME to reselect an SGW-C that has not failed for the third session; the MME is also used for receiving a second message from the PGW-C and reselecting a fourth SGW-C which does not fail for a third session according to the second message; and the fourth SGW-C is used for reselecting the fourth SGW-U which does not fail for the third session. Through the specific design, the SGW-U can be timely reselected for the session affected by the abnormal SGW-U through triggering of the downlink data message sent by the network side, and timely recovery of the affected service is facilitated.

In a specific design, the first PGW-U is further configured to record, in the session, an identification of an SGW-U used by the session; the specific implementation manner of the first PGW-U determining the third session affected by the first SGW-U is as follows: the first PGW-U determines that a session having an identification of the first SGW-U among the sessions is a third session affected by the first SGW-U. Based on the specific design, the first PGW-U can accurately determine the first session affected by the first SGW-U.

In a second aspect, an embodiment of the present application provides a gateway selection method, where the method includes: when detecting that a first serving gateway user plane functional entity SGW-U docked with a first SGW-C is abnormal, a first serving gateway control plane functional entity SGW-C sends an identifier of the first SGW-U to a mobile management entity MME; the MME receives the identification of the first SGW-U from the first SGW-C, and determines a first session influenced by the first SGW-U according to the identification of the first SGW-U; the MME reselects a second SGW-C which does not fail for the first session; the second SGW-C reselects the second SGW-U that has not failed for the first session.

In one particular design, the MME records an identification of the SGW-U used for the session in the session; the specific implementation manner of the MME determining the first session affected by the first SGW-U according to the identifier of the first SGW-U is as follows: the MME determines a session of the sessions having an identification of the first SGW-U as a first session affected by the first SGW-U.

In a specific design, the MME reselects the second SGW-C that has not failed for the first session in the following manner: when the MME receives the service request of the first session, the MME reselects the second SGW-C which does not fail for the first session.

In one particular design, the method further includes the steps of: the method comprises the steps that a first PGW-C receives identification of a first SGW-U with an abnormality from the first PGW-U or the first SGW-C, wherein the first PGW-C is a PGW-C in butt joint with the first SGW-C, and the first PGW-U is a PGW-U in butt joint with the first SGW-U and the first PGW-C; the first PGW-C determines a second session influenced by the first SGW-U according to the identifier of the first SGW-U; when receiving a downlink signaling of a second session, a first PGW-C sends a first message to an MME, wherein the first message is used for triggering the MME to reselect an SGW-C which does not fail for the second session; the MME receives a first message from the first PGW-C, and reselects a third SGW-C which does not fail for the second session according to the first message; the third SGW-C reselects the non-failed third SGW-U for the second session.

In one particular design, the method further includes the steps of: the first SGW-C sends an identification of an SGW-U used by the session to the first PGW-C; the first PGW-C receives the identification of the SGW-U used by the session from the first SGW-C, and records the identification of the SGW-U corresponding to the session in the session; the specific implementation manner of the first PGW-C determining, according to the identifier of the first SGW-U, the second session affected by the first SGW-U is as follows: the first PGW-C determines that a session of the sessions having an identification of the first SGW-U is a second session affected by the first SGW-U.

In one particular design, the method further includes the steps of: when the first SGW-C detects that the first SGW-U is abnormal, determining a second session influenced by the first SGW-U; a first PGW-C receives a first signaling corresponding to a second session, and sends a second signaling corresponding to the second session to the first SGW-C according to the first signaling, wherein the first PGW-C is a PGW-C which is in butt joint with the first SGW-C; the first SGW-C receives a second signaling from the first PGW-C and sends a response of the second signaling to the first PGW-C, wherein the response of the second signaling carries a reason value for indicating that the context of the second session is not found; the first PGW-C receives a response of the second signaling from the first SGW-C and sends a first message to the MME, wherein the first message is used for triggering the MME to reselect an SGW-C which does not fail for the second session; the MME receives a first message from the first PGW-C, and reselects a third SGW-C which does not fail for the second session according to the first message; the third SGW-C reselects the non-failed third SGW-U for the second session.

In one particular design, a first SGW-C records an identification of an SGW-U used by a session in the session; the first SGW-C determines that the second session affected by the first SGW-U is implemented as: the first SGW-C determines that a session of the sessions having the identity of the first SGW-U is a second session affected by the first SGW-U.

In a specific design, the first SGW-C deletes the second session after receiving a request for creating a dedicated bearer corresponding to the second session.

In one particular design, the method further includes the steps of: when the first PGW-U detects that the first SGW-U is abnormal, determining a third session influenced by the first SGW-U, wherein the first PGW-U is a PGW-U which is in butt joint with the first SGW-U; when receiving a downlink data message of a third session, a first PGW-U sends a notification message to a first PGW-C, wherein the notification message is used for notifying the third session of receiving the downlink data message, and the first PGW-C is a PGW-C which is in butt joint with the first PGW-U; the first PGW-C receives the notification message from the first PGW-U and sends a second message to the MME, wherein the second message is used for triggering the MME to reselect an SGW-C which does not fail for a third session; the MME receives a second message from the PGW-C, and reselects a fourth SGW-C which does not fail for a third session according to the second message; the fourth SGW-C reselects the fourth SGW-U that did not fail for the third session.

In one particular design, the first PGW-U may also record, in the session, an identification of the SGW-U used by the session; the specific implementation manner of the first PGW-U determining the third session affected by the first SGW-U is as follows: the first PGW-U determines that a session having an identification of the first SGW-U among the sessions is a third session affected by the first SGW-U.

Based on the same inventive concept, the specific design advantages of the second aspect or the second aspect may refer to the specific design advantages of the first aspect or the first aspect, and repeated details are not repeated.

In a third aspect, an embodiment of the present application provides a gateway selection method, where the method includes: a mobile management entity MME receives an identifier of a first service gateway user plane functional entity SGW-U with an abnormality from a first service gateway control plane functional entity SGW-C, and the first SGW-C is in butt joint with the first SGW-U; the MME determines a first session influenced by the first SGW-U according to the identifier of the first SGW-U; the MME reselects the second SGW-C that has not failed for the first session.

In a specific design, the MME may further record, in the session, an identifier of an SGW-U corresponding to the session; the specific implementation manner of the MME determining the first session affected by the first SGW-U according to the identifier of the first SGW-U is as follows: the MME determines a session of the sessions having an identification of the first SGW-U as a first session affected by the first SGW-U.

In a specific design, the MME reselects the second SGW-C that has not failed for the first session in the following manner: and when the MME receives the uplink signaling of the first session, the MME reselects the second SGW-C which does not fail for the first session.

Based on the same inventive concept, the beneficial effects of the third aspect or the specific design of the third aspect may refer to the beneficial effects of the first aspect or the specific design of the first aspect, and repeated details are not repeated.

In a fourth aspect, an embodiment of the present application provides a gateway selection method, where the method includes: a first packet data network gateway control plane function entity (PGW-C) receives an identifier of an abnormal first serving gateway user plane function entity (SGW-U) from a first packet data network gateway user plane function entity (PGW-U) or a first serving gateway control plane function entity (SGW-C), wherein the first PGW-C is in butt joint with the first PGW-U, the first PGW-C is in butt joint with the first SGW-C, the first SGW-U is in butt joint with the first PGW-U, and the first SGW-U is in butt joint with the first SGW-C; the first PGW-C determines a second session influenced by the first SGW-U according to the identifier of the first SGW-U; and when receiving the downlink signaling of the second session, the first PGW-C sends a first message to the MME, wherein the first message is used for triggering the MME to reselect an SGW-C which does not fail for the second session.

In one particular design, the first PGW-C may also receive, from the first SGW-C, an identification of an SGW-U used for the session; the first PGW-C may also record in the session an identification of the SGW-U used by the session; the specific implementation manner of the first PGW-C determining, according to the identifier of the first SGW-U, the second session affected by the first SGW-U is as follows: the first PGW-C determines that a session of the sessions having an identification of the first SGW-U is a second session affected by the first SGW-U.

Based on the same inventive concept, the beneficial effects of the fourth aspect or the specific design of the fourth aspect can be referred to the beneficial effects of the specific design of the first aspect, and repeated details are not repeated.

In a fifth aspect, an embodiment of the present application provides a gateway selection method, where the method includes: when a first service gateway control plane functional entity SGW-C detects that a butted first service gateway user plane functional entity SGW-U is abnormal, the first SGW-C determines a second session influenced by the first SGW-U; the first SGW-C receives a second signaling corresponding to a second session from a first packet data network gateway control plane function entity (PGW-C); and the first SGW-C sends a response of the second signaling to the first PGW-C, wherein the response of the second signaling carries a reason value for indicating that the context of the second session is not found.

In one particular design, the first SGW-C may also record, in the session, an identification of the SGW-U used by the session; the first SGW-C determines that the second session affected by the first SGW-U is implemented as: the first SGW-C determines that a session of the sessions having the identity of the first SGW-U is a second session affected by the first SGW-U.

In a particular design, the first SGW-C deletes the second session after receiving a create-specific bearer request corresponding to the second session.

Based on the same inventive concept, the beneficial effects of the fifth aspect or the specific design of the fifth aspect can be referred to the beneficial effects of the specific design of the first aspect, and repeated details are not repeated.

In a sixth aspect, an embodiment of the present application provides a gateway selection method, where the method includes: a first packet data network gateway control plane function entity (PGW-C) receives a first signaling; the first PGW-C sends a second signaling to a first serving gateway control plane function (SGW-C) according to the first signaling; the first PGW-C receives a response of the second signaling from the first SGW-C, the response of the second signaling carries a cause value used for indicating that a context of a second session is not found, and the second session is a session influenced by an abnormal first service gateway user plane function entity (SGW-U) in the first SGW-C; the first PGW-C sends a first message to a Mobility Management Entity (MME), wherein the first message is used for triggering the MME to reselect a third SGW-C which does not fail for the second session.

Based on the same inventive concept, the beneficial effects of the sixth aspect can be found in the beneficial effects of the specific design of the first aspect, and repeated details are not repeated.

In a seventh aspect, an embodiment of the present application provides a gateway selection method, where the method includes: when a first packet data network gateway user plane functional entity (PGW-U) detects that a butted first service gateway user plane functional entity (SGW-U) is abnormal, the first PGW-U determines a third session influenced by the first SGW-U; and when receiving the downlink data message of the third session, the first PGW-U sends a notification message to the first packet data network gateway control plane function entity (PGW-C), wherein the notification message is used for notifying the third session of receiving the downlink data message.

Based on the same inventive concept, the beneficial effects of the seventh aspect or the specific design of the seventh aspect can be referred to the beneficial effects of the specific design of the first aspect, and repeated details are not repeated.

In an eighth aspect, an embodiment of the present application provides a gateway selection method, where the method includes: a first packet data network gateway control plane function entity (PGW-C) receives a notification message from a first packet data network gateway user plane function entity (PGW-U), wherein the notification message is used for notifying that a downlink data message of a third session is received, and the third session is a session influenced by an abnormal first service gateway user plane function entity (SGW-U) in the first PGW-U; the first PGW-C sends a first message to a Mobility Management Entity (MME), wherein the first message is used for triggering the MME to reselect an SGW-C which does not fail for a third session.

Based on the same inventive concept, the beneficial effects of the eighth aspect may refer to the beneficial effects of the specific design of the first aspect, and repeated details are not repeated.

In a ninth aspect, an embodiment of the present application provides a gateway selection system, where the gateway selection system includes a mobility management entity MME, multiple serving gateway control plane functional entities SGW-C, multiple serving gateway user plane functional entities SGW-U, multiple packet data network gateway control plane functional entities PGW-C, and multiple packet data network gateway user plane functional entities PGW-U, where: the first PGW-U is used for receiving an identification of a first SGW-C with an abnormality from a first PGW-C or the first SGW-U, the first PGW-U is one of the PGW-Us, the first PGW-C and the first SGW-U are respectively docked with the first PGW-U, and the first PGW-C and the first SGW-U are respectively docked with the first SGW-C; the first PGW-U is used for determining a first session influenced by the first SGW-C according to the identification of the first SGW-C; the first PGW-U is further configured to send a notification message to the first PGW-C when receiving the downlink data packet of the first session, where the notification message is used to notify that the downlink data packet of the first session is received; the first PGW-C is used for receiving a notification message from the first PGW-U and sending a first message to the MME, wherein the first message is used for triggering the MME to reselect an SGW-C which does not fail for the first session; the MME is further used for receiving a first message from the first PGW-C and reselecting a second SGW-C which does not fail for the first session according to the first message. Based on the method described in the ninth aspect, the reselection of the SGW-C for the session affected by the abnormal SGW-C can be triggered in time through the downlink data packet sent by the network side, which is beneficial to timely recovery of the affected service.

In a specific design, the first PGW-C is further configured to send, to the first PGW-U, an identifier of an SGW-C used for the session; the first PGW-U is further used for receiving the identification of the SGW-C used by the session from the first PGW-C and recording the identification of the SGW-C used by the session in the session; the specific implementation manner of the first PGW-U determining, according to the identifier of the first SGW-C, the first session affected by the first SGW-C is as follows: the first PGW-U determines that a session of the sessions having an identification of the first SGW-C is a first session affected by the first SGW-C. Based on the specific design, the first PGW-U can accurately determine the first session affected by the first SGW-C.

In a tenth aspect, an embodiment of the present application provides a gateway selection method, where the method includes: a first packet data network gateway user plane function entity PGW-U receives an identifier of a first service gateway control plane function entity SGW-C with an abnormality from the first packet data network gateway control plane function entity PGW-C or the first service gateway user plane function entity SGW-U, the first PGW-C and the first SGW-U are respectively butted with the first PGW-U, and the first PGW-C and the first SGW-U are respectively butted with the first SGW-C; the first PGW-U determines a first session influenced by the first SGW-C according to the identification of the first SGW-C; when receiving a downlink data message of a first session, a first PGW-U sends a notification message to a first PGW-C, wherein the notification message is used for notifying that the downlink data message of the first session is received; the first PGW-C receives a notification message from the first PGW-U and sends a first message to the MME, wherein the first message is used for triggering the MME to reselect an SGW-C which does not fail for the first session; the MME receives the first message from the first PGW-C, and reselects a second SGW-C which does not fail for the first session according to the first message.

In one particular design, the method may further include the steps of: the first PGW-C sends an identification of the SGW-C used by the session to the first PGW-U; the first PGW-U receiving, from the first PGW-C, an identification of an SGW-C used for the session; the first PGW-U records the identification of the SGW-C used by the session in the session; the specific implementation manner of the first PGW-U determining, according to the identifier of the first SGW-C, the first session affected by the first SGW-C is as follows: the first PGW-U determines that a session of the sessions having an identification of the first SGW-C is a first session affected by the first SGW-C.

Based on the same inventive concept, the specific design advantages of the tenth aspect or the tenth aspect may refer to the specific design advantages of the ninth aspect or the ninth aspect, and repeated details are not repeated.

In an eleventh aspect, an embodiment of the present application provides a gateway selection method, where the method includes: a first packet data network gateway user plane function entity (PGW-U) receives an identifier of a first serving gateway control plane function entity (SGW-C) with an abnormality from the first packet data network gateway control plane function entity (PGW-C) or the first serving gateway user plane function entity (SGW-U), wherein the first PGW-C and the first SGW-U are respectively butted with a first PGW-U, and the first PGW-C and the first SGW-U are respectively butted with the first SGW-C; the first PGW-U determines a first session influenced by the first SGW-C according to the identification of the first SGW-C; when receiving the downlink data packet of the first session, the first PGW-U sends a notification message to the first PGW-C, where the notification message is used to notify that the downlink data packet of the first session is received.

In one particular design, the first PGW-U may also receive, from the first PGW-C, an identification of an SGW-C used for the session; the first PGW-U can also record the identification of the SGW-C used by the session in the session; the specific implementation manner of the first PGW-U determining, according to the identifier of the first SGW-C, the first session affected by the first SGW-C is as follows: the first PGW-U determines that a session of the sessions having an identification of the first SGW-C is a first session affected by the first SGW-C.

In a twelfth aspect, an embodiment of the present application provides a gateway selection method, where the method includes: a first packet data network gateway control plane function entity (PGW-C) receives a notification message from a first packet data network gateway user plane function entity (PGW-U), wherein the notification message is used for notifying that a downlink data message of a first session is received, and the first session is a session influenced by an abnormal first service gateway control plane function entity (SGW-C) in the first PGW-U; the first PGW-C sends a first message to a Mobile Management Entity (MME), wherein the first message is used for triggering the MME to reselect a non-failed SGW-C for the first session.

In a specific design, when the first PGW-C determines that the first SGW-C is abnormal, the first PGW-C sends an identifier of the first SGW-C to the first PGW-U.

In a particular design, the first PGW-C sends, to the first PGW-U, an identification of an SGW-C corresponding to the session.

Based on the same inventive concept, the specific design of the eleventh aspect or the eleventh aspect may have the advantages described in the ninth aspect or the ninth aspect, and repeated details are not repeated.

In a thirteenth aspect, an embodiment of the present application provides a service recovery system, where the service recovery system includes a mobility management entity MME, multiple serving gateway control plane functional entities SGW-C, multiple serving gateway user plane functional entities SGW-U, multiple packet data network gateway control plane functional entities PGW-C, and multiple packet data network gateway user plane functional entities PGW-U, where: the first SGW-C is used for receiving an identifier of a first PGW-U with an abnormality from a first PGW-C or the first SGW-U, the first SGW-C is one of the multiple SGW-Cs, the first PGW-C is one of the multiple PGW-Cs and is in butt joint with the first SGW-C, the first SGW-U is one of the multiple SGW-Us and is in butt joint with the first SGW-C, and the first PGW-U is in butt joint with the first PGW-C and the first SGW-C respectively; the first SGW-C is used for sending an identifier of the first PGW-U with the abnormality to the MME; the MME is used for receiving the identification of the first PGW-U from the first SGW-C and determining a first session influenced by the first PGW-U according to the identification of the first PGW-U; and the MME is also used for deleting the first session and indicating the user equipment corresponding to the first session to reestablish the PDN connection of the packet data network. Based on the system described in the thirteenth aspect, the affected service can be recovered.

In a particular design, the MME is further configured to record, in the session, an identity of a PGW-U used for the session; the specific implementation manner of the MME determining the first session affected by the first PGW-U according to the identifier of the first PGW-U is: the MME determines a session with the identification of the first PGW-U in the session to be the first session affected by the first PGW-U. Based on this specific design, the MME can accurately determine the first session affected by the first PGW-U.

In a specific design, a specific implementation manner in which the MME deletes the first session and instructs the ue corresponding to the first session to reestablish the PDN connection is as follows: and when the MME receives the uplink signaling of the first session, deleting the first session and indicating the user equipment corresponding to the first session to reestablish the PDN connection. In the specific design, when the MME receives the signaling sent by the user equipment side, the service corresponding to the signaling can be triggered to be recovered, so that by the design, it is beneficial to recover the emergency service in time.

In a specific design, the first SGW-U is configured to determine, when it is determined that the first PGW-U is abnormal, a second session affected by the first PGW-U; the first SGW-U is used for sending a notification message to the first SGW-C when receiving the uplink data message of the second session, wherein the notification message is used for notifying the second session of receiving the uplink data message; a first SGW-C for receiving a notification message from the first SGW-U; and the first SGW-C is also used for deleting the second session and indicating the user equipment corresponding to the second session to reestablish the PDN connection. Based on the design, the first SGW-U can trigger the recovery of the service of the second session in time when receiving the uplink data packet of the second session.

In a specific design, the first SGW-U is further configured to record, in the session, an identification of a PGW-U used by the session; the first SGW-U determines that the second session affected by the first PGW-U is implemented in the following specific manner: the first SGW-U determines that the session with the identification of the first PGW-U in the sessions is a second session affected by the first PGW-U. Based on the specific design, the first SGW-U can accurately determine the second session affected by the first PGW-U.

In a specific design, the first PGW-C is configured to, when detecting that the first PGW-U is abnormal, determine a third session affected by the first PGW-U; the first PGW-C is further configured to delete the third session and instruct the user equipment corresponding to the third session to reestablish the PDN connection when receiving the downlink signaling of the third session. Based on the design, the first PGW-C can trigger the recovery of the service of the third session in time when receiving the signaling of the third session.

In a specific design, the first PGW-C is further configured to record, in the session, an identification of a PGW-U used by the session; the first PGW-C determines that the third session affected by the first PGW-U is implemented in the following specific manner: the first PGW-C determines that a session with the identification of the first PGW-U in the sessions is a third session affected by the first PGW-U. Based on the specific design, the first PGW-C can accurately determine the third session affected by the first PGW-U.

In a specific design, the first PGW-C is further configured to send, when receiving a signaling of a fourth session, a message to a second PGW-U used by the fourth session according to the signaling of the fourth session; the first PGW-C is further configured to delete the fourth session and instruct the ue corresponding to the fourth session to reestablish the PDN connection, after retransmitting the message to the second PGW-U, if the first PGW-C does not receive the response message of the message within the preset time period. Based on the specific design, when the first PGW-C does not sense the abnormality of the second PGW-U in time, the service influenced by the second PGW-U can be recovered in time.

In a fourteenth aspect, an embodiment of the present application provides a service recovery method, where the method includes: a first serving gateway control plane function (SGW-C) receives an identifier of a first packet data network gateway user plane function entity (PGW-U) with an abnormality from a first packet data network gateway control plane function entity (PGW-C) or a first serving gateway user plane function entity (SGW-U), wherein the first PGW-C is in butt joint with the first SGW-C, the first SGW-U is in butt joint with the first SGW-C, the first PGW-U is in butt joint with the first PGW-C, and the first PGW-U is in butt joint with the first SGW-U; the first SGW-C sends an identifier of a first PGW-U with an abnormality to a mobile management entity MME; the MME receives an identification of a first PGW-U from a first SGW-C; the MME determines a first session influenced by the first PGW-U according to the identification of the first PGW-U; and the MME deletes the first session and instructs the user equipment corresponding to the first session to reestablish the PDN connection of the packet data network.

In one particular design, the MME records an identification of PGW-U used for the session in the session; the specific implementation manner of the MME determining the first session affected by the first PGW-U according to the identifier of the first PGW-U is: the MME determines a session with the identification of the first PGW-U in the session to be the first session affected by the first PGW-U.

In a specific design, a specific implementation manner in which the MME deletes the first session and instructs the ue corresponding to the first session to reestablish the PDN connection is as follows: and when the MME receives the uplink signaling of the first session, deleting the first session and indicating the user equipment corresponding to the first session to reestablish the PDN connection.

In one particular design, the method further includes the steps of: when the first SGW-U determines that the first PGW-U is abnormal, determining a second session influenced by the first PGW-U; when receiving an uplink data message of a second session, a first SGW-U sends a notification message to a first SGW-C, wherein the notification message is used for notifying the second session of receiving the uplink data message; the first SGW-C receives a notification message from the first SGW-U; and the first SGW-C deletes the second session and instructs the user equipment corresponding to the second session to reestablish the PDN connection.

In one particular design, the first SGW-U may also record, in the session, an identification of a PGW-U used by the session; the first SGW-U determines that the second session affected by the first PGW-U is implemented in the following specific manner: the first SGW-U determines that the session with the identification of the first PGW-U in the sessions is a second session affected by the first PGW-U.

In one particular design, the method further includes the steps of: the first PGW-C determines a third conversation influenced by the first PGW-U when detecting that the first PGW-U is abnormal; and when receiving the downlink signaling of the third session, the first PGW-C deletes the third session and instructs the user equipment corresponding to the third session to reestablish the PDN connection.

In one particular design, the first PGW-C records, in the session, an identification of a PGW-U used by the session; the first PGW-C determines that the third session affected by the first PGW-U is implemented in the following specific manner: the first PGW-C determines that a session with the identification of the first PGW-U in the sessions is a third session affected by the first PGW-U.

In a specific design, when the first PGW-C receives a signaling of a fourth session, a message is sent to a second PGW-U used by the fourth session according to the signaling of the fourth session; after the first PGW-C retransmits the message to the second PGW-U, if the first PGW-C does not receive the response message of the message within the preset time period, the fourth session is deleted, and the user equipment corresponding to the fourth session is instructed to reestablish the PDN connection.

Based on the same inventive concept, the specific design of the fourteenth aspect or the fourteenth aspect may have the advantages described in the above thirteenth aspect or the thirteenth aspect, and repeated descriptions are omitted.

In a fifteenth aspect, an embodiment of the present application provides a service recovery method, where the method includes: a mobile management entity MME receives an identifier of a first packet data network gateway user plane functional entity PGW-U with an abnormality from a first serving gateway control plane functional entity SGW-C; the MME determines a first session influenced by the first PGW-U according to the identification of the first PGW-U; and the MME deletes the first session and instructs the user equipment corresponding to the first session to reestablish the PDN connection of the packet data network.

In one particular design, the MME may also record in the session the identity of the PGW-U used by the session; the specific implementation manner of the MME determining the first session affected by the first PGW-U according to the identifier of the first PGW-U is: the MME determines a session with the identification of the first PGW-U in the session to be the first session affected by the first PGW-U.

Based on the same inventive concept, the specific design of the fifteenth aspect or the fifteenth aspect may have the advantages described in the thirteenth aspect or the thirteenth aspect, and repeated descriptions are omitted.

In a sixteenth aspect, an embodiment of the present application provides a service recovery method, where the method includes: when determining that a first packet data network gateway user plane functional entity (PGW-U) is abnormal, a first service gateway user plane functional entity (SGW-U) determines a second session influenced by the first PGW-U; when receiving the uplink data packet of the second session, the first SGW-U sends a notification message to the first serving gateway control plane function entity SGW-C, where the notification message is used to notify the second session of receiving the uplink data packet.

Based on the same inventive concept, the beneficial effects of the sixteenth aspect or the specific design of the sixteenth aspect can be referred to the beneficial effects of the specific design of the thirteenth aspect, and repeated details are not repeated.

In a seventeenth aspect, an embodiment of the present application provides a service recovery method, where the method includes: a first service gateway control plane function entity SGW-C receives a notification message from a first service gateway user plane function entity SGW-U, wherein the notification message is used for notifying a second session to receive an uplink data message, and the second session is a session influenced by an abnormal first packet data network gateway user plane function entity PGW-U; and the first SGW-C deletes the second session and instructs the user equipment corresponding to the second session to reestablish the PDN connection.

Based on the same inventive concept, the beneficial effects of the seventeenth aspect can be seen from the beneficial effects of the specific design of the thirteenth aspect, and repeated details are not repeated.

In an eighteenth aspect, an embodiment of the present application provides a service recovery method, where the method includes: when detecting that a first packet data network gateway control plane functional entity (PGW-C) is abnormal, determining a third session influenced by a first packet data network gateway user plane functional entity (PGW-U); and when receiving the downlink signaling of the third session, the first PGW-C deletes the third session and instructs the user equipment corresponding to the third session to reestablish the PDN connection.

In one particular design, the first PGW-C may also record, in the session, an identification of a PGW-U used by the session; the first PGW-C determines that the third session affected by the first PGW-U is implemented in the following specific manner: the first PGW-C determines that a session with the identification of the first PGW-U in the sessions is a third session affected by the first PGW-U.

Based on the same inventive concept, the specific design advantageous effects of the eighteenth aspect or the eighteenth aspect can be referred to the specific design advantageous effects of the thirteenth aspect, and repeated details are not repeated.

A nineteenth aspect provides a network device that can perform the method according to any one of the third to eighth aspects, the specific designs of the third to eighth aspects, the eleventh, twelfth, fifteenth to eighteenth, and fifteenth to eighteenth aspects. The function can be realized by hardware, and can also be realized by executing corresponding software by hardware. The hardware or software includes one or more units corresponding to the above functions. The unit may be software and/or hardware. Based on the same inventive concept, the principle and the beneficial effects of the root node for solving the problem can be seen in any one of the methods and the beneficial effects of the third to eighth aspects, the specific designs of the third to eighth aspects, the eleventh, twelfth, fifteenth to eighteenth aspects, and the specific designs of the fifteenth to eighteenth aspects, and the repeated parts are not repeated.

In a twentieth aspect, there is provided a network device, comprising: a processor, a memory, a communication interface; the processor, the communication interface and the memory are connected; wherein the communication interface may be a transceiver. The communication interface is used for realizing communication with other network elements. Wherein one or more programs are stored in the memory, and the processor calls the program stored in the memory to implement the method of any one of the third to eighth aspects, the specific designs of the third to eighth aspects, the eleventh aspect, the twelfth aspect, the specific designs of the eleventh aspect, the specific designs of the twelfth aspect, the fifteenth to eighteenth aspects, and the specific designs of the fifteenth to eighteenth aspects, and the embodiments and advantages of the network device for solving the problem can be seen in the specific designs of the third to eighth aspects, the eleventh aspect, the twelfth aspect, the fifteenth to eighteenth aspects, and the method of the seventh to eighteenth aspects, The method and the advantageous effects of any one of the specific designs of the fifteenth aspect to the eighteenth aspect are not repeated.

A twenty-first aspect provides a computer program product which, when run on a computer, causes the computer to perform the method of any one of the third to eighth aspects described above, the specific designs of the third to eighth aspects, the eleventh, twelfth, fifteenth to eighteenth aspects, and the specific designs of the fifteenth to eighteenth aspects.

A twenty-second aspect provides a chip product, wherein the method of any one of the third to eighth aspects, the specific designs of the third to eighth aspects, the eleventh, twelfth, fifteenth to eighteenth aspects, and the fifteenth to eighteenth aspects is performed.

A twenty-third aspect provides a computer-readable storage medium having stored therein instructions which, when run on a computer, cause the computer to perform the method of any one of the above-described third to eighth aspects, the specific designs of the eleventh aspect, the twelfth aspect, the specific designs of the eleventh aspect, the twelfth aspect, the fifteenth to eighteenth aspects, and the specific designs of the fifteenth aspect to the eighteenth aspect.

Drawings

Fig. 1 is a schematic diagram of a conventional communication system;

fig. 2 is a schematic diagram of a communication system provided in an embodiment of the present application;

fig. 3 is a schematic flowchart of a gateway selection method according to an embodiment of the present application;

fig. 4 is a schematic flowchart of another gateway selection method provided in an embodiment of the present application;

fig. 5 is a schematic flowchart of another gateway selection method provided in an embodiment of the present application;

fig. 6 is a schematic flowchart of another gateway selection method provided in an embodiment of the present application;

fig. 7 is a schematic flowchart of another gateway selection method provided in an embodiment of the present application;

fig. 8 is a flowchart illustrating a service recovery method according to an embodiment of the present application;

fig. 9 is a schematic flowchart of another service recovery method provided in an embodiment of the present application;

fig. 10 is a schematic flowchart of another service recovery method provided in an embodiment of the present application;

fig. 11 is a schematic structural diagram of a network device according to an embodiment of the present application;

fig. 12 is a schematic structural diagram of another terminal device provided in an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the embodiments of the present invention will be described below with reference to the accompanying drawings.

In order to better understand the embodiments of the present application, the following description is provided for a system architecture to which the embodiments of the present application can be applied:

referring to fig. 2, fig. 2 is a schematic diagram of a communication system according to an embodiment of the present disclosure. As shown in fig. 2, the communication system at least includes a Mobility Management Entity (MME), a plurality of serving gateway control plane function entities (SGW-C), a plurality of serving gateway user plane function entities (SGW-U), a plurality of packet data network gateway control plane function entities (PGW-C), and a plurality of packet data network gateway user plane function entities (PGW-U). In fig. 2, the communication system includes 2 SGW-cs, 4 SGW-us, 2 PGW-cs and 4 PGW-us as an example, but the communication system may include more SGW-C, SGW-U, PGW-cs and PGW-us. Alternatively, the communication system may include less than 4 SGW-us and less than 4 PGW-us, which is not limited in the embodiments of the present application.

One SGW-C can be connected with one or more SGW-Us, one SGW-U can also be connected with one or more SGW-Cs, and the SGW-Cs and the SGW-Us communicate through Sxa interfaces. One SGW-C can be connected with one or more PGW-Cs, one PGW-C can also be connected with one or more SGW-Cs, and the SGW-C and the PGW-C communicate through an S5/8-C interface. One PGW-C can be connected with one or more PGW-U, one PGW-U can also be connected with one or more PGW-C, and the PGW-C and the PGW-U are communicated through an Sxb interface. One SGW-U can be connected with one or more PGW-Us, one PGW-U can also be connected with one or more SGW-Us, and the SGW-U and the PGW-U are communicated through an S5/8-U interface.

The MME can be used for access control, mobility management, attachment and detachment, session management functions, SGW-C and PGW-C selection and the like. For other functions and descriptions of the MME, reference may be made to descriptions in relevant 3GPP TS 23.401 and other standards, which are not described herein again.

The SGW-C is used for managing or selecting the SGW-U or realizing other non-session-level management functions. The PGW-C is used to manage or select the PGW-U, or implement other non-session level management functions. The SGW-U and the PGW-U are used for forwarding IP packets, carrying/APN flow attention, service detection and the like. Other functions and descriptions of the SGW-C, PGW-C, SGW-U and the PGW-U can be found in the descriptions of 3GPP TS 23.214 and other standards, which are not repeated herein.

As shown in fig. 2, the communication system may further include an evolved universal mobile telecommunications system terrestrial radio access network (E-UTRAN) and a User Equipment (UE). Fig. 2 illustrates an example including one UE, and certainly, the communication system may further include a plurality of UEs, which is not limited in the embodiment of the present application. Wherein the SGW-U communicates with the E-UTRAN over the S1-U interface. The E-UTRAN communicates with the UE over an LTE-Uu interface.

Wherein, the E-UTRAN comprises an access network device. The access network device may provide communication coverage for a particular geographic area and may communicate with user devices located within the coverage area, may support different systems of communication protocols, or may support different communication modes. For example, the access network device may be an evolved node B (eNB or eNodeB) in an LTE system, or a radio network controller in a Cloud Radio Access Network (CRAN), or may be an access network device in a 5G network, such as a gNB, or may be a small station, a micro station, or a Transmission Reception Point (TRP), or may be an access network device in a relay station, an access point, or a Public Land Mobile Network (PLMN) for future evolution, and the like.

The UE may refer to, among other things, an access terminal, a subscriber unit, a subscriber station, a mobile station, a remote terminal, a terminal device, a mobile terminal, a user terminal, a wireless communication device, a user agent, or a user equipment. An access terminal may be a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA), a handheld device with wireless communication capability, a computing device or other processing device connected to a wireless modem, a vehicle mounted device, a wearable device, a terminal device in the internet of things, a virtual reality device, a terminal device in a fifth generation (5G) network, a terminal device in a future evolved Public Land Mobile Network (PLMN), or the like.

Wherein, the MME, the SGW-C, SGW-U, PGW-C and the PGW-U are devices in a core network. The UE, the E-UTRAN, and the core network constitute an Evolved Packet System (EPS). As shown in fig. 2, the communication system may also include a Packet Data Network (PDN). And the PDN and the PGW-U are communicated through an SGi interface.

In an LTE network a UE must connect to at least one PDN in order to perform the work of data communication. The PDN connection established between the UE and the PDN may also be referred to as a session. The UE may receive data from the PDN and may also send data to the PDN in one session, which transfers data between the UE and the PDN over an EPS bearer. The EPS bearer can bear data of services such as video phones, network videos, VoLTE and the like. If the SGW-U is abnormal, a channel between the UE and the PDN is interrupted, so that the service of the session cannot be normally carried out. Therefore, the SGW-U needs to be selected again for the affected session.

Therefore, the embodiment of the application provides a gateway selection system and method, which can reselect an SGW-U for a session affected by an abnormal SGW-U when the SGW-U fails. Wherein the gateway selection system may comprise a network element in the communication system.

Based on the above communication system, please refer to fig. 3, and fig. 3 is a flowchart illustrating a gateway selection method according to an embodiment of the present application. The method described in fig. 3 is used for timely reselecting the SGW-U for the session affected by the abnormal SGW-U when the SGW-U is abnormal. In the method depicted in fig. 3, the first session affected by the failed SGW-U is logged by the MME. The MME may actively reselect the SGW-U for the first session. Alternatively, the MME may initiate a procedure of reselecting the SGW-U for the first session upon receiving signaling of the first session from the user equipment side. As shown in FIG. 3, the gateway selection method comprises the following parts 301-304, wherein:

301. and when detecting that a first SGW-U docked with the first SGW-C is abnormal, the first SGW-C sends an identifier of the first SGW-U to the MME.

In the embodiment of the application, the first SGW-C is any one of the SGW-cs in the gateway selection system. The first SGW-U is any one of the SGW-Us interfaced with the first SGW-C. For example, the first SGW-C may be SGW-C1 or SGW-C2 in FIG. 2. When the first SGW-C is SGW-C1, the first SGW-U is SGW-U1 or SGW-U2 in FIG. 2. When the first SGW-C is SGW-C2, the first SGW-U is SGW-U3 or SGW-U4 in FIG. 2.

The identification of the first SGW-U can be an IP address of the first SGW-U.

The first SGW-U is abnormal, which means that the first SGW-U fails or the first SGW-U is restarted.

Wherein, the first SGW-C can determine whether the first SGW-U is abnormal or not through a HeatBeat message of the Sx interface. For example, the first SGW-C may send a HeatBeat request message to the first SGW-U at a preset time period. After receiving the heatbed request message, the first SGW-U returns a heatbed response message to the first SGW-C. And if the first SGW-C does not receive the HeatBeat response message sent by the first SGW-U within the preset time length, the first SGW-C determines that the first SGW-U fails. If the first SGW-C receives the HeatBeat response message, but the value of the recovery counter in the HeatBeat response message is not the same as the value of the recovery counter in the HeatBeat response message received last time, the first SGW-C determines that the first SGW-U is restarted. If the first SGW-U reboots, the value of the reset counter in the HeatBeat response message is incremented by 1. Thus, when the first SGW-C detects that the value in the revert counter in the heattreat response message is not the same as the value of the revert counter in the last received heattreat response message, the first SGW-C may determine that the first SGW-U has restarted.

Of course, the first SGW-C may also determine whether the first SGW-U is abnormal through other manners, which is not limited in the embodiment of the present application.

The first SGW-C can send the identifier of the first SGW-U to the MME through an Echo message of a Node level existing in a protocol. Alternatively, the first SGW-C may send the identity of the first SGW-U to the MME by protocol an existing session-level session management/mobility management message. Alternatively, the first SGW-C may send the identity of the first SGW-U to the MME via a private Node-level message defined outside the protocol.

302. And the MME determines a first session influenced by the first SGW-U according to the identification of the first SGW-U.

In the embodiment of the application, after the MME receives the identifier of the first SGW-U from the first SGW-C, the MME determines, according to the identifier of the first SGW-U, the first session affected by the first SGW-U from the sessions included in the MME. Due to the failure of the first SGW-U, the traffic using the session in the first SGW-U may be affected. Thus, the first session affected by the first SGW-U refers to a session using the MME and the first SGW-U.

In one particular design, the MME may record the identity of the SGW-U used for the session in the session. The specific implementation manner of the MME determining the first session affected by the first SGW-U according to the identifier of the first SGW-U is as follows: the MME determines a session of the sessions having an identification of the first SGW-U as a first session affected by the first SGW-U. Based on this specific design, the MME can accurately determine the first session affected by the first SGW-U.

For example, session 1, session 2, and session 3 use the MME, and thus the MME has session 1, session 2, and session 3 therein. Session 1 and Session 2 also use SGW-C1, SGW-U1, PGW-C1, and PGW-U1. Session 3 also uses SGW-C2, SGW-U3, PGW-C2, and PGW-U3. Then the MME will record the identity of SGW-U1 in session 1 and session 2 and record the identity of SGW-U3 in session 3. If the first SGW-U is the SGW-U1, after the MME receives the identification of the SGW-U1, the MME searches the identification of the SGW-U1 in the session of the MME. The MME finds the identity of SGW-U1 in session 1 and session 2. Therefore, the MME determines session 1 and session 2 as the first session determined to be affected by the SGW-U1.

Throughout the embodiments of the present application, an SGW-U used in a session may also be referred to as an SGW-U in which the session is located or referred to as an SGW-U corresponding to the session.

In a specific design, after the MME determines the first session affected by the first SGW-U according to the identifier of the first SGW-U, the MME needs to record the first session affected by the first SGW-U so as to subsequently reselect the second non-failed SGW-C for the first session. For example, the MME may record the first session affected by the first SGW-U by: the MME marks the first session. Alternatively, the MME flags the first session. Alternatively, the MME creates a list of sessions affected by the first SGW-U and adds the first session to the list. For example, after the MME determines session 1 and session 2 as the first session affected by the SGW-U1, session 1 and session 2 are marked. Alternatively, the MME flags the session 1 and session 2 bits. Alternatively, the MME adds session 1 and session 2 to the list of sessions affected by the first SGW-U.

In a particular design, the MME de-registers the first session after the MME reselects the second SGW-C for the first session that is not failed. Therefore, the MME can subsequently perform normal processing on the service of the first session. For example, after the MME reselects the second SGW-C that has not failed for the first session, the MME unmarks the first session. Or, the MME cancels the flag bit set for the first session. Alternatively, the MME removes the first session from the list of sessions affected by the first SGW-U.

303. The MME reselects the second SGW-C that has not failed for the first session.

In the embodiment of the application, after the MME determines the first session, the second SGW-C which is not failed is reselected for the first session, so that the subsequent second SGW-C can provide services for the service of the first session. The second SGW-C may be the same as or different from the first SGW-C, and the embodiments of the present application are not limited thereto.

For example, the first SGW-C is SGW-C1. Session 1 and session 2 are the first sessions affected by SGW-U1. Then the MME reselects the non-failed second SGW-C for session 1 and the non-failed second SGW-C for session 2. The second SGW-C may be the SGW-C1 or the SGW-C2, i.e., the MME may select the original SGW-C1 to continue serving the first session or select the other non-failed SGW-C2 to continue serving the first session. The second SGW-C reselected by the MME for session 1 may be the same as or different from the second SGW-C reselected for session 2, which is not limited in this embodiment of the present application.

The MME may reselect the second SGW-C that did not fail for the first session in two ways.

In a first way, there are multiple affected first sessions, and even if the MME does not receive signaling of the first session, the MME may actively reselect the second SGW-C that does not fail for the multiple first sessions according to a preset rate.

In the second mode, when the MME receives the uplink signaling of the first session, the MME preferentially reselects the second SGW-C which does not fail for the first session. For example, the uplink signaling may be a service request or the like. For example, the affected first session includes session 1 to session 100, and the MME reselects the second SGW-C that does not fail for session 1 to session 100, respectively, according to a preset rate. The MME preferentially selects the second SGW-C that does not fail for the session 100 when receiving a service request for the session 100 from the user equipment. Due to the large number of affected first sessions, the MME will spend a long time to re-select the non-failed second SGW-C for these 100 sessions in bulk, which may result in some emergency traffic not being able to be recovered in time. In the specific design, when the MME receives a signaling sent by the user equipment side, it may trigger reselection of the SGW-C for the corresponding session, which is beneficial to timely recovery of the emergency service through the specific design.

304. The second SGW-C reselects the second SGW-U that has not failed for the first session.

In the embodiment of the application, after the MME reselects the second SGW-C that has not failed for the first session to provide service, the second SGW-C reselects the second SGW-U that has not failed for the first session, so that the subsequent second SGW-U can provide service for the service of the first session. For example, as shown in fig. 2, the SGW-U1 is abnormal, and session 1 in the MME is affected by SGW-U1. The MME reselects the non-failed SGW-C2 for session 1. The SGW-C2 serves the first session to reselect the non-failed SGW-U3 or SGW-U4.

After the MME reselects the second SGW-C which does not fail for the first session, and the second SGW-C reselects the second SGW-U which does not fail for the first session, some related operations are needed to recover the service. For example, after the MME reselects the second SGW-C that does not fail for the first session, the MME may send a create session request (create session request) to the second SGW-C that does not fail to modify the bearer for the first session. And after receiving the session creation request, the second SGW-C sends a modified bearer request (modify bearer request) to the first PGW-C. And the first PGW-C is a PGW-C which is in butt joint with the first SGW-C. Specifically, after receiving the bearer modification request, the first PGW-C sends a session modification request (session modification request) to the first PGW-U. The first PGW-U is a PGW-U which is in butt joint with the first PGW-C. Specifically, after receiving the session modification request, the first PGW-U updates S5S8 the tunnel of the interface, and sends a session modification response to the first PGW-C. For example, the tunnel may be a general packet radio service tunneling protocol (GTPU) tunnel. And after receiving the session modification response, the first PGW-C sends a bearer modification response to the second SGW-C. And after the second SGW-C receives the bearer modification response, the second SGW-C sends a session creation response to the MME.

For another example, after the second SGW-C reselects a second SGW-U that does not fail for the first Session, the second SGW-C may further send a Packet Forwarding Control Protocol (PFCP) Session initiation request to the second SGW-U to establish a Session between the second SGW-C and the second SGW-U. Accordingly, after the second SGW-U receives the PFCP Session initiation request, a PFCP Session initiation response may be sent to the second SGW-C.

For how to recover the service after the MME reselects the second SGW-C that has not failed for the first session, and after the second SGW-C reselects the second SGW-U that has not failed for the first session, reference may be made to the procedures in the standards of 3GPP TS 23.007, 3GPP TS 23.214, 3GPP TS 29.244, and the like, which are not described herein again.

It can be seen that by implementing the method described in fig. 3, the MME can reselect the second SGW-C that has not failed for the first session, and the second SGW-C can reselect the second SGW-U that has not failed for the first session. The first session may then be subsequently serviced by the second SGW-C and the second SGW-U.

Based on the above communication system, please refer to fig. 4, and fig. 4 is a flowchart illustrating another gateway selection method according to an embodiment of the present application. The method described in figure 4 is used to timely reselect SGW-us for sessions affected by an abnormal SGW-U when the SGW-U experiences an abnormality. In the method depicted in fig. 4, the second session affected by the failed SGW-U is recorded by the PGW-C. And when receiving the downlink signaling of the second session, the PGW-C initiates a process of reselecting the SGW-U for the second session. As shown in fig. 4, the gateway selection method includes the following parts 401 to 405, wherein:

401. the first PGW-C receives an identification of the first SGW-U where the exception occurred from the first PGW-U or the first SGW-C.

Wherein the first SGW-C is any one of the gateway selection systems. The first PGW-C is docked with the first PGW-U and the first SGW-C, respectively. The first SGW-U interfaces with the first PGW-U and the first SGW-C, respectively.

For example, as shown in FIG. 2, the first SGW-C is SGW-C1 and the first PGW-C is PGW-C1. The first SGW-U is SGW-U1, and the first PGW-U is PGW-U1. Whether an SGW-U1 is abnormal or not can be detected by the PGW-U1, and when the SGW-U1 is detected to be abnormal, the identification of the SGW-U1 is sent to the PGW-C1 so as to inform the PGW-C1 of the SGW-U1 with the abnormality. The SGW-C1 may also detect whether an SGW-U1 is abnormal, and when detecting that SGW-U1 is abnormal, send an identifier of SGW-U1 to PGW-C1 to notify PGW-C1 of the abnormal SGW-U1.

The first PGW-U can detect whether the first SGW-U is abnormal or not through an Echo request message. For example, the first PGW-U may send an Echo request message to the first SGW-U at a preset time period. After receiving the Echo request message, the first SGW-U returns an Echo response message to the first PGW-U. And if the first PGW-U does not receive the Echo response message sent by the first SGW-U within the preset time length, the first PGW-U determines that the first SGW-U fails. If the first PGW-U receives the Echo response message, but the value of a Recovery counter (Recovery counter) in the Echo response message is not the same as the value of the Recovery counter in the Echo response message received last time, the first PGW-U determines that the first SGW-U is restarted. If the first SGW-U reboots, the value of the reset counter in the Echo response message is incremented by 1. Therefore, when the first PGW-U detects that the value of the reversion counter in the Echo response message is not the same as the value of the reversion counter in the last received Echo response message, the first PGW-U may determine that the first SGW-U is restarted.

Of course, the first PGW-U may also determine whether the first SGW-U is abnormal through other manners, which is not limited in the embodiment of the present application.

The first PGW-U may send the identifier of the first SGW-U to the first PGW-C through a Node level message. For example, the Node level message may be a Node report request message. Alternatively, the first PGW-U may send the identity of the first SGW-U to the first PGW-C via a private Node-level message defined outside the protocol.

For a specific implementation manner of how the first SGW-C detects whether the first SGW-U is abnormal, reference may be made to the description corresponding to step 301, which is not described herein again.

The first SGW-C can send the identifier of the first SGW-U to the first PGW-C through an Echo message of a Node level existing in a protocol. Alternatively, the first SGW-C may send the identity of the first SGW-U to the first PGW-C by protocol an existing session-level session management/mobility management message. Alternatively, the first SGW-C may send the identity of the first SGW-U to the first PGW-C via a private Node-level message defined outside the protocol.

402. And the first PGW-C determines a second session influenced by the first SGW-U according to the identification of the first SGW-U.

In the embodiment of the application, after receiving the identifier of the first SGW-U, the first PGW-C determines, according to the identifier of the first SGW-U, a second session affected by the first SGW-U. The second session affected by the first SGW-U is a session using the first PGW-C and the first SGW-U.

In a particular design, the first SGW-C may further send, to the first PGW-C, an identification of the SGW-U used for the session; after the first PGW-C receives the identification of the SGW-U used by the session from the first SGW-C, recording the identification of the SGW-U corresponding to the session in the session; the specific implementation manner of the first PGW-C determining, according to the identifier of the first SGW-U, the second session affected by the first SGW-U is as follows: the first PGW-C determines that a session of the sessions having an identification of the first SGW-U is a second session affected by the first SGW-U. Based on this specific design, the first PGW-C can accurately determine the second session affected by the first SGW-U.

For example, the first PGW-C is PGW-C1, and the first SGW-U is SGW-U1. Session 1, session 2 and session 3 use PGW-C1. Session 1 and session 2 also use SGW-U1. Session 3 also uses SGW-U2. The SGW-C1 may send the SGW-U1 used for session 1, the SGW-U1 used for session 2, and the SGW-U2 used for session 3 to PGW-C1. After PGW-C1 receives SGW-U1 used for session 1, SGW-U1 used for session 2, and SGW-U2 used for session 2, PGW-C1 records the identity of SGW-U1 in session 1 and session 2, and records the identity of SGW-U2 in session 3. After receiving the identification of the SGW-U1, the PGW-C1 searches for the identification of the SGW-U1 in sessions 1 to 3 of the PGW-C1. The PGW-C1 finds the identity of the SGW-U1 in session 1 and session 2. Accordingly, the PGW-C1 determines session 1 and session 2 as the second session determined to be affected by the SGW-U1.

In a specific design, after the first PGW-C determines, according to the identifier of the first SGW-U, the second session affected by the first SGW-U, the first PGW-C needs to record the second session affected by the first SGW-U, so as to subsequently trigger a process of reselecting an non-failed SGW-U for the second session. For example, the first PGW-C may record the second session affected by the first SGW-U by: the first PGW-C marks the second session. Alternatively, the first PGW-C flags the second session. Alternatively, the first PGW-C creates a list of sessions affected by the first SGW-U and adds the second session to the list.

In a specific design, after the first PGW-C receives, from the third SGW-C, a modified bearer request corresponding to the second session, the PGW-C cancels the recording of the second session. For example, after the first PGW-C receives the bearer modification request corresponding to the second session from the third SGW-C, the first PGW-C cancels marking of the second session. Alternatively, the first PGW-C cancels the flag set for the second session. Alternatively, the first PGW-C removes the second session from the list of sessions affected by the first SGW-U. Thus, the first PGW-C may subsequently perform normal processing on the traffic of the second session.

403. And the first PGW-C sends a first message to the MME when receiving the downlink signaling of the second session.

For example, the first PGW-C may send a first message to the MME upon receiving downlink signaling for the second session from a Policy and Charging Rules Function (PCRF). Alternatively, the first PGW-C may send the first message to the MME upon receiving downlink signaling of the second session from another device on the network side.

Specifically, the first PGW-C sends the first message to the MME through the SGW-C when receiving the downlink signaling of the second session. The first PGW-C may send the first message to the MME through the SGW-C where the second session is located or send the first message to the MME through other non-failed SGW-Cs.

Wherein the first message is used to trigger the MME to reselect an non-failed SGW-C for the second session. For example, the first message may be a packet data network gateway downlink triggering notification (PDTN) message. Accordingly, after receiving the PDTN message, the MME may send a packet data network gateway downlink triggering acknowledgement (PDTA) message to the first PGW-C. Alternatively, the first message may also be other messages, which is not limited in this embodiment of the present application.

For example, the second session includes session 1 and session 2. And when receiving the signaling of the session 1, the first PGW-C sends a first message to the MME, wherein the first message is used for triggering the MME to reselect a non-failed third SGW-C for the session 1. When receiving signaling of session 2, the first PGW-C sends a first message to the MME, where the first message is used to trigger the MME to reselect a non-failed third SGW-C for session 2.

404. And the MME reselects a third non-failed SGW-C for the second session according to the first message.

In the embodiment of the application, after the MME receives the first message from the first PGW-C, the MME reselects a third SGW-C that has not failed for the second session according to the first message. The third SGW-C may be the same as or different from the first SGW-C, and the embodiments of the present application are not limited thereto.

The principle of the MME reselecting a non-failed third SGW-C for the second session is the same as the principle of the MME reselecting a non-failed second SGW-C for the first session in fig. 3, and details are not described herein.

405. The third SGW-C reselects the non-failed third SGW-U for the second session.

The principle of the third SGW-C reselecting a non-failed third SGW-U for the second session is the same as the principle of the second SGW-C reselecting a non-failed second SGW-U for the first session in fig. 3, which is not described herein again.

In a specific design, steps 403-405 may not be performed. When the first PGW-C receives the downlink signaling of the second session, the first PGW-C may directly send a target message to the SGW-C used by the second session, where the target message is used to trigger the SGW-C used by the second session to reselect a non-failed SGW-U message for the second session. And after the SGW-C used by the second session receives the target message, reselecting a third SGW-U which does not fail for the second session. In this particular design, rather than reselecting a third non-failed SGW-C, the SGW-C originally used for the second session may reselect a non-failed SGW-U for the second session.

By implementing the method described in fig. 4, the first PGW-C can send a first message to the MME when receiving signaling of the affected second session, so that the MME reselects a third SGW-C that is not failed for the second session, and the third SGW-C reselects a third SGW-U that is not failed for the second session. It can be seen that by implementing the method described in fig. 4, the reselection of the SGW-U affected by the abnormal SGW-U can be triggered in time through the signaling sent by the network side, which is beneficial to timely recovering the affected service.

Referring to fig. 5, fig. 5 is a schematic diagram of another gateway selection method according to an embodiment of the present application. The method described in fig. 5 is used to timely reselect SGW-U for a session affected by an abnormal SGW-U when the SGW-U is abnormal. In the method depicted in fig. 5, the second session affected by the failed SGW-U is recorded by the SGW-C. And the SGW-C initiates a process of reselecting the SGW-U for the second session when receiving the downlink signaling of the second session. As shown in fig. 5, the gateway selection method includes the following parts 501-507, wherein:

501. and the first SGW-C determines a second session influenced by the first SGW-U when detecting that the first SGW-U is abnormal.

The first SGW-C detects whether the first SGW-U is abnormal or not, and determines a second session influenced by the first SGW-U when the first SGW-U is detected to be abnormal. The second session affected by the first SGW-U refers to a session using the first SGW-C and the first SGW-U. For a specific implementation of the first SGW-C detecting whether the first SGW-U is abnormal, reference may be made to the description of the specific implementation of the first SGW-C detecting whether the first SGW-U is abnormal in the embodiment corresponding to fig. 3, which is not described herein again.

In one particular design, the first SGW-C may record in the session an identification of the SGW-U used by the session; the first SGW-C determines that the second session affected by the first SGW-U is implemented as: the first SGW-C determines that a session of the sessions having the identity of the first SGW-U is a second session affected by the first SGW-U. Based on the specific design, the first SGW-C can accurately determine the second session affected by the first SGW-U.

For example, the first SGW-C is SGW-C1, and the first SGW-U is SGW-U1. Session 1, session 2 and session 3 use SGW-C1. Session 1 and session 2 also use SGW-U1. Session 3 also uses SGW-U2. Then the SGW-C1 records the identity of SGW-U1 in session 1 and session 2 and the identity of SGW-U2 in session 3. And if the SGW-C1 detects that the SGW-U1 is abnormal, searching the SGW-C1 session for the identification of the SGW-U1. The SGW-C1 finds the identity of SGW-U1 in session 1 and session 2. Therefore, the SGW-C1 determines session 1 and session 2 as the second session determined to be affected by the SGW-U1.

In one specific design, after the first SGW-C determines the second session affected by the first SGW-U, the first SGW-C needs to record the second session affected by the first SGW-U, so as to detect whether signaling of the session affected by the first SGW-U is received subsequently. For example, the first SGW-C may record the second session affected by the first SGW-U by: the first SGW-C marks the second session. Alternatively, the first SGW-C flags the second session. Alternatively, the first SGW-C creates a list of sessions affected by the first SGW-U and adds the second session to the list.

502. The first PGW-C receives first signaling corresponding to the second session.

The first PGW-C is a PGW-C docked with the first SGW-C, and the first PGW-C has a second session. The first signaling may be sent by the PCRF to the first PGW-C, or may be sent by other devices to the first PGW-C, which is not limited in this embodiment of the application.

503. And the first PGW-C sends a second signaling corresponding to the second session to the first SGW-C according to the first signaling.

In this embodiment of the application, after the first PGW-C receives the first signaling corresponding to the second session, the first PGW-C sends the second signaling corresponding to the second session to the first SGW-C.

504. And the first SGW-C sends a response of the second signaling to the first PGW-C.

In this embodiment of the application, after receiving the second signaling from the first PGW-C, the first SGW-C determines that the second signaling is a second signaling corresponding to the second session. Therefore, the first SGW-C sends a response of the second signaling to the first PGW-C. The response of the second signaling carries a reason value for indicating that the context of the second session is not found.

The first signaling may be a re-authentication request (RAR), and the second signaling may be a create bearer request (create bearer request) or an update bearer request (update bearer request). And the creating bearing request corresponding to the second session is used for creating the exclusive bearing of the second session. The update bearer request corresponding to the second session is used for updating the bearer of the second session. Alternatively, the first signaling and the second signaling may also be other signaling, and the embodiment of the present application is not limited.

In a particular design, the first SGW-C may also delete the second session after receiving the create-dedicated-bearer request from the first PGW-C. The first SGW-C may delete the second session while sending the create-dedicated bearer response to the first PGW-C, or delete the second session after sending the create-dedicated bearer response to the first PGW-C, which is not limited in the embodiment of the present application. Since the SGW-C needs to be reselected for the second session, the first SGW-C keeps the second session as it is, and the first SGW-C needs to delete the second session.

505. The first PGW-C sends a first message to the MME.

In the embodiment of the present application, after receiving the response of the second signaling from the first SGW-C, the first PGW-C sends the first message to the MME. The first message is used to trigger the MME to reselect an non-failed SGW-C for the second session. For example, the first message may be a packet data network gateway downlink triggering notification (PDTN) message. Accordingly, after receiving the PDTN message, the MME may send a packet data network gateway downlink triggering acknowledgement (PDTA) message to the first PGW-C. Alternatively, the first message may also be other messages, which is not limited in this embodiment of the present application.

Specifically, the first PGW-C sends a first message to the MME through the SGW-C. The first PGW-C may send the first message to the MME through the first SGW-C or send the first message to the MME through other non-failed SGW-Cs.

506. And the MME reselects a third non-failed SGW-C for the second session according to the first message.

In the embodiment of the application, after receiving the first message from the first PGW-C, the MME reselects a third SGW-C that does not fail for the second session according to the first message. The third SGW-C may be the same as or different from the first SGW-C, and the embodiments of the present application are not limited thereto. The principle of the MME reselecting a non-failed third SGW-C for the second session is the same as the principle of the MME reselecting a non-failed second SGW-C for the first session in fig. 3, and details are not described herein.

507. The third SGW-C reselects the non-failed third SGW-U for the second session.

In a specific design, steps 504-507 may not be performed. And after the first SGW-C receives a second signaling corresponding to the second session from the first PGW-C, the first SGW-C directly reselects an unserviced SGW-U for the second session. That is, in this particular design, rather than reselecting an un-failed third SGW-C, an un-failed SGW-U may be reselected for the second session from the first SGW-C originally used for the second session.

By implementing the method described in fig. 5, the first SGW-C is able to send, when receiving the second signaling of the affected second session, a response of the second signaling to the first PGW-C, where the response of the second signaling carries a cause value indicating that the context of the second session is not found. After the first PGW-C receives the response of the second signaling, the first PGW-C sends a first message to the MME to trigger the MME to reselect the non-failed third SGW-C for the second session, and reselect the non-failed third SGW-U for the second session by the third SGW-C. It can be seen that by implementing the method described in fig. 5, the reselection of the SGW-U for the session affected by the abnormal SGW-U can be triggered in time through the signaling sent by the network side, which is beneficial to timely recovering the affected service.

Referring to fig. 6, fig. 6 is another gateway selection method provided in the embodiment of the present application. The method described in fig. 6 is used to timely reselect SGW-U for a session affected by an abnormal SGW-U when the SGW-U is abnormal. In the method depicted in fig. 6, the third session affected by the failed SGW-U is recorded by the PGW-U. And when receiving the downlink data message of the third session, the PGW-U initiates a process of reselecting the SGW-U for the third session. As shown in fig. 6, the gateway selection method includes the following sections 601-605, wherein:

601. and the first PGW-U determines a third session influenced by the first SGW-U when detecting that the first SGW-U is abnormal.

The first PGW-U is any one PGW-U of the gateway selection system. The first SGW-U is any one SGW-U which is in butt joint with the first PGW-U in the gateway selection system. For example, the first PGW-U is PGW-U1, and the first SGW-U is SGW-U1. The first PGW-U is PGW-U3, and the first SGW-U is SGW-U3.

The specific implementation of the first PGW-U detecting the abnormality of the first SGW-U may specifically refer to the description in the embodiment corresponding to fig. 4, and is not described herein again.

Wherein the third session affected by the first SGW-U is a session using the first PGW-U and the first SGW-U.

In one particular design, the first PGW-U may record, in the session, an identification of the SGW-U used by the session; the specific implementation manner of the first PGW-U determining the third session affected by the first SGW-U is as follows: the first PGW-U determines that a session having an identification of the first SGW-U among the sessions is a third session affected by the first SGW-U. Based on the specific design, the first PGW-U can accurately determine the third session affected by the first SGW-U.

For example, the first PGW-U is PGW-U1, and the first SGW-U is SGW-U1. Session 1, session 2 and session 3 use PGW-U1. Session 1 and session 2 also use SGW-U1. Session 3 also uses SGW-U2. Then PGW-U1 will record the identity of SGW-U1 in session 1 and session 2 and the identity of SGW-U2 in session 3. And if the PGW-U1 detects that the SGW-U1 is abnormal, searching the session of the PGW-U1 for the identification of the SGW-U1. The PGW-U1 finds the identity of the SGW-U1 in session 1 and session 2. Accordingly, the PGW-U1 determines session 1 and session 2 as the third session determined to be affected by the SGW-U1.

In a specific design, after the first PGW-U determines the second session affected by the first SGW-U, the first PGW-U needs to record a third session affected by the first SGW-U, so as to detect whether a downlink data packet of the session affected by the first SGW-U is received subsequently. For example, the first PGW-U may record the third session affected by the first SGW-U by: the first PGW-U marks the third session. Alternatively, the first PGW-U flags the third session. Alternatively, the first PGW-U creates a list of sessions affected by the first SGW-U and adds the third session to the list.

In one particular design, a first PGW-U receives a session modification request from a first PGW-C; the first PGW-U cancels the recording of the third session. For example, the first PGW-U cancels the marking of session 1. Alternatively, the first PGW-U cancels the flag bit set to session 1. Alternatively, the first PGW-U removes the session list affected by the first SGW-U for session 1. Based on the design, the first PGW-U receiving the session modification request from the first PGW-C indicates that the fourth SGW-U that has not failed has been reselected for the third session, and the traffic of the subsequent third session can be restored. Therefore, at this time, the first PGW-U needs to cancel the recording of the third session, and the subsequent first PGW-U can normally process the service of the third session.

602. And when receiving the downlink data message of the third session, the first PGW-U sends a notification message to the first PGW-C.

In the embodiment of the application, after the first PGW-U determines the third session affected by the first SGW-U, if a downlink data packet of the third session is received, the first PGW-U sends a notification message to the first PGW-C. The notification message is used for notifying that the third session receives a downlink data packet, and the first PGW-C is a PGW-C docked with the first PGW-U. The notification message may be session-level information, such as a session report request (session report request).

603. And the first PGW-C sends a second message to the MME.

In this embodiment, after receiving the notification message from the first PGW-U, the first PGW-C sends a second message to the MME. The second message is used to trigger the MME to reselect an non-failed SGW-C for the third session. For example, the second message may be a packet data network gateway downlink triggering notification (PDTN) message. Accordingly, after receiving the PDTN message, the MME may send a packet data network gateway downlink triggering acknowledgement (PDTA) message to the first PGW-C. Alternatively, the second message may also be other messages, which is not limited in this embodiment of the present application.

Specifically, after receiving the notification message from the first PGW-U, the first PGW-C sends a second message to the MME through the SGW-C. The first PGW-C may send the first message to the MME through the SGW-C used for the third session or send the first message to the MME through other non-failed SGW-cs.

604. And the MME reselects a fourth non-failed SGW-C for the third session according to the second message.

In the embodiment of the present application, after receiving the second message from the PGW-C, the MME reselects a fourth SGW-C that has not failed for the third session according to the second message. The principle of the MME reselecting a fourth SGW-C that does not fail for the third session is the same as the principle of the MME reselecting a second SGW-C that does not fail for the first session in fig. 3, and details are not described herein.

605. The fourth SGW-C reselects the fourth SGW-U that did not fail for the third session.

The principle of the fourth SGW-C reselecting a non-failed fourth SGW-U for the third session is the same as the principle of the second SGW-C reselecting a non-failed second SGW-U for the first session in fig. 3, and details are not described herein.

In a specific design, steps 603 to 605 may not be performed. When receiving a notification message for notifying that the third session receives the downlink data packet from the first PGW-U, the first PGW-C may directly send a target message to the SGW-C used by the third session, where the target message is used to trigger the SGW-C used by the third session to reselect an SGW-U that has not failed for the third session. And after the SGW-C used by the third session receives the target message, reselecting an SGW-U which does not fail for the third session. In this particular design, rather than reselecting an un-failed third SGW-C, an un-failed SGW-U may be reselected for the third session from the SGW-C originally used for the third session.

By implementing the method described in fig. 6, the first PGW-U may send the second message to the MME when receiving the downlink data packet of the affected third session, so as to trigger the MME to reselect the fourth SGW-C that has not failed for the third session, and reselect the fourth SGW-U that has not failed for the third session by the fourth SGW-C. It can be seen that by implementing the method described in fig. 6, the reselection of the SGW-U for the session affected by the abnormal SGW-U can be triggered in time through the downlink data packet sent by the network side, which is beneficial to timely recovering the affected service.

It is worth mentioning that an SGW-U exception affects all sessions using that SGW-U. Each session may receive a signaling from the user equipment side, a signaling from the network side, or a downlink data packet. Thus, the methods described in fig. 3, 4, 6 may be combined with each other. Alternatively, the methods described in fig. 3, 5, 6 are combined with each other. Thus, upon receiving upstream signaling for the affected session, the process of reselecting the SGW-U for the affected session may be initiated by the method described in fig. 3. Upon receiving the downlink signaling of the affected session, the process of reselecting the SGW-U for the affected session may be initiated by the method described in fig. 4 or fig. 5. Upon receiving the downlink data message for the affected session, the process of reselecting the SGW-U for the affected session may be initiated by the method described in fig. 6. By combining the methods described in fig. 3, 4 and 6 with each other or combining the methods described in fig. 3, 5 and 6 with each other, the process of reselecting an SGW-U for an affected session can be initiated in a number of scenarios.

For example, the methods described in fig. 3, 4 and 6 are combined with each other. The MME may determine a first session affected by the first SGW-U and actively reselect the SGW-U for the first session by the MME. Alternatively, the MME may initiate a procedure of reselecting the SGW-U for the first session upon receiving signaling of the first session from the user equipment side. The first PGW-C may also determine a second session affected by the first SGW-U, and initiate a procedure of reselecting the SGW-U for the first session upon receiving signaling of the second session from the network side. The first PGW-U may also determine a third session affected by the first SGW-U, and initiate a process of reselecting the SGW-U for the third session when receiving a downlink data packet of the third session.

As another example, the methods described in fig. 3, 5 and 6 are combined with each other. The MME may determine a first session affected by the first SGW-U and actively reselect the SGW-U for the first session by the MME. Alternatively, the MME may initiate a procedure of reselecting the SGW-U for the first session upon receiving signaling of the first session from the user equipment side. The first SGW-C may also determine a second session affected by the first SGW-U and initiate a procedure of reselecting the SGW-U for the first session upon receiving signaling of the second session from the network side. The first PGW-U may also determine a third session affected by the first SGW-U, and initiate a process of reselecting the SGW-U for the third session when receiving a downlink data packet of the third session.

In the above-described communication system, not only the SGW-U but also the SGW-C may be abnormal in some cases. If the SGW-C is abnormal, the channel between the UE and the PDN is interrupted, so that the service of the session cannot be normally carried out. Therefore, the SGW-C needs to be selected again for the affected session as well. Therefore, the embodiment of the application also provides a gateway selection system and a method, which are used for reselecting the SGW-C for the session affected by the abnormal SGW-C. The gateway selection system comprises a network element in the communication system.

Referring to fig. 7, fig. 7 is a schematic diagram of another gateway selection method according to an embodiment of the present application. The method described in fig. 7 is used to timely reselect SGW-C for a session affected by an abnormal SGW-C when the SGW-C is abnormal. In the method depicted in FIG. 7, the first session affected by the failed SGW-C is recorded by the PGW-U. And when receiving the downlink data message of the first session, the PGW-U initiates a process of reselecting the SGW-C for the first session. As shown in fig. 7, the gateway selection method includes the following parts 701-705, wherein:

701. the first PGW-U receives an identification of the first SGW-C where the anomaly occurred from the first PGW-C or the first SGW-U.

The first PGW-U may be any one PGW-U in the gateway selection system. The first PGW-C is any one PGW-C which is in butt joint with the first PGW-U. The first SGW-U is any one SGW-U which is in butt joint with the first PGW-U. The first SGW-C is any one SGW-C which is in butt joint with the first PGW-C or the first SGW-U.

For example, as shown in FIG. 2, the first PGW-U is PGW-U1, the first PGW-C is PGW-C1, the first SGW-U is SGW-U1, and the first SGW-C is SGW-C1. PGW-U1 may receive an identification of the SGW-C1 that the anomaly occurred from PGW-C1 or SGW-U1. That is, PGW-C1 or SGW-U1 will detect if an abnormality occurs in SGW-C1. And if the SGW-C1 is detected to be abnormal, sending the identification of the SGW-C1 to the PGW-U1.

The first PGW-C can detect whether the first SGW-C is abnormal through an Echo request message. For example, the first PGW-C may send an Echo request message to the first SGW-C at a preset time period. After receiving the Echo request message, the first SGW-C returns an Echo response message to the first PGW-C. And if the first PGW-C does not receive the Echo response message sent by the first SGW-C within the preset time length, the first PGW-C determines that the first SGW-C fails. If the first PGW-C receives the Echo response message, but the value of a Recovery counter (Recovery counter) in the Echo response message is not the same as the value of the Recovery counter in the Echo response message received last time, the first PGW-C determines that the first SGW-C is restarted. If the first SGW-C restarts, the value of the reset counter in the Echo response message is incremented by 1. Therefore, when the first PGW-C detects that the value of the reversion counter in the Echo response message is not the same as the value of the reversion counter in the last received Echo response message, the first PGW-C may determine that the first SGW-C reboots.

Of course, the first PGW-C may also detect whether the first SGW-C is abnormal through other methods, which is not limited in the embodiment of the present application.

The first PGW-C may send the identifier of the first SGW-C in which the exception occurs to the first PGW-U through a message defined by an existing protocol, for example, an Sx Association Update Request message. Or, the first PGW-C may send the identifier of the first SGW-C in which the abnormality occurs to the first PGW-U through other messages.

Wherein, the first SGW-U can detect whether the first SGW-C is abnormal or not through a HeatBeat message. For example, the first SGW-U may send a HeatBeat request message to the first SGW-C at a preset time period. After receiving the heatbed request message, the first SGW-C returns a heatbed response message to the first SGW-U. And if the first SGW-U does not receive the HeatBeat response message sent by the first SGW-C within the preset time length, the first SGW-U determines that the first SGW-C fails. If the first SGW-U receives the HeatBeat response message, but the value in the Recovery counter (Recovery counter) in the HeatBeat response message is not the same as the value of the Recovery counter in the HeatBeat response message last received, the first SGW-U determines that a restart has occurred at the first SGW-C. If the first SGW-C reboots, the value of the reset counter in the HeatBeat response message is incremented by 1. Thus, when the first SGW-U detects that the value in the recovery counter in the heattreat response message is not the same as the value of the recovery counter in the last received heattreat response message, the first SGW-U may determine that the first SGW-C has restarted.

Of course, the first SGW-U may also detect whether the first SGW-C is abnormal through other methods, which is not limited in this embodiment of the present application.

The first SGW-U can send the identifier of the first SGW-C to the first PGW-U through an Echo message of a Node level existing in a protocol. Alternatively, the identity of the first SGW-C may be sent to the first PGW-U by a session management/mobility management message of the existing session level of the protocol. Alternatively, the identity of the first SGW-C may be sent to the first PGW-U via a private Node-level message defined outside the protocol.

702. And the first PGW-U determines a first session influenced by the first SGW-C according to the identification of the first SGW-C.

In the embodiment of the application, after the first PGW-U receives the identifier of the first SGW-C with the abnormality from the first PGW-C or the first SGW-U, the first session influenced by the first SGW-C is determined according to the identifier of the first SGW-C. The first session affected by the first SGW-C is a session using the first PGW-U and the first SGW-C.

In one particular design, a first PGW-U receives, from a first PGW-C, an identification of an SGW-C used for a session; a first PGW-U records the identification of an SGW-C corresponding to the session in the session; the specific implementation manner of the first PGW-U determining, according to the identifier of the first SGW-C, the first session affected by the first SGW-C is as follows: the first PGW-U determines that a session with the identification of the first SGW-C among the sessions is a first session affected by the first SGW-C. Based on the specific design, the first PGW-U can accurately determine the first session affected by the first SGW-C.

For example, the first PGW-U is PGW-U1, and the first SGW-C is SGW-C1. Session 1, session 2 and session 3 use PGW-U1. Session 1 and session 2 also use SGW-C1. Session 3 also uses SGW-C2. Then PGW-U1 would record the identity of SGW-C1 in session 1 and session 2 and the identity of SGW-C2 in session 3. If the first SGW-C is SGW-C1, after the MME receives the identification of SGW-C1, the MME searches the identification of SGW-C1 in the session of the MME. The MME finds the identity of SGW-C1 in session 1 and session 2. Therefore, the MME determines session 1 and session 2 as the first session determined to be affected by SGW-C1.

Throughout the embodiments of the present application, the SGW-C used for a session may also be referred to as the SGW-C where the session is located or referred to as the SGW-C corresponding to the session.

In one particular design, the first PGW-U records the first session after determining the affected first session. For example, the first PGW-U may record the first session affected by the first SGW-C by: the first PGW-U tags the first session. Alternatively, the first PGW-U flags the first session. Alternatively, the first PGW-U creates a list of sessions affected by the first SGW-C and adds the first session to the list. In one particular design, the first PGW-U cancels recording of the first session after the first PGW-U receives the session modification request from the second SGW-C. For example, the first PGW-U cancels the marking of the first session. Alternatively, the first PGW-U cancels the flag bit set for the first session. Alternatively, the first PGW-U removes the first session from the list of sessions affected by the first SGW-U. Thus, the first PGW-U can subsequently perform normal processing on the service of the first session.

703. And when receiving the downlink data message of the first session, the first PGW-U sends a notification message to the first PGW-C.

The notification message is used for notifying that the downlink data message of the first session is received. The notification message may be a session level message, such as a session report request (session report request), or other message.

704. The first PGW-C sends a first message to the MME.

Specifically, after receiving the notification message from the first PGW-U, the first PGW-C sends a first message to the MME. The first message is used to trigger the MME to reselect an non-failed SGW-C for the first session. For example, the first message may be a packet data network gateway downlink triggering notification (PDTN) message. Accordingly, after receiving the PDTN message, the MME may send a packet data network gateway downlink triggering acknowledgement (PDTA) message to the first PGW-C. Alternatively, the first message may also be other messages, which is not limited in this embodiment of the present application.

Specifically, the first PGW-C sends a first message to the MME through the non-failed SGW-C.

705. And the MME reselects a second SGW-C which does not fail for the first session according to the first message.

In the embodiment of the application, after receiving the first message, the MME reselects a second SGW-C that has not failed for the first session according to the first message. The principle of the MME reselecting a second SGW-C that does not fail for the first session is the same as the principle of the MME reselecting a second SGW-C that does not fail for the first session in fig. 3, and is not described herein again.

By implementing the method described in fig. 7, the first PGW-U can send a first message to the MME when receiving the downlink data packet of the affected first session, so that the MME reselects the second SGW-C that has not failed for the first session. It can be seen that by implementing the method described in fig. 7, the reselection of the SGW-C for the session affected by the abnormal SGW-C can be triggered in time through the downlink data packet sent by the network side, which is beneficial to timely recovery of the affected service.

In the above-described communication system, not only the SGW-U and SGW-C but also the PGW-U may be abnormal in some cases. If the PGW-U is abnormal, the channel between the UE and the PDN may be interrupted, which may cause the service of the session to not be performed normally. Therefore, the embodiment of the application also provides a service recovery system and a method, which are used for recovering the service affected by the abnormal PGW-U. The service recovery system may include a network element in the communication system.

Referring to fig. 8, fig. 8 is a diagram of a service recovery method according to an embodiment of the present application. The method described in fig. 8 is used for recovering the service affected by the abnormal PGW-U when the PGW-U is abnormal. In the method described in fig. 8, the MME records the first session affected by the failed PGW-U, and the MME may actively initiate a flow for recovering the traffic of the first session. Or, the MME may also initiate a procedure of recovering the service of the first session when receiving the uplink signaling of the first session. As shown in fig. 8, the service recovery method includes the following 801-804 parts, wherein:

801. the first SGW-C receives an identification of the first PGW-U where the abnormality occurs from the first PGW-C or the first SGW-U.

The first SGW-C is any one of the service recovery systems, the first PGW-C is a PGW-C in the service recovery system and in butt joint with the first SGW-C, the first SGW-U is an SGW-U in the service recovery system and in butt joint with the first SGW-C, and the first PGW-U is in butt joint with the first PGW-C and the first SGW-C respectively. For example, as shown in FIG. 2, the first SGW-C is GW-C1, the first SGW-U is SGW-U1, the first PGW-C is PGW-C1, and the first PGW-U is PGW-U1.

The first SGW-C may receive the identification of the first PGW-U where the exception occurred in two ways.

In a first mode, the first SGW-C receives the identification of the first PGW-U with the exception from the first SGW-U. For example, the first SGW-U may detect whether the first PGW-U has failed. And if the first SGW-U detects that the first PGW-U fails, the first SGW-U sends the identifier of the first PGW-U to the first SGW-C. The first SGW-U may send the identifier of the first PGW-U to the first SGW-C through a Node level message or a session management/mobility management message. For example, the Node-level message may be a Node report request message. Alternatively, the first SGW-U may send the identity of the first PGW-U to the first SGW-C by defining a private Node-level message outside the protocol.

The first SGW-U can detect whether the first PGW-U is abnormal or not through an Echo request message. For example, the first SGW-U may send an Echo request message to the first PGW-U at a preset time period. After receiving the Echo request message, the first PGW-U returns an Echo response message to the first SGW-U. And if the first SGW-U does not receive the Echo response message sent by the first PGW-U within the preset time length, the first SGW-U determines that the first PGW-U fails. If the first SGW-U receives the Echo response message, but the value of a Recovery counter (Recovery counter) in the Echo response message is not the same as the value of the Recovery counter in the Echo response message received last time, the first SGW-U determines that the first PGW-U is restarted. If the first PGW-U restarts, the value of the reset counter in the Echo response message is incremented by 1. Therefore, when the first SGW-U detects that the value of the recovery counter in the Echo response message is not the same as the value of the recovery counter in the last received Echo response message, the first SGW-U may determine that the first PGW-U is restarted.

Of course, the first SGW-U may also detect whether the first PGW-U is abnormal through other methods, which is not limited in the embodiment of the present application.

In a second mode, the first SGW-C receives the identification of the first PGW-U with the exception from the first PGW-C. For example, the first PGW-C may detect whether the first PGW-U has failed. And if the first PGW-C detects that the first PGW-U fails, the first PGW-C sends the identifier of the first PGW-U to the first SGW-C. The first PGW-C may send the identifier of the first PGW-U to the first SGW-C through a Node level message or a session management/mobility management message. For example, the Node-level message may be a Node report request message. Alternatively, the first PGW-C may send the identifier of the first PGW-U to the first SGW-C by defining a private Node-level message outside the protocol.

The first PGW-C can determine whether the first PGW-U is abnormal through a HeatBeat message of the Sx interface. For example, the first PGW-C may transmit a heatbed request message to the first PGW-U at a preset time period. After receiving the heatbed request message, the first PGW-U returns a heatbed response message to the first PGW-C. And if the first PGW-C does not receive a HeatBeat response message sent by the first PGW-U within the preset time length, the first PGW-C determines that the first PGW-U fails. If the first PGW-C receives the HeatBeat response message but the value of the Recovery counter (Recovery counter) in the HeatBeat response message is not the same as the value of the Recovery counter in the last received HeatBeat response message, the first PGW-C determines that the first PGW-U is restarted. If the first PGW-U restarts, the value of the reset counter in the HeatBeat response message is incremented by 1. Therefore, when the first PGW-C detects that the value of the reset counter in the heattreat response message is not the same as the value of the reset counter in the heattreat response message received last time, the first PGW-C may determine that the first PGW-U is restarted.

Of course, the first PGW-C may also determine whether the first PGW-U is abnormal through other manners, which is not limited in the embodiment of the present application.

802. And the first SGW-C sends the identification of the abnormal first PGW-U to the MME.

In this embodiment of the application, specifically, after receiving the identifier of the abnormal first PGW-U, the first SGW-C sends the identifier of the abnormal first PGW-U to the MME.

803. And the MME determines a first session influenced by the first PGW-U according to the identification of the first PGW-U.

In the embodiment of the application, after receiving the identifier of the first PGW-U from the first SGW-C, the MME determines, according to the identifier of the first PGW-U, a first session affected by the first PGW-U. The first session affected by the first PGW-U refers to a session using the MME and the first PGW-U.

In one particular design, the MME records an identification of PGW-U used for the session in the session; the specific implementation manner of the MME determining the first session affected by the first PGW-U according to the identifier of the first PGW-U is: the MME determines a session with the identification of the first PGW-U in the session to be the first session affected by the first PGW-U. Based on this specific design, the MME can accurately determine the first session affected by the first PGW-U.

For example, session 1, session 2, and session 3 use MMEs. Session 1 and Session 2 also use SGW-C1, SGW-U1, PGW-C1, and PGW-U1. Session 3 also uses SGW-C2, SGW-U3, PGW-C2, and PGW-U3. Then the MME will record the identity of PGW-U1 in session 1 and session 2 and the identity of PGW-U3 in session 3. If the first PGW-U is PGW-U1, after the MME receives the identification of PGW-U1, the MME searches the identification of PGW-U1 in the session of the MME. The MME finds the identity of PGW-U1 in session 1 and session 2. Therefore, the MME determines session 1 and session 2 as the first session determined to be affected by PGW-U1.

Throughout the embodiments of the present application, a PGW-U used for a session may also be referred to as a PGW-U where the session is located or referred to as a PGW-U corresponding to the session.

In a particular design, the MME may record the first session affected by the first PGW-U after determining the first session based on the identity of the first PGW-U. For example, the MME may record the first session affected by the first PGW-U by: the MME marks the first session. Alternatively, the MME flags the first session. Alternatively, the MME creates a list of sessions affected by the first PGW-U and adds the first session to the list.

804. And the MME deletes the first session and instructs the user equipment corresponding to the first session to reestablish the PDN connection.

In the embodiment of the application, after determining the first session affected by the first PGW-U according to the identifier of the first PGW-U, the MME deletes the first session, and instructs the ue corresponding to the first session to reestablish the PDN connection. That is, the MME will delete the first session and instruct the user equipment to re-establish the session with the PDN. After the user equipment and the session between the user equipment and the PDN are reestablished, the service of the user equipment can be recovered to be normal.

In one particular design, the MME may also instruct the first SGW-C to delete the first session. And after receiving the indication of the MME for deleting the first session, the first SGW-C deletes the first session. And the first SGW-C sending an indication to the first SGW-U and the first PGW-C to delete the first session. After the first SGW-U and the first PGW-C receive the indication to delete the first session, the first session is also deleted.

The MME may delete the first session and instruct the ue corresponding to the first session to reestablish the PDN connection in the following two ways.

In a first manner, the affected first sessions have multiple sessions, and the MME deletes the multiple first sessions according to a preset rate, and instructs the ue corresponding to the first session to reestablish the PDN connection.

In a second mode, when the MME receives the uplink signaling of the first session, the first session is deleted, and the user equipment corresponding to the first session is instructed to reestablish the PDN connection. The uplink signaling may be a service request or other uplink signaling. For example, the affected first session includes session 1 to session 100, and the MME deletes session 1 to session 100 according to the preset rate and instructs the ue corresponding to the session to reestablish the PDN connection. When receiving a service request of the session 100 from the user equipment, the MME preferentially deletes the session 100 and instructs the user equipment corresponding to the session 100 to reestablish the PDN connection. Due to the large number of affected first sessions, the service recovery of these 100 sessions will take a long time, which may result in some emergency services not being recovered in time. In the specific design, when the MME receives the signaling sent by the user equipment side, the service corresponding to the signaling can be triggered to be recovered, so that by the design, it is beneficial to recover the emergency service in time.

It can be seen that by implementing the method described in fig. 8, the affected traffic can be recovered.

Referring to fig. 9, fig. 9 is a diagram illustrating another service recovery method according to an embodiment of the present application. The method described in fig. 9 is used to timely recover the service affected by the abnormal PGW-U when the PGW-U is abnormal. In the method depicted in fig. 9, the SGW-U records the second session affected by the failed PGW-U. And when the SGW-U receives the uplink data message of the second session, initiating a process of recovering the service of the second session. As shown in fig. 9, the service recovery method includes the following components 901 to 903, where:

901. and when the first SGW-U determines that the first PGW-U is abnormal, determining a second session influenced by the first PGW-U.

The first SGW-U is any one SGW-U, and the first PGW-U is any one PGW-U butted with the SGW-U. For example, as shown in FIG. 2, when the first SGW-U is SGW-U1, the first PGW-U is PGW-U1. When the first SGW-U is SGW-U3, the first PGW-U is PGW-U3.

For a specific implementation of how the first SGW-U determines that the first PGW-U is abnormal, reference may be made to the description in the embodiment corresponding to fig. 8, which is not described herein again.

Wherein the second session affected by the first PGW-U is a session using the first SGW-U and the first PGW-U.

In one particular design, the first SGW-U records, in the session, an identification of a PGW-U used by the session; the specific implementation manner of the first SGW-U determining, according to the identifier of the first PGW-U, the second session affected by the first PGW-U is as follows: the first SGW-U determines that the session with the identification of the first PGW-U in the sessions is a second session affected by the first PGW-U. Based on the specific design, the first SGW-UE can accurately determine the second session affected by the first PGW-U.

For example, session 1, session 2, session use SGW-U1. Session 1 and session 2 also use PGW-U1. Session 3 also uses PGW-U2. Then the SGW-U1 would record the identity of PGW-U1 in session 1 and session 2 and the identity of PGW-U2 in session 3. If the first PGW-U is PGW-U1, after the SGW-U1 receives the identification of PGW-U1, the identification of PGW-U1 is looked up in the session of SGW-U1. The SGW-U1 finds the identity of PGW-U1 in session 1 and session 2. Therefore, the SGW-U1 determines session 1 and session 2 as the second session determined to be affected by the PGW-U1.

In a specific design, after the first SGW-U determines, according to the identifier of the first PGW-U, a second session affected by the first PGW-U, the second session may be recorded. For example, the first SGW-U may record the second session affected by the first PGW-U by: the first SGW-U marks the second session. Alternatively, the first SGW-U flags the second session. Or the first SGW-U creates a session list influenced by the first PGW-U and adds the second session to the list.

902. And the first SGW-U sends a notification message to the first SGW-C when receiving the uplink data message of the second session.

In the embodiment of the application, after determining a second session affected by a first PGW-U, a first SGW-U sends a notification message to a first SGW-C when receiving an uplink data packet of the second session. The notification message is used to notify the second session that the uplink data packet is received. The uplink data message refers to a data message from a user side. The notification message may be a message defined by an existing protocol, for example, the notification message may be a session report request message. The notification message may also be a message defined privately outside the protocol.

903. And the first SGW-C deletes the second session and instructs the user equipment corresponding to the second session to reestablish the PDN connection.

Specifically, after receiving the notification message from the first SGW-U, the first SGW-C deletes the second session, and instructs the ue corresponding to the second session to reestablish the PDN connection. That is, the first SGW-C may delete the second session and instruct the user equipment to re-establish the session with the PDN. After the user equipment and the session between the user equipment and the PDN are reestablished, the service of the user equipment can be recovered to be normal.

In a particular design, the first SGW-C may also instruct the MME, the first SGW-U, and the first PGW-C to delete the second session. And after the MME receives the indication for deleting the second session, deleting the second session. And after receiving the indication for deleting the second session, the first SGW-U deletes the second session. And after receiving the indication for deleting the second session, the first PGW-C deletes the second session.

Through the method described in fig. 9, the first SGW-U can trigger the recovery of the service of the second session in time when receiving the uplink data packet of the second session.

Referring to fig. 10, fig. 10 is a diagram illustrating another service recovery method according to an embodiment of the present application. The method described in fig. 10 is used for recovering the service affected by the abnormal PGW-U in time when the PGW-U is abnormal. In the method depicted in fig. 10, the PGW-C records a third session affected by the failed PGW-U. And when the PGW-C receives the uplink data message of the third session, initiating a process of recovering the service of the third session. As shown in fig. 10, the service recovery method includes the following 1001-1002 parts:

1001. and the first PGW-C determines a third session influenced by the first PGW-U when detecting that the first PGW-U is abnormal.

In the embodiment of the application, the first PGW-C is any one PGW-C in service restoration. The first PGW-U is any one of the PGW-U interfacing with the first PGW-C. For example, as shown in FIG. 2, the first PGW-C is PGW-C1, and the first PGW-U is PGW-U1 or PGW-U2. The first PGW-C is PGW-C2, and the first PGW-U is PGW-U3 or PGW-U4.

Wherein, the third session influenced by the first PGW-U refers to the session existing in the first PGW-C and the first PGW-U.

For a specific implementation of the first PGW-C detecting the abnormality of the first PGW-U, reference may be made to the description in the embodiment corresponding to fig. 7, which is not repeated herein.

In one particular design, the first PGW-C records, in the session, an identification of a PGW-U used by the session; the specific implementation manner of the first PGW-C determining, according to the identifier of the first PGW-U, the third session affected by the first PGW-U is as follows: the first PGW-C determines that a session with the identification of the first PGW-U in the sessions is a third session affected by the first PGW-U. Based on the specific design, the first PGW-C can accurately determine the third session affected by the first PGW-U.

For example, the first PGW-C includes session 1, session 2, session 3. Session 1 and session 2 use PGW-U1. Session 3 also uses PGW-U2. Then the first PGW-C will record the identity of the PGW-U1 in session 1 and session 2 and the identity of the PGW-U2 in session 3. If the first PGW-U is the PGW-U1, the first PGW-C searches the identification of the PGW-U1 in the session of the first PGW-C after receiving the identification of the PGW-U1. The first PGW-C finds the identity of the PGW-U1 in session 1 and session 2. Thus, the first PGW-C determines session 1 and session 2 as the third session determined to be affected by PGW-U1.

In a specific design, after the first PGW-C determines, according to the identifier of the first PGW-U, a third session affected by the first PGW-U, the third session may be recorded. For example, the first PGW-C may record the third session affected by the first PGW-U by: the first PGW-C marks the third session. Alternatively, the first PGW-C flags the third session. Alternatively, the first PGW-C creates a list of sessions affected by the first PGW-U, and adds the third session to the list.

1002. And when receiving the downlink signaling of the third session, the first PGW-C deletes the third session and instructs the user equipment corresponding to the third session to reestablish the PDN connection.

Specifically, after the first PGW-C determines a third session affected by the first PGW-U, if a downlink signaling of the third session is received, the third session is deleted, and the user equipment corresponding to the third session is instructed to reestablish the PDN connection. That is, the first PGW-C may delete the third session and instruct the user equipment to re-establish the session with the PDN. After the user equipment and the session between the user equipment and the PDN are reestablished, the service of the user equipment can be normally carried out.

The signaling may be signaling from the PCRF, for example, a re-authentication request message. Or the signaling may also be signaling from other network elements on the network side, which is not limited in this embodiment of the application.

In one particular design, the first PGW-C may also instruct the first SGW-C to delete the third session. And after receiving the indication for deleting the third session, the first SGW-C deletes the third session. After receiving the indication to delete the third session, the first SGW-U, MME is also instructed to delete the third session. And after the MME receives the indication for deleting the third session, deleting the third session. And after receiving the indication for deleting the third session, the first SGW-U deletes the third session.

In a specific design, when a first PGW-C receives a signaling of a fourth session, the first PGW-C sends a message to a second PGW-U according to the signaling of the fourth session; after the first PGW-C retransmits the message to the second PGW-U, if the first PGW-C does not receive the response message of the message within the preset time period, the fourth session is deleted, and the user equipment corresponding to the fourth session is instructed to reestablish the PDN connection. The second PGW-U is any one PGW-U except the first PGW-U and butted with the first PGW-C. The fourth session is a session existing in the first PGW-C and in the second PGW-U. For example, as shown in FIG. 2, the first PGW-C is PGW-C1, the first PGW-U is PGW-U1, and the second PGW-U is PGW-U2. The PGW-C1 detects an abnormality in the PGW-U1 and determines that the session affected by the PGW-U1 is session 1. The PGW-C1 deletes session 1 and instructs the user equipment of session 1 to re-establish the PDN connection to resume the traffic of session 1. If the PGW-U2 is also abnormal but the PGW-C1 has not sensed it, at this time, if the PGW-C1 receives the signaling of session 2, such as the re-authentication request message, the PGW-C1 sends information to the PGW-U2 according to the signaling of session 2. If the PGW-C1 does not receive a response from the PGW-U2 for the information, the PGW-C1 resends the information to the PGW-U2. If the PGW-C1 is not receiving the response message of the message within the preset time period after the PGW-C1 retransmits the message to the PGW-U2, the PGW-C1 deletes the session 2 and instructs the user equipment corresponding to the session 2 to reestablish the PDN connection. Based on the specific design, when the PGW-C1 does not timely sense the occurrence of the abnormality of the PGW-U2, the service influenced by the PGW-U2 can be timely recovered.

It can be seen that, by the method described in fig. 10, the first PGW-C can trigger the recovery of the traffic of the third session in time when receiving the signaling of the third session.

It is worth mentioning that an exception to a PGW-U will affect all sessions using that PGW-U. Each session may receive an uplink signaling, an uplink data packet, or a downlink signaling. Thus, the methods described in fig. 8, 9, 10 may be combined with each other. Thus, receiving the upstream session, the service affected by the abnormal PGW-U can be recovered by the method described in fig. 8. After receiving the session of the uplink data packet, the method described in fig. 9 may be used to recover the service affected by the abnormal PGW-U. The session receiving the downlink signaling can recover the service affected by the abnormal PGW-U by the method described in fig. 10. By combining the methods described in fig. 8, 9, and 10, the service affected by the abnormal PGW-U can be recovered in a plurality of scenarios.

For example, the methods described in fig. 8, 9, and 10 are combined. The MME may determine a first session affected by the abnormal first PGW-U, and actively initiate a procedure for recovering a service of the first session by the MME. Or, the MME may initiate a procedure for recovering the service of the first session when receiving the uplink signaling of the first session. The first SGW-U may also determine a second session affected by the anomalous first PGW-U. And when receiving the uplink data message of the second session, the first SGW-U initiates a process of recovering the service of the second session. The first PGW-C may also determine a third session affected by the abnormal first PGW-U, and initiate a procedure of recovering the service of the third session when receiving a downlink signaling of the third session.

In the embodiment of the present invention, the device may be divided into the functional modules according to the method example, for example, each functional module may be divided corresponding to each function, or two or more functions may be integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. It should be noted that, the division of the modules in the embodiment of the present invention is schematic, and is only a logic function division, and there may be another division manner in actual implementation.

Referring to fig. 11, fig. 11 is a schematic structural diagram of a network device according to an embodiment of the present application. The network device shown in fig. 11 may be configured to perform part or all of the functions of the MME in the method embodiment described in fig. 3. The network device shown in fig. 11 may include a processing module 1101 and a communication module 1102. Wherein:

a communication module 1102, configured to receive, from a first serving gateway control plane functional entity SGW-C, an identifier of a first serving gateway user plane functional entity SGW-U where an abnormality occurs, where the first SGW-C is docked with the first SGW-U; a processing module 1101, configured to determine, according to an identifier of the first SGW-U, a first session affected by the first SGW-U; the processing module 1101 is further configured to reselect a second SGW-C that has not failed for the first session.

In one particular design, the processing module 1101 is further configured to record, in the session, an identification of an SGW-U used by the session; the way for the processing module 1101 to determine the first session affected by the first SGW-U according to the identifier of the first SGW-U is specifically: determining that a session of the sessions having an identification of the first SGW-U is a first session affected by the first SGW-U.

In a specific design, the processing module 1101 reselects the second SGW-C that has not failed for the first session in a manner that: when the communication module 1102 receives upstream signaling for the first session, the processing module 1101 reselects the non-failed second SGW-C for the first session.

Referring to fig. 11, fig. 11 is a schematic structural diagram of a network device according to an embodiment of the present application. The network device shown in fig. 11 may be configured to perform some or all of the functions of the first PGW-C in the method embodiment described in fig. 4 above. The network device shown in fig. 11 may include a processing module 1101 and a communication module 1102. Wherein:

a communication module 1102, configured to receive, from a first packet data network gateway user plane functional entity PGW-U or a first serving gateway control plane functional entity SGW-C, an identifier of a first serving gateway user plane functional entity SGW-U in which an abnormality occurs, where a network device is docked with the first PGW-U, and the network device is docked with the first SGW-C, the first SGW-U is docked with the first PGW-U, and the first SGW-U is docked with the first SGW-C; a processing module 1101, configured to determine, according to the identifier of the first SGW-U, a second session affected by the first SGW-U; the communication module 1102 is further configured to send a first message to the mobility management entity MME when receiving the downlink signaling of the second session, where the first message is used to trigger the MME to reselect an non-failed SGW-C for the second session.

In a particular design, the communication module 1102 is further configured to receive, from the first SGW-C, an identification of an SGW-U used for the session; the processing module 1101 is further configured to record, in a session, an identifier of an SGW-U corresponding to the session; the way for the processing module 1101 to determine, according to the identifier of the first SGW-U, the second session affected by the first SGW-U is specifically: determining a session having an identification of the first SGW-U among the sessions as a second session affected by the first SGW-U.

Referring to fig. 11, fig. 11 is a schematic structural diagram of a network device according to an embodiment of the present application. The network device shown in fig. 11 may be used to perform some or all of the functions of the first SGW-C in the method embodiment described above in fig. 5. The network device shown in fig. 11 may include a processing module 1101 and a communication module 1102. Wherein:

the processing module 1101 is configured to determine, when detecting that an SGW-U of a first service gateway user plane functional entity that is docked is abnormal, a second session that is affected by the first SGW-U; a communication module 1102, configured to receive a second signaling corresponding to a second session from a first packet data network gateway control plane function entity PGW-C; the communication module 1102 is further configured to send a response of the second signaling to the first PGW-C, where the response of the second signaling carries a cause value indicating that the context of the second session is not found.

In one particular design, the processing module 1101 is further configured to record, in the session, an identification of an SGW-U used by the session; the way for the processing module 1101 to determine the second session affected by the first SGW-U is specifically: determining a session having an identification of the first SGW-U among the sessions as a second session affected by the first SGW-U.

In a specific design, the processing module 1101 is further configured to delete the second session after the communication module 1102 receives a request for creating a dedicated bearer corresponding to the second session.

The embodiment of the application provides network equipment. The network device may be configured to perform some or all of the functionality of the first PGW-C in the method embodiment described above with reference to fig. 5. The network device may include a communication module. Wherein:

a communication module, configured to receive a first signaling; the communication module is further configured to send a second signaling corresponding to a second session to a first serving gateway control plane functional entity SGW-C according to the first signaling; the communication module is further configured to receive a response of a second signaling from the first SGW-C, where the response of the second signaling carries a cause value used for indicating that a context of a second session is not found, and the second session is a session in the first SGW-C that is affected by an abnormal first serving gateway user plane function entity SGW-U; the communication module is further configured to send a first message to the mobility management entity MME, where the first message is used to trigger the MME to reselect a non-failed third SGW-C for the second session.

Referring to fig. 11, fig. 11 is a schematic structural diagram of a network device according to an embodiment of the present application. The network device shown in fig. 11 may be configured to perform some or all of the functions of the first PGW-U in the method embodiment described above with reference to fig. 6. The network device shown in fig. 11 may include a processing module 1101 and a communication module 1102. Wherein:

the processing module 1101 is configured to determine, when detecting that an SGW-U of a first service gateway user plane functional entity that is docked is abnormal, a third session that is affected by the first SGW-U; the communication module 1102 is configured to send a notification message to the first packet data network gateway control plane function entity PGW-C that is docked when the downlink data packet of the third session is received, where the notification message notifies the third session of receiving the downlink data packet.

In one particular design, the processing module 1101 is further configured to record, in the session, an identification of an SGW-U used by the session; the way for the processing module 1101 to determine the third session affected by the first SGW-U is specifically: determining a session having an identification of the first SGW-U among the sessions as a third session affected by the first SGW-U.

The embodiment of the application provides network equipment. The network device may be configured to perform some or all of the functionality of the first PGW-C in the method embodiment described above with reference to fig. 6. The network device may include a communication module. Wherein:

the communication module is used for receiving a notification message from a first packet data network gateway user plane functional entity (PGW-U), wherein the notification message is used for notifying that a downlink data message of a third session is received, and the third session is a session influenced by an abnormal first service gateway user plane functional entity (SGW-U) in the first PGW-U; the communication module is further configured to send a first message to the mobility management entity MME, where the first message is used for the MME to reselect an non-failed SGW-C for the third session.

Referring to fig. 11, fig. 11 is a schematic structural diagram of a network device according to an embodiment of the present application. The network device shown in fig. 11 may be configured to perform some or all of the functions of the first PGW-U in the method embodiment described above with reference to fig. 7. The network device shown in fig. 11 may include a processing module 1101 and a communication module 1102. Wherein:

a communication module 1102, configured to receive, from a first packet data network gateway control plane functional entity PGW-C or a first serving gateway user plane functional entity SGW-U, an identifier of the first serving gateway control plane functional entity SGW-C where an abnormality occurs, where the first PGW-C and the first SGW-U are respectively docked with a network device, and the first PGW-C and the first SGW-U are respectively docked with the first SGW-C; a processing module 1101, configured to determine, according to an identifier of the first SGW-C, a first session affected by the first SGW-C; the communication module 1102 is further configured to send a notification message to the first PGW-C when receiving the downlink data packet of the first session, where the notification message is used to notify that the downlink data packet of the first session is received.

In a particular design, the communication module 1102 is further configured to receive, from the first PGW-C, an identification of an SGW-C used for the session; a processing module 1101, further configured to record, in the session, an identification of an SGW-C used by the session; the way for the processing module 1101 to determine, according to the identifier of the first SGW-C, the first session affected by the first SGW-C is specifically: determining that a session of the sessions having an identification of the first SGW-C is a first session affected by the first SGW-C.

The embodiment of the application provides network equipment. The network device may be configured to perform some or all of the functionality of the first PGW-C in the method embodiment described above with reference to fig. 7. The network device may include a communication module. Wherein:

a communication module, configured to receive a notification message from a first packet data network gateway user plane function entity PGW-U, where the notification message is used to notify that a downlink data packet of a first session is received, where the first session is a session in the first PGW-U that is affected by an abnormal first serving gateway control plane function entity SGW-C; the communication module is further configured to send a first message to the mobility management entity MME, where the first message is used to trigger the MME to reselect an non-failed SGW-C for the first session.

In a specific design, the communication module is further configured to send, to the first PGW-U, an identifier of the first SGW-C when it is determined that the first SGW-C is abnormal.

In a specific design, the communication module is further configured to send, to the first PGW-U, an identification of an SGW-C corresponding to the session.

Referring to fig. 11, fig. 11 is a schematic structural diagram of a network device according to an embodiment of the present application. The network device shown in fig. 11 may be configured to perform part or all of the functions of the MME in the method embodiment described in fig. 8. The network device shown in fig. 11 may include a processing module 1101 and a communication module 1102. Wherein:

a communication module 1102, configured to receive, from a first serving gateway control plane functional entity SGW-C, an identifier of a first packet data network gateway user plane functional entity PGW-U where an exception occurs; the processing module 1101 is configured to determine, according to the identifier of the first PGW-U, a first session affected by the first PGW-U; the processing module 1101 is further configured to delete the first session, and instruct the ue corresponding to the first session to reestablish the PDN connection.

In one particular design, the processing module 1101 is further configured to record, in the session, an identification of a PGW-U used by the session; the specific way for the processing module 1101 to determine the first session affected by the first PGW-U according to the identifier of the first PGW-U is as follows: determining that a session having an identification of a first PGW-U among the sessions is a first session affected by the first PGW-U.

In a specific design, the manner for the processing module 1101 to delete the first session and instruct the ue corresponding to the first session to reestablish the PDN connection is specifically: when the communication module 1102 receives the uplink signaling of the first session, the first session is deleted, and the user equipment corresponding to the first session is instructed to reestablish the PDN connection.

Referring to fig. 11, fig. 11 is a schematic structural diagram of a network device according to an embodiment of the present application. The network device shown in fig. 11 may be used to perform some or all of the functions of the first SGW-U in the method embodiment described above in fig. 9. The network device shown in fig. 11 may include a processing module 1101 and a communication module 1102. Wherein:

the processing module 1101 is configured to determine a second session affected by a first packet data network gateway user plane function entity PGW-U when determining that the PGW-U is abnormal; the communication module 1102 is configured to send a notification message to the first serving gateway control plane function entity SGW-C when receiving the uplink data packet of the second session, where the notification message is used to notify the second session of receiving the uplink data packet.

In one particular design, the processing module 1101 is further configured to record, in the session, an identification of a PGW-U used by the session; the processing module 1101 determines the manner of the second session affected by the first PGW-U specifically is: and determining that the session with the identification of the first PGW-U in the sessions is a second session influenced by the first PGW-U.

Referring to fig. 11, fig. 11 is a schematic structural diagram of a network device according to an embodiment of the present application. The network device shown in fig. 11 may be used to perform some or all of the functions of the first SGW-C in the method embodiment described above in fig. 9. The network device shown in fig. 11 may include a processing module 1101 and a communication module 1102. Wherein:

a communication module 1102, configured to receive a notification message from a first serving gateway user plane function entity SGW-U, where the notification message is used to notify a second session to receive an uplink data packet, where the second session is a session affected by an abnormal first packet data network gateway user plane function entity PGW-U; a processing module 1101, configured to delete the second session, and instruct the ue corresponding to the second session to reestablish the PDN connection.

The embodiment of the application provides network equipment. The network device may be configured to perform some or all of the functionality of the first PGW-C in the method embodiment described above with reference to fig. 10. The network device may include a processing module. Wherein:

the processing module is used for determining a third session influenced by the first packet data network gateway user plane function entity (PGW-U) when detecting that the first PGW-U is abnormal; and the processing module is further configured to delete the third session and instruct the user equipment corresponding to the third session to reestablish the PDN connection when the downlink signaling of the third session is received.

In a specific design, the processing module is further configured to record, in the session, an identification of a PGW-U used by the session; the method for determining, by the processing module, the third session affected by the first PGW-U specifically includes: and determining the session with the identification of the first PGW-U in the sessions as a third session influenced by the first PGW-U.

In one particular design, the network device further includes a communication module. When the communication module receives a signaling of a fourth session, sending a message to a second PGW-U used by the fourth session according to the signaling of the fourth session; after the communication module retransmits the message to the second PGW-U, if the first network device does not receive the response message of the message within the preset time period, the processing module deletes the fourth session, and instructs the user device corresponding to the fourth session to reestablish the PDN connection.

Referring to fig. 12, fig. 12 is a schematic structural diagram of a network device disclosed in the embodiment of the present application. As shown in fig. 12, the network device includes a processor 1201, a memory 1202, and a communication interface 1203. Wherein the processor 1201, the memory 1202 and the communication interface 1203 are connected.

The processor 1201 may be a Central Processing Unit (CPU), a general purpose processor, a coprocessor, a Digital Signal Processor (DSP), an application-specific integrated circuit (ASIC), a Field Programmable Gate Array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. The processor 1201 may also be a combination of computing functions, e.g., comprising one or more microprocessors, a combination of a DSP and a microprocessor, or the like.

The communication interface 1203 is used for implementing communication with an access network device.

Wherein, the processor 1201 calls the program code stored in the memory 1202 to execute the steps executed by the MME, the SGW-C, SGW-U, PGW-C, or the SGW-U in the above method embodiments.

Based on the same inventive concept, the principle of the network device to solve the problem provided in the embodiment of the present application is similar to the principle of the MME, the SGW-C, SGW-U, PGW-C, or the SGW-U to solve the problem in the embodiment of the method of the present application, so that the implementation of each device may refer to the implementation of the method, and is not described herein again for brevity.

In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

Finally, it should be noted that: the above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A gateway selection system, comprising a mobility management entity MME, a plurality of serving gateway control plane function entities SGW-C, and a plurality of serving gateway user plane function entities SGW-U, wherein the MME interfaces with the plurality of SGW-C, the MME communicates with SGW-C via an S11 interface, one SGW-C interfaces with one or more SGW-U, one SGW-U interfaces with one or more SGW-C, and the SGW-C communicates with SGW-U via a Sxa interface, wherein:

the first SGW-C is used for sending an identifier of the first SGW-U to the MME when detecting that the first SGW-U is abnormal, wherein the first SGW-C is one of the SGW-Cs, and the first SGW-U is the SGW-U which is butted with the first SGW-C in the SGW-Us;

the MME is used for recording the identification of the SGW-U used by the session in the session;

the MME further configured to receive an identification of the first SGW-U from the first SGW-C;

the MME is further configured to determine that the session with the identifier of the first SGW-U is a first session affected by the first SGW-U, and reselect a second SGW-C which does not fail for the first session when the MME receives uplink signaling of the first session;

the second SGW-C is used for reselecting a second SGW-U which does not fail for the first session.

2. The gateway selection system of claim 1, further comprising a plurality of packet data network gateway control plane function entities, PGW-C, and a plurality of packet data network gateway user plane function entities, PGW-U, wherein:

a first PGW-C, configured to receive, from a first PGW-U or the first SGW-C, an identifier of the first SGW-U where an abnormality occurs, where the first PGW-C is a PGW-C of the multiple PGW-cs that is docked with the first SGW-C, and the first PGW-U is a PGW-U of the multiple PGW-us that is docked with the first SGW-U and the first PGW-C;

the first PGW-C is further configured to determine, according to an identifier of the first SGW-U, a second session affected by the first SGW-U, and send, when receiving downlink signaling of the second session, a first message to the MME, where the first message is used to trigger the MME to reselect an non-failed SGW-C for the second session;

the MME is further configured to receive the first message from the first PGW-C, and reselect a third SGW-C that has not failed for the second session according to the first message;

the third SGW-C is configured to reselect a non-failed third SGW-U for the second session.

3. The gateway selection system of claim 2,

the first SGW-C is further configured to send, to the first PGW-C, an identifier of an SGW-U used for a session;

the first PGW-C is further configured to receive, from the first SGW-C, an identifier of an SGW-U used by the session, and record, in the session, the identifier of an SGW-U corresponding to the session;

the determining, by the first PGW-C, a second session affected by the first SGW-U according to the identifier of the first SGW-U, including:

the first PGW-C determines that a session with the identification of the first SGW-U in the sessions is a second session affected by the first SGW-U.

4. The gateway selection system of claim 1, further comprising a plurality of packet data network gateway control plane function entities, PGW-C, and a plurality of packet data network gateway user plane function entities, PGW-U, wherein:

the first SGW-C is further used for determining a second session influenced by the first SGW-U when the first SGW-U is detected to be abnormal;

a first PGW-C, configured to receive a first signaling corresponding to the second session, and send a second signaling corresponding to the second session to the first SGW-C according to the first signaling, where the first PGW-C is a PGW-C that is docked with the first SGW-C among the multiple PGW-cs;

the first SGW-C is further configured to receive the second signaling from the first PGW-C, and send a response of the second signaling to the first PGW-C, where the response of the second signaling carries a cause value used for indicating that a context of the second session is not found;

the first PGW-C is further configured to receive a response to the second signaling from the first SGW-C, and send a first message to the MME, where the first message is used to trigger the MME to reselect an non-failed SGW-C for the second session;

5. The gateway selection system of claim 4,

the first SGW-C is also used for recording the identification of the SGW-U used by the session in the session;

the first SGW-C determining a second session affected by the first SGW-U, comprising:

the first SGW-C determines that a session with the identity of the first SGW-U among the sessions is a second session affected by the first SGW-U.

6. The gateway selection system according to any of claims 1 to 5, wherein the gateway selection system further comprises a plurality of packet data network gateway control plane function entities (PGW-C) and a plurality of packet data network gateway user plane function entities (PGW-U), wherein:

the first PGW-U is configured to determine, when it is detected that the first SGW-U is abnormal, a third session affected by the first SGW-U, where the first PGW-U is a PGW-U in the plurality of PGW-Us, and the PGW-U is docked with the first SGW-U;

the first PGW-U is further configured to send, when receiving the downlink data packet of the third session, a notification message to a first PGW-C, where the notification message is used to notify that the third session receives the downlink data packet, and the first PGW-C is a PGW-C in the multiple PGW-cs, where the PGW-C is docked with the first PGW-U;

the first PGW-C is further configured to receive the notification message from the first PGW-U, and send a second message to the MME, where the second message is used to trigger the MME to reselect an non-failed SGW-C for the third session;

the MME is further configured to receive the second message from the PGW-C, and reselect a fourth SGW-C that has not failed for the third session according to the second message;

the fourth SGW-C is configured to reselect a non-failed fourth SGW-U for the third session.

7. The gateway selection system of claim 6,

the first PGW-U is further used for recording the identification of the SGW-U used by the session in the session;

the first PGW-U determining a third session affected by the first SGW-U, including:

the first PGW-U determines that a session with the identification of the first SGW-U in the sessions is a third session affected by the first SGW-U.

8. A method for gateway selection, the method further comprising:

when detecting that a first serving gateway user plane functional entity SGW-U docked with a first SGW-C is abnormal, a first serving gateway control plane functional entity SGW-C sends an identifier of the first SGW-U to a mobile management entity MME;

the MME records the identification of the SGW-U used by the session in the session;

the MME receiving an identification of the first SGW-U from the first SGW-C;

the MME determines that the session with the identification of the first SGW-U is a first session influenced by the first SGW-U, and reselects a second SGW-C which does not fail for the first session when the MME receives uplink signaling of the first session;

the second SGW-C reselects a non-failed second SGW-U for the first session.

9. The method of claim 8, further comprising:

a first PGW-C receives an identification of a first SGW-U with an abnormality from a first PGW-U or the first SGW-C, wherein the first PGW-C is a PGW-C in butt joint with the first SGW-C, and the first PGW-U is a PGW-U in butt joint with the first SGW-U and the first PGW-C;

the first PGW-C determines a second session influenced by the first SGW-U according to the identifier of the first SGW-U;

when receiving a downlink signaling of the second session, the first PGW-C sends a first message to the MME, where the first message is used to trigger the MME to reselect an non-failed SGW-C for the second session;

the MME receives the first message from the first PGW-C, and reselects a third SGW-C which does not fail for the second session according to the first message;

the third SGW-C reselects a non-failed third SGW-U for the second session.

10. The method of claim 9, further comprising:

the first SGW-C sends an identification of an SGW-U used by a session to the first PGW-C;

the first PGW-C receives an identification of an SGW-U used by the session from the first SGW-C, and records the identification of the SGW-U corresponding to the session in the session;

11. The method of claim 8, further comprising:

the first SGW-C determines a second session influenced by the first SGW-U when detecting that the first SGW-U is abnormal;

a first PGW-C receives a first signaling corresponding to the second session, and sends a second signaling corresponding to the second session to the first SGW-C according to the first signaling, where the first PGW-C is a PGW-C docked with the first SGW-C;

the first SGW-C receives the second signaling from the first PGW-C, and sends a response of the second signaling to the first PGW-C, where the response of the second signaling carries a cause value used for indicating that a context of the second session is not found;

the first PGW-C receiving a response to the second signaling from the first SGW-C, and sending a first message to the MME, where the first message is used to trigger the MME to reselect an non-failed SGW-C for the second session;

the third SGW-C reselects a non-failed third SGW-U for the second session.

12. The method of claim 11, further comprising:

the first SGW-C records the identification of the SGW-U used by the session in the session;

13. The method according to any one of claims 8 to 12, further comprising:

when detecting that the first SGW-U is abnormal, a first PGW-U determines a third session affected by the first SGW-U, wherein the first PGW-U is a PGW-U which is in butt joint with the first SGW-U;

when receiving the downlink data packet of the third session, the first PGW-U sends a notification message to a first PGW-C, where the notification message is used to notify that the third session receives the downlink data packet, and the first PGW-C is a PGW-C docked with the first PGW-U;

the first PGW-C receiving the notification message from the first PGW-U, and sending a second message to the MME, where the second message is used to trigger the MME to reselect an non-failed SGW-C for the third session;

the MME receives the second message from the PGW-C, and reselects a fourth SGW-C which does not fail for the third session according to the second message;

the fourth SGW-C reselects a non-failed fourth SGW-U for the third session.

14. The method of claim 13, further comprising:

the first PGW-U records the identification of the SGW-U used by the session in the session;