WO2024078696A1

WO2024078696A1 - User plane entity enhancement for supporting in-network computing in a mobile network

Info

Publication number: WO2024078696A1
Application number: PCT/EP2022/078169
Authority: WO
Inventors: Qing Wei
Original assignee: Huawei Technologies Co., Ltd.
Priority date: 2022-10-11
Filing date: 2022-10-11
Publication date: 2024-04-18

Abstract

This disclosure relates to a user plane entity and a control plane entity. The user plane entity is configured for performing in-network computing (INC) in a mobile network. The control plane entity supports the INC of the user plane entity. The user plane entity receives a network traffic flow, which comprises a service data flow from an application. It determines whether the service data flow is intended for application layer processing at the user plane entity and/or whether a condition for the application layer processing is fulfilled. If fulfilled, it sends the service data flow from a network layer to an application layer logic, and performs first processing operations at the application layer logic on the received service data flow. The control pane entity determines the configuration information of the user plane entity for the first processing operations, and provides configuration information to the user plane entity.

Description

USER PLANE ENTITY ENHANCEMENT FOR SUPPORTING IN-NETWORK COMPUTING IN A MOBILE NETWORK

TECHNICAL FIELD

The present disclosure relates to mobile communication networks, for instance, to New Radio (NR) or 5^th generation (5G) mobile networks. The disclosure proposes a user plane entity, a control plane entity, and corresponding methods. The user plane entity is configured in view of in-network computing (INC) to be performed in the mobile communication network. The control plane entity is configured to support the INC functions of the user plane entity.

BACKGROUND

Emerging applications, such as XR (Virtual Reality (VR), Argument Reality (AR), Mixed Reality (MR)), or Metaverse, or Machine Learning (ML) enabled applications, pose new requirements to mobile communication networks. For instance, VR glasses need to perform intensive computation tasks for video rendering, while at the same time they have to be designed with a light weight, with a constrained size, and have limited computing and battery capacity.

INC is a solution that offloads application layer computation tasks from light end devices to network entities. On the other hand, XR, Metaverse and ML enabled applications introduce much more network traffic than before, considering high resolution videos, and continue data collection from a big amount of end devices in parallel. Also for this INC is a solution, as it may improve the efficiency of resource usage, e.g., via traffic aggregation, compression, content caching, multicasting, and raw data pre-processing such as feature subtraction.

In a 3GPP network, service traffic from a User Equipment (UE) is transferred via the Radio Access Network (RAN) and User Plane Functions (UPFs) to the application server, which is deployed at a Mobile Edge Platform (MEP) or a Data Network (DN) for application layer processing.

It may be envisioned to integrate INC into next generation mobile networks (e.g., 5G+, 6G). According to the 3GPP’s definition, computational tasks at the application layer (in contrast to the network layer) are considered as application services provided in the DN or the MEP, i.e., outside the 3 GPP network domain. In other words, the 3 GPP network provides only a transport service for an application traffic between the UE and the DN or MEP at the N6 interface. The UPF(s) perform(s) only network layer treatment for the packets in a traffic flow, i.e., the packets are forwarded to a certain port of the UPF connecting to other network entities, are prioritized for network transport or dropped in case of network resource shortage, or are duplicated for better transport reliability.

SUMMARY

This disclosure and its solutions are based on the idea that a possible deployment option for INC could also be to integrate a light-weight MEP at user plane entities, e.g., UPFs, wherein such a UPF may be called a computing UPF (cUPF). An application could deploy application layer computing logic at the user plane entity on a forwarding path of a UE. With this intermediate computing logic, one or more computing tasks could start much earlier, i.e., before arriving at the endpoint where the actual application server is located. This would reduce the traffic volume in the transport network and aggregated at a datacenter. Further, service latency for applications could be reduced. In addition, resource utilization of the network could be improved, and the application layer computation load at light end devices or the application server could be released.

Based on this idea, the disclosure focuses particularly on the technical problem of how to support traffic that would be transferred to an application layer computation logic, which is deployed at the user plane entity for processing, and how the network layer transport would continue after the processing by the application layer computation logic.

In current 3 GPP, incoming application traffic is classified into different traffic flows at the UPF based on a Packet Detection Rule (PDR). The classified traffic flows are treated according to a set of instructions that binds to that PDR (e.g., Forwarding Action Ruler (FAR), Quality of Service (QoS) enforcement Rule (QER), Usage Reporting Rule (URR)) at the UPF.

A FAR specifies packet treatments such as drop, forwarding, buffer, duplicate, eliminate duplicated packets. A QER specifies the rules on QoS enforcement. A URR specifies the rules on related notification and/or reporting to a control plane entity, e.g., Control Plane Network Function (CP -NF). The set of instructions may be provided by the control plane entity, which may be the Session Management Function (SMF) in the 5G system. In the current 3GPP specification, all the packet treatments are made at the network layer only. That is, the FAR does not support a packet transfer to a local application layer at a user plane entity. Further, a PDR based on a current flow ID will not work anymore, if the end-to-end (e2e) traffic flow is broken due to the intermediate processing at one or more user plane entities (e.g., cUPFs). The application traffic (service data flow) after the intermediate processing at such a user plane entity may terminate at the user plane entity, or may become a new application traffic flow (with another set of application packets). The user plane entities of 3GPP are not able to identify such type of processed application traffic.

In view of the above, this disclosure aims to provide a solution that supports the transport of packets, which require to be processed by application logic located at a user plane entity for application layer processing (e.g., a cUPF). An objective is particularly to provide a solution for detecting, at the user plane entity, traffic requiring local application layer processing, and for further transferring that traffic to the application layer for processing. Another objective is particularly to provide a solution for associating the traffic before and after the local application layer processing at the user plane entity, and for continuing the network layer transport after the processing.

These and other objectives are achieved by this disclosure as described in the independent claims. Advantageous implementations are further defined in the dependent claims.

Notably, in this disclosure, a user plane entity with application layer processing capability may be referred to as cUPF for simplicity. A cUPF may contain computing logical that is defined and deployed according to the requirement s) of an application.

A first aspect of this disclosure provides a user plane entity for a network capable of application layer processing of a service data flow (i.e., application traffic) transported through the network, the user plane entity being configured to: receive, from a second user plane entity of the network, a network traffic flow comprising a service data flow from an application; determine whether the received service data flow is intended for application layer processing at the user plane entity; and/or determine whether a condition for performing the application layer processing is fulfilled; and if the received service data flow is intended for application layer processing and/or the condition for performing the application layer processing is fulfilled: send the received service data flow from a network layer to an application layer logic at the application layer of the user plane entity; and perform one or more first processing operations at the application layer logic on the received service data flow to obtain an application-layer processed service data flow.

The user plane entity of the first aspect may be a cUPF as explained above. The network traffic flow may carry the service data flow to the user plane entity, wherein after performing the application layer processing, the network traffic may be terminated at the user plane entity or the user plane entity may forward the processed network traffic. The user plane entity of the first aspect provides a solution for detecting network traffic containing a service data flow that requires local application layer processing. Further, the user plane entity of the first aspect provides a solution for transferring that (detected) service data flow to the application layer for processing, particularly, by the application layer logic deployed in the application layer of the user plane entity. The user plane entity of the first aspect also provides a solution for associating the service data flow of the network traffic before and after the local application layer processing at the user plane entity, and a solution for continuing the network layer transport after the processing of the service data flow at the application layer logic.

In an implementation form of the first aspect, the user plane entity is further configured to send the application-layer processed service data flow back to the network layer; and forward the network traffic comprising the application-layer processed service data flow.

For instance, the user plane entity may forward the network traffic including the processed service data flow to a third user plane entity (which may perform further application layer processing on the processed service data flow) or to a data network entity. Notably, also the service data flow received at the user plane entity of the first aspect may already have been application layer processed by the second user plane entity.

In an implementation form of the first aspect, the user plane entity is further configured to determine whether the service data flow is intended for application layer processing based on a first packet detection rule (PDR) associated with the received service data flow.

This provides a simple implementation for detecting the relevant service data flows. In an implementation form of the first aspect, the condition for performing the application layer processing includes one or more of: a processing load of an application logic at the application layer of the user plane entity is lower than a threshold; a queuing delay at the application layer and/or application logic of the user plane entity is lower than a threshold; an availability status of the application logic at the application layer of the user plane entity indicates that the application logic is available; a weight factor and/or selection probability associated to the application logic at the user plane entity is higher than a threshold.

Notably, whether a queuing delay at the application layer or the application logic is considered may depend on the implementation. For instance, whether there is a shared queue for multiple application logics or whether there is an individual queue per application logic. Overall, the conditions allow a flexible user plane entity selection on a UE path. For instance, load balancing is enabled.

In an implementation form of the first aspect, the user plane entity is further configured to associate the application-layer processed service data flow with a second PDR.

In this way, traffic association before and after the application layer processing at the user plane entity is implemented. This enables e2e transport service assurance.

In an implementation form of the first aspect, the user plane entity is further configured to perform one or more second processing operations at the network layer on the network traffic flow comprising the application-layer processed service data flow before forwarding it to the third user plane entity or to the data network entity.

In an implementation form of the first aspect, the user plane entity is further configured to, if the received service data flow is not intended for application layer processing at the user plane entity and/or the condition for performing the application layer processing is not fulfilled: perform one or more third processing operations at the network layer on the received network traffic flow comprising the service data flow; and forward the processed network traffic flow.

For example, the network traffic flow may be forwarded to a third user plane entity or to a data network entity. In this case, a conventional processing at the user plane entity may be performed. In an implementation form of the first aspect, the received network traffic flow comprising the service data flow is associated with a first FAR, and the user plane entity is configured to associate the application-layer processed service data flow with a second FAR.

In an implementation form of the first aspect, the one or more second processing operations at the network layer are performed according to the second FAR.

In an implementation form of the first aspect, the one or more third computation operations at the network layer are performed according to the first FAR.

In an implementation form of the first aspect, the user plane entity is further configured to generate the second PDR based on a PDR generating configuration included in configuration information received from a control plane entity.

This may reduce the complexity at the application layer, as well the message overhead between control plane entities, and between a control plane entity and an application.

In an implementation form of the first aspect, the user plane entity is further configured to derive the second PDR from configuration information received from the control plane entity.

In an implementation form of the first aspect, the configuration information comprises one or more of a flow description of the service data flow; an end-to-end flow description of the service data flow; a local source address; a local flow ID generated by the user plane entity.

In an implementation form of the first aspect, the user plane entity is further configured to derive the second FAR from the configuration information received from the control plane entity.

A second aspect of this disclosure provides a control plane entity for controlling an application layer processing of a service data flow transported through a user plane of a network, the control plane entity being configured to: determine for each user plane entity of one or more selected user plane entities, configuration information indicating a first PDR for the user plane entity to detect a service data flow requiring application layer processing at the user plane entity and a second PDR to be associated by the user plane entity with the application-layer processed service data flow; and provide the determined configuration information to each user plane entity of one or more selected user plane entities.

In addition, the control plane entity may be configured to determine one or more conditions regarding the first PDR for the user plane entity. This enables control of a further selection of user plane entities among the one or more selected user plane entities without the involvement of the control plane entity. That is, the one or more conditions may be used to control the selection of user plane entities for application layer processing. The control plane entity of the second aspect supports the above-mentioned solutions provided by the user plane entity of the first aspect and its respective implementation forms. The control plane entity of the second aspects and its following implementation forms can configure the selected user plane entities to work according to these solutions.

In an implementation form of the second aspect, the configuration information provided to each user plane entity further indicates a FAR to be associated by the user plane entity to the application-layer processed service data flow.

In an implementation form of the second aspect, the control plane entity is further configured to derive the first PDR for each selected user plane entity based on a condition for performing the application layer processing received from an application; wherein the condition for performing the application layer processing includes one or more of a processing load of an application logic at the application layer of the user plane entity is lower than a threshold; a queuing delay at the application layer and/or application logic of the user plane entity is lower than a threshold; an availability status of the application logic at the application layer of the user plane entity indicates that the application logic is available; a weight factor and/or selection probability associated to the application logic at the user plane entity is higher than a threshold.

The conditions may be referred to as INC conditions, as they may determine whether application layer processing is to be performed in the network or not.

In an implementation form of the second aspect, the control plane entity is further configured to derive the second PDR for each selected user plane entity from a list of service description of the service data flow received from an application. In an implementation form of the second aspect, the control plane entity is further configured to generate the second PDR or a PDR generation configuration for each selected user plane entity based on a service description of the service data flow and a PDR generating instruction received from an application or another control plane entity.

The other control plane entity may be PCF. In this case, the control plane entity may comprise a SMF.

In an implementation form of the second aspect, the PDR generation instruction or the PDR generation configuration comprises one or more of an end-to-end flow description of the service data flow; an instruction to use a new source and/or destination port; an instruction to use a local source IP address; an instruction to use a local flow ID generated by the user plane entity.

In an implementation form of the second aspect, the configuration information provided to each selected user plane entity includes a PDR generation configuration for configuring the user plane entity to generate the second PDR that is to be associated by the user plane entity to the application-layer processed service data flow.

The PDR generation configuration may comprise an instruction for the user plane entity to generate a local flow ID (e.g., mapped to a PDU session ID and global flow ID) for the service data flow requiring application layer processing, provide the local flow ID to the application layer, and generate the second PDR using the local flow ID.

A third aspect of this disclosure provides a method for a user plane entity of a network capable of application layer processing of a service data flow transported through the network, method comprising: receiving, from a second user plane entity of the network, a network traffic flow comprising a service data flow from an application; determining whether the received service data flow is intended for application layer processing at the user plane entity; and/or determining whether a condition for performing the application layer processing is fulfilled; and if the received service data flow is intended for application layer processing and/or the condition for performing the application layer processing is fulfilled: sending the received service data flow from a network layer to an application layer logic at the application layer of the user plane entity; performing one or more first processing operations at the application layer logic on the received service data flow.

In an implementation form of the third aspect, the method further comprises sending the application-layer processed service data flow back to the network layer; and forwarding the network traffic comprising the application-layer processed service data flow.

In an implementation form of the third aspect, the method further comprises determining whether the service data flow is intended for application layer processing based on a first PDR associated with the received service data flow.

In an implementation form of the third aspect, the condition for performing the application layer processing includes one or more of a processing load of an application logic at the application layer of the user plane entity is lower than a threshold; a queuing delay at the application layer and/or application logic of the user plane entity is lower than a threshold; an availability status of the application logic at the application layer of the user plane entity indicates that the application logic is available; a weight factor and/or selection probability associated to the application logic at the user plane entity is higher than a threshold.

In an implementation form of the third aspect, the method further comprises associating the application-layer processed service data flow with a second PDR.

In an implementation form of the third aspect, the method further comprises performing one or more second processing operations at the network layer on the network traffic flow comprising the application-layer processed service data flow before forwarding it to the third user plane entity or to the data network entity.

In an implementation form of the third aspect, the method further comprises, if the received service data flow is not intended for application layer processing at the user plane entity and/or the condition for performing the application layer processing is not fulfilled: performing one or more third processing operations at the network layer on the received network traffic flow comprising the service data flow; and forwarding the processed network traffic flow. In an implementation form of the third aspect, the received network traffic flow comprising the service data flow is associated with a first forwarding action rule, FAR, and the method further comprises associating the application-layer processed service data flow with a second FAR.

In an implementation form of the third aspect, the one or more second processing operations at the network layer are performed according to the second FAR.

In an implementation form of the third aspect, the one or more third processing operations at the network layer are performed according to the first FAR.

In an implementation form of the third aspect, the method further comprises generating the second PDR based on a PDR generating configuration included in configuration information received from a control plane entity.

In an implementation form of the third aspect, the method further comprises deriving the second PDR from configuration information received from the control plane entity.

In an implementation form of the third aspect, the configuration information comprises one or more of a flow description of the service data flow; an end-to-end flow description of the service data flow; a local source address; a local flow ID generated by the user plane.

In an implementation form of the first aspect, the method further comprises deriving the second FAR from the configuration information received from the control plane entity.

The method of the third aspect and its implementation forms achieve the same advantages described above for the user plane entity of the first aspect and its respective implementations forms.

A fourth aspect of this disclosure provides a method for a control plane entity for controlling an application layer processing of a service data flow transported through a user plane of a network, the method comprising: determining, for each user plane entity of one or more selected user plane entities, configuration information indicating a first PDR for the user plane entity to detect a service data flow requiring application layer processing at the user plane entity and a second PDR to be associated by the user plane entity with an application-layer processed service data flow; and providing the determined configuration information to each user plane entity of the one or more selected user plane entities.

In addition, the method may comprise determining, for each user plane entity of the one or more selected user plane entities, one or more conditions regarding the first PDR for the user plane entity.

In an implementation form of the fourth aspect, the configuration information provided to each user plane entity further indicates a FAR to be associated by the user plane entity to the application-layer processed service data flow.

In an implementation form of the fourth aspect, the method further comprises deriving the first PDR for each selected user plane entity based on a condition for performing the application layer processing received from an application; wherein the condition for performing the application layer processing includes one or more of a processing load of an application logic at the application layer of the user plane entity is lower than a threshold; a queuing delay at the application layer and/or application logic of the user plane entity is lower than a threshold; an availability status of the application logic at the application layer of the user plane entity indicates that the application logic is available; a weight factor and/or selection probability associated to the application logic at the user plane entity is higher than a threshold.

In an implementation form of the fourth aspect, the method further comprises deriving the second PDR for each selected user plane entity from a list of service description of the service data flow received from an application.

In an implementation form of the fourth aspect, the method further comprises generating the second PDR for each selected user plane entity based on a service description of the service data flow and a PDR generating instruction received from an application or another control plane entity.

In an implementation form of the fourth aspect, the PDR generation instruction or PDR generation configuration comprises one or more of an end-to-end flow description of the service data flow; an instruction to use a new source and/or destination port; an instruction to use a local source IP address; an instruction to use a local flow ID generated by the user plane entity.

In an implementation form of the fourth aspect, the configuration information provided to each selected user plane entity includes a PDR generation configuration for configuring the user plane entity to generate the second PDR that is to be associated by the user plane entity to the application-layer processed service data flow.

The method of the fourth aspect and its implementation forms achieve the same advantages described above for the control plane entity of the second aspect and its respective implementations forms.

A fifths aspect of this disclosure provides a computer program comprising instructions which, when the program is executed by a computer, cause the computer to perform the method according to the third aspect or fourth aspect or any implementation forms thereof.

A sixths aspect of this disclosure provides a non-transitory storage medium storing executable program code which, when executed by a processor, causes the method according to the third aspect or fourth aspect or any of its implementation forms to be performed.

It has to be noted that all devices, elements, units and means described in the present application could be implemented in the software or hardware elements or any kind of combination thereof. All steps which are performed by the various entities described in the present application as well as the functionalities described to be performed by the various entities are intended to mean that the respective entity is adapted to or configured to perform the respective steps and functionalities. Even if, in the following description of specific embodiments, a specific functionality or step to be performed by external entities is not reflected in the description of a specific detailed element of that entity which performs that specific step or functionality, it should be clear for a skilled person that these methods and functionalities can be implemented in respective software or hardware elements, or any kind of combination thereof.

BRIEF DESCRIPTION OF DRAWINGS

The above described aspects and implementation forms will be explained in the following description of specific embodiments in relation to the enclosed drawings, in which FIG. 1 shows control of a user plane entity (UPF) in a 5G mobile network.

FIG. 2 shows a user plane entity (cUPF) and a control plane entity (control NF) according to this disclosure.

FIG. 3 shows a user plane entity (cUPF) and a control plane entity (control NF) according to this disclosure.

FIG. 4 shows an exemplary deployment of a user plane entity (cUPF) of this disclosure in a 3 GPP network architecture.

FIG. 5 shows an exemplary traffic transfer between an application layer and network layer at the user plane entity (cUPF) of this disclosure.

FIG. 6 shows an exemplary implementation at a user plane entity (cUPF) of this disclosure with a second FAR and a second PDR associated with the processed service data flow, and a decision making at the user plane entity.

FIG. 7 shows exemplary scenarios with multiple user plane entities (cUPFs) according to this disclosure, which are deployed in a network for the processing with application layer logic.

FIG. 8 shows a first option for obtaining the second PDR, wherein the second PDR is derived based on a flow description.

FIG. 9 shows an exemplary message flow for the first option, wherein an AF provides the flow description and QoS requirements to a 5GS.

FIG. 10 shows an exemplary message flow for the first option, wherein the 5GS determines the configurations and configures the user plane function (cUPF) of this disclosure.

FIG. 11 shows a second option for obtaining the second PDR, wherein the second PDR is automatically generated at a control plane entity (SMF) of this disclosure.

FIG. 12 shows a message flow for the second option, wherein the 5GS determines the configurations and configures a user plane entity (cUPF) of this disclosure.

FIG. 13 shows a third option for obtaining the second PDR, wherein the second PDR is automatically generated at a user plane entity (cUPF) of this disclosure.

FIG. 14 shows a N4 procedure for a configuration of a user plane entity (cUPF) according to this disclosure.

FIG. 15 shows a method for a user plane entity according to this disclosure.

FIG. 16 shows a method for a control plane entity according to this disclosure. DETAILED DESCRIPTION OF EMBODIMENTS

This disclosure describes a set of enhancements for a user plane entity 210, and a control plane entity 220 that supports these enhancements. A user plane entity 210 and a control plane entity 220 according to this disclosure are shown in FIG. 2. The entities 210, 220 enables a local processing of a service data flow 211 of an application at the application layer 213 of the user plane entity 210.

According to this disclosure, generally, the user plane entity 210 is configured to perform a conditional and local traffic transfer support between the network layer 212 of the user plane entity 210 and the application layer 213 of the user plane entity.

The user plane entity 210 of this disclosure is configured to receive a network traffic flow comprising the service data flow 211. For instance, the network traffic flow may be received from a second user plane entity (with or without application-layer processing capability) or from a UE. The user plane entity 210 is then configured to determine whether the received service data flow 211 is intended for application layer processing at the user plane entity 210 and/or to determine whether a condition for performing the application layer processing is fulfilled.

If the received service data flow 211 is intended for application layer processing and/or if the condition for performing the application layer processing is fulfilled, the user plane entity 210 is configured to send the received service data flow 211 from the network layer 212 to an application layer logic 213a deployed at its application layer 213. The application layer logic 213a is configured to perform one or more first processing operations on the received service data flow 211, in order to obtain an application-layer processed service data flow 214. That is, the result of the one or more first processing operations is the application-layer processed service data flow 214.

As further shown in FIG. 2, the user plane entity 210 may be also configured to send the application-layer processed service data flow 214 back to the network layer 212, and then to forward network traffic comprising the application-layer processed service data flow 214 (at the network layer). Notably, the service data flow 211 being forwarded from the network layer 212 to the application layer 213 may mean that the network traffic flow comprising the service data flow 211 is forwarded, and likewise the application-layer processed service data flow 214 being forwarded from the application layer 213 to the network layer 212 may mean that the network traffic flow comprising the application-layer processed service data flow 214.

The control plane entity 220 may be configured to control the user plane entity 210 and/or to configure the user plane entity 210. In particular, the control plane entity 220 is configured to control the selection of the user plane entity 210 to perform the one or more first processing operations at the application layer 213 on the service data flow 214.

The condition for performing the application layer processing, which may be evaluated at the user plane entity 210, may comprise one or more criteria for performing the local processing operations, which criteria may be defined by the application. For instance, the criteria may comprise at least one of a processing load of the application logic 213a deployed at the application layer 213 of the user plane entity 210, and a weight factor defined by the application, and a processing status of the service data flow 211, 214.

As shown in FIG. 3, the service data flow 211 may be associated with a first PDR 301 (“PDR” in FIG. 3). This first PDR 301 may be used by the user plane entity 210 to determine, whether the service data flow 211 is intended for application layer processing (or not). Further, the user plane entity 210 may be further configured to associate the application-layer processed service data flow 214 with a second PDR 302 (“associated PDR” in FIG. 3). Optionally, the network traffic flow comprising the service data flow 211 is associated with a first FAR 303 (“FAR i” in FIG. 3), and the user plane entity 210 may be further configured to associate the network traffic comprising the application-layer processed service data flow 214 with a second FAR 304 (“FAR e” in FIG. 3). That is, the user plane entity 210 may perform network traffic association for the service data flow 211, 214 before and after the local application layer processing, by using the second PDR 302, and optionally the second FAR 304.

The control plane entity 220 (e.g., a SMF) may have functionalities to derive the associated PDR 302 for the application-layer processed service data flow 214 at the user plane entity 210 and/or to indicate to the user plane entity 210 the second PDR 302 that it should use. In particular, the control plane entity 220 is configured to provide, to the user plane entity 210 (and in fact to each user plane entity 210 which it selects for application layer processing of the service data flow 211), an indication of the first PDR 301 and the second PDR 302. The first PDR 310 is for the user plane entity 210 to detect that the service data flow 211 requires application layer processing at the user plane entity 210. The second PDR 302 is to be associated by the user plane entity 210 with the application-layer processed service data flow 214.

Accordingly, control rules between the control plane entity 220 and the user plane entity 210 can be enhanced. These enhancements may be related to the first FAR 303, which may include the forwarding destination/ source address “local DNN”, and may be related to the second FAR 304, which may include new criteria/parameters used to determine the action of traffic transfer to the local application layer 213 at the user plane entity 210. The enhancements may also be related to the PDRs, specifically to the second PDR 302 that may be used for identifying the application-layer processed service data flow 214 by the local application layer 213 at the user plane entity 210. The enhancements may also relate to the interactions between different control plane entities 220, and between the control plane entity 220 and the application, for example, to obtain information related to the second PDR 302 and information related to the first FAR 303 and second FAR 304 (e.g., criteria for local user plane entity processing, associated traffic descriptions, associated PDR generation instructions), as will be described further below. FIG. 3 shows an exemplary implementation including the first FAR 303 and the second FAR 304, and including the association with the second PDR 302 at the user plane entity 210.

The user plane entity 210 and/or the control plane entity 220 may each comprise a processor or processing circuitry (not shown) configured to perform, conduct or initiate the various operations of the user plane entity 210 and/or the control plane entity 220 described herein. For instance, the processing circuitry at the user plane entity 210 may comprise the application layer logic 213a. The processing circuitry may comprise hardware and/or the processing circuitry may be controlled by software. The hardware may comprise analog circuitry or digital circuitry, or both analog and digital circuitry. The digital circuitry may comprise components such as application-specific integrated circuits (ASICs), field-programmable arrays (FPGAs), digital signal processors (DSPs), or multi-purpose processors. The user plane entity 210 and/or the control plane entity 220 may further each comprise memory circuitry, which stores one or more instruction(s) that can be executed by the processor or by the processing circuitry, in particular under control of the software. For instance, the memory circuitry may comprise a non-transitory storage medium storing executable software code which, when executed by the processor or the processing circuitry, causes the various operations of the user plane entity 210 and/or the control plane entity 220 to be performed. In one embodiment, the processing circuitry comprises one or more processors and a non-transitory memory connected to the one or more processors. The non-transitory memory may carry executable program code which, when executed by the one or more processors, causes the user plane entity 210 and/or the control plane entity 220 to perform, conduct or initiate the operations or methods described herein.

FIG. 4 shows an exemplary 3 GPP network architecture with enhanced user plane entities (UPFs) that are provided with additional application layer logic 213 (i.e., shows user plane entities 210, here referred to as cUPF, according to this disclosure). Any cUPF 210 may connect to the RAN via N3, may connect to DN or MEP via N6, and may connect to one or more other cUPFs 210 or to one or more conventional UPFs via N9. Any UPF or cUPF may be controlled by the control plane entity 220, which here in this example comprises the SMF, viaN4 interface, and the PCF. The PCF may derive traffic descriptions of the network traffic flow comprising service data flow 211 intended for application layer processing at the cUPFs (also referred to as “INC traffic”) and may derive related service parameters based on the interaction with the application related to the service data flow 211 (e.g., via UDR, NEF, or AF). The PCF may further provide the derived information to the SMF. The control plane entity 220 may control a selection of the cUPF(s) 210, may determine the configuration s) of the selected cUPF(s) 210, and may configure these cUPF(s) accordingly (e.g., with the first FAR 303 and the second FAR 304 and with the second PDR 302).

FIG. 5 shows an exemplary network traffic transfer between the application layer 213 and the network layer 212 at the user plane entity 210 (cUPF) of this disclosure. For packets that are transferred from the network layer 212 to the application layer 213, the following treatment may apply:

1. The user plane entity 210 determines whether the network traffic flow comprising the service data flow 211 should be transferred to the local application layer 213 based on application related local traffic transfer conditions. Details of the local traffic transfer conditions are explained below. The forwarding rule for such a network traffic flow may use a local address (e.g., IP with 0.0.0.0) and/or an application specific port number as the destination/target of the forwarding.

2. If the result is yes, the user plane entity may de-encapsulate the packets in the service data flow 211 (e.g., remove the network header) and forward them to the application layer 213 with the local address and/or application specific port. The user plane entity 210 may store the contents of the removed related network header (e.g., PDU session ID, flow ID) and may use that or part of the stored contents to generate the encapsulation header of the processed service data flow 214 afterwards. The stored contents may be removed after a certain time period (e.g., if there is no correspondent processed traffic generated by the local application layer). The step of storing the contents the network header can be skipped, if the user plane entity 210 knows that the network traffic flow will be terminated at it (e.g., via local configuration at the user plane entity 210 for this application).

For packets that are transferred from the application layer 213 to the network layer 212, the following treatments may be performed:

1. The user plane entity 210 applies a second PDR 302 to the processed service data flow 214. The second PDR 302, similar to the first PDR 301, may be derived from the traffic description of the (end-to-end) network traffic flow. It may be used to identify the correspondent network traffic flow after the application layer processing at the application layer logic 213a. Details on how to determine the second PDR 302 are explained below.

2. The user plane entity 210 re-encapsulates the packets with the stored network layer header and optionally additional sector indicator (e.g., associated sector ID or ID of the user plane entity 210). The original network layer header may include both the session information and flow information.

FIG. 6 shows an example for how these functionalities of the user plane entity 210 may work in a 5G system. As shown in FIG. 3, FAR may be split into the first FAR 303 (“FAR i”, refers to ingress FAR) and the second FAR 304 (“FAR e”, refers to egress FAR) at the user plane entity 210. The first PDR 301 (“PDR”) and the first FAR 303 may define the treatment of a network traffic flow comprising a service traffic flow 211, which needs to transfer to the application layer 213 for performing the one or more first processing operations. The second PDR 302 and the second FAR 304 may define the treatments of network traffic flow transfer back from the local application layer 213 to the network layer 212 after the one or more first processing operations have been performed. The first FAR 303 may use a local address as the destination of traffic transfer. The second FAR 304 may use the same destination as a conventional FAR (e.g., a network layer port number of the user plane entity 210). The first FAR 303 may include the conditions for the traffic to be forward to the application layer 213 for processing (i.e., INC conditions). FIG. 6 shows, in particular, a flow chart of an exemplary network traffic flow processing in the user plane entity 210. An incoming network traffic flow with the service data flow 211 matching the first PDR 301 may be proceeded by the user plane entity 210 with a network layer treatment defined by the first FAR 303. If the received service data flow 211 is not intended for application layer processing at the user plane entity 210 and/or if the condition for performing the application layer processing is not fulfilled, the network traffic flow is forward to the correspondent network layer target after a network layer treatment (e.g., a de-encapsulation) defined in the first FAR 301. In particular, the user plane entity 210 is configured to perform 607 one or more third processing operations at the network layer 212 on the received network traffic flow comprising the service data flow 211, and then forward 608 the processed network traffic flow.

If the received service data flow 211 is intended for application layer processing at the user plane entity 210 and/or if a condition for performing the application layer processing is fulfilled, the network traffic flow is forward to the correspondent application layer target (an application layer logic 213a) for processing. In particular, the user plane entity 210 is configured to send 603 the received service data flow 211 from the network layer 212 to the application layer logic 213a at the application layer 213 of the user plane entity 210, and then to perform 604 one or more first processing operations at the application layer logic 213a on the received service data flow 211, to obtain an application-layer processed service data flow 214.

When the service data flow 214 is sent back to the network layer 212 after the processing 604 (e.g., detected by the second PDR 302), the network traffic flow may be forward to a network layer target after a network layer treatment defined in the second FAR 303 (e.g., an encapsulation). The first FAR 303 and/or the second FAR 304 may include an additional network layer treatment (e.g., packet duplication, duplication elimination) except for the forwarding and encapsulation operation. The conditional check could be done before or after such additional network layer treatment described in the first FAR 303. In particular, the user plane entity 210 may be configured to perform 605 one or more second processing operations at the network layer 212 on the network traffic flow comprising the application-layer processed service data flow 214, and then forward 606 it to a third user plane entity or a DN entity, for instance. FIG. 7 shows that an application may deploy multiple INC instances at the user plane entities (UPFs) to serve one or more UEs. A network traffic flow from a certain UE may be processed by the application layer logic 213a at different user plane entities 210 in the UE data path. As shown for the “Data compression/Feature extraction” use case in FIG. 7, the UE traffic may be processed by the application layer logic 213a at the cUPF2 instead of cUPFl (both are user plane entities 210 according to this disclosure), since the application layer logic 213a of cUPFl may be overloaded. In the “data aggregation” use case of FIG. 7, the UE traffic may be either terminated at cUPF 1 or cUPF2 or at an application server (AS) deployed at the MEP or the DN, which may depend on the processing status in the application layer logic 213a (e.g., whether UE is able to retrieve the requested data on one of the user plane entities 210 on the way), or may depend on an application deployment configuration (e.g., whether the deployed application layer logic 213 at this user plane entity 210 is the termination point of the application.

According to the above, this disclosure proposes introducing conditional network traffic flow transfer to the application layer 213 based on additional INC conditions, which may be defined by the application related to the service data flow 211, wherein the INC conditions may include one or more of the following:

1. A weight factor and/or probability to select the (candidate) user plane entity 210 (e.g., for load balancing).

2. A computation load and/or queuing delay threshold at the application layer 213 or the application layer logic 213a at the user plane entity 210 (to satisfy latency requirements of the application traffic, i.e., the service data flow 211).

3. A status or states of the application layer logic 213a at the user plane entity 210, e.g., content available.

The SMF may provide such application related INC conditions in enhanced rules (e.g., the first FAR 303) to the user plane entity 210 using the procedure shown in FIG. 10 (from step 2). The SMF may obtain the INC conditions from an application via the PCF using the procedure shown in FIG. 9 and according to step 1 in FIG. 10. The SMF and the PCF are each control plane entities, and may form a control plane entity 220 of this disclosure, individually or together.

For an incoming network traffic flow comprising a service data flow 211 matching the first PDR 302, the user plane entity 210 may check its configuration or local monitoring results (e.g., the status or states indication from the application layer logic 213a, computation load or queuing delay) against the INC conditions in the first FAR 303, and may decide whether the network traffic flow should be forwarded to the local application layer 213 for processing.

There are multiple ways to derive second PDRs 302 at a user plane entity 210. In a first option, shown in FIG. 8, the second PDR 302 may be derived by an SMF 802 from a list of associated service flow description(s) from an application (via AF). The user plane entity 210 may be further configured to derive the second PDR 302 from configuration information 803 received from the control plane entity 220, here from the SMF 802. That is, the control plane entity 220 may provide to the user plane entity 210 the configuration information 803, which indicates the first PDR 301 and the second PDR 302. The configuration information 803 may further indicate the second FAR 304.

The configuration information 803 may also comprise one or more of the flow description of the service data flow 211, an end-to-end flow description of the service data flow 211, a local source address, and a local flow ID generated by the user plane entity 210. Notably, a network traffic flow, or specifically the service data flow 211 from the application, may be split into multiple sectors, and each sector may hay have its own flow description. A first flow description may describe, in this case, the network traffic flow when it enters the mobile network, and one or more second flow descriptions may follow after each application layer processing step (e.g., at the user plane entity 210).

An example procedure for the first option is shown in FIG. 9 and FIG. 10, and is as follows:

1. The application (AF) provides a list of associated flow descriptions and QoS requirements, in addition to the flow description, INC conditions, to the PCF 801 using the procedure shown in FIG. 9. Each associated flow description and QoS requirement may apply to one user plane entity 210 on the traffic path in the network. Alternatively, the PCF 801 may obtain the list of associated flow descriptions, INC conditions from other network entities (e.g., User Data Repository (UDR)) or via local configuration. Optionally, each associated flow description could also be a set of flow description(s) for a set of flows, wherein the set of flows require the same treatment in the network.

2. The PCF 801 provides derived PCC rules (including associated flow description(s), INC conditions) to the SMF 802 using a policy association shown in FIG. 10 step 1. 3. The SMF 802 derives the second PDR 302 based on the PCC rule from the PCF 801, as in FIG. 10 step 2. The SMF 802 derives an instruction set (e.g., including the first FAR 303, and the second FAR 304) as explained above.

4. The SMF 802 configures the user plane entity 210 with enhanced UPF rules (e.g., including the second PDR 302 and the instruction set) during the PDU session modification procedure as shown in FIG. 10 step 3.

5. In case of one or more incoming network traffic flow(s), the user plane entity 210 applies the enhanced UPF rules accordingly on these network traffic flow(s), as shown in FIG. 10, steps 4, 5 and 6.

The first option applies to the case in which the application knows the exact service data flow description(s) after each of the user plane entities 210 processing steps. However, the application may not know exactly which of the user plane entities 210 is taking care of the application layer processing of the network traffic flow comprising the service data flow 211 from a certain UE, considering dynamic load situation in the network and UE mobility. Introducing the list of associated flow descriptions may increase the complexity for the application slightly, and also the message overhead between the application and the control plane entity 220 and among the control plane entity functions (e.g., PCF 801 and SMF 802). The flow descriptions for each sector of a traffic flow in a mobile network can be very similar.

In a second option, shown in FIG. 11, the second PDR 302 may be generated by the SMF 802 based on the flow description and based on PDR generation instruction 1101. The second option overcomes any limitations of the first option. Instead of providing a list of associated flow descriptions, the AF/PCF 801 provides associated PDR generation instruction 1101 to the SMF 802. The SMF 802 can derive the second PDR 302 for each selected user plane entity 210 based on the PDR generation instruction 1101. The PDR generation instruction 1101 may be independent of the UE location and each individual user plane entity 210. It can be, e.g., generated from the flow description without change, or can be generated from the flow description(s) with the source IP address changed to local (e.g., aggregated traffic), or can be generated from the flow description with a new source and/or destination port, or can be generated from the flow description with an INC sequence number or INC indicator. An example procedure of the second option is as shown in FIG. 12:

1-2. The PDR generation instruction 1101 is configured at the SMF/PCF (control plane entity 220), or is obtained from the AF via the PCF 801 using the same procedure as shown for the first option, without the need for additional information on list of associated flow description(s).

3. The SMF 802 generates the second PDR(s) 302 based on the flow description using the PDR generation instruction 1101. The SMF 802 derives an instructions set (e.g., including the first FAR 303 and second FAR 304), as explained above.

4-5. These steps are the same as for the first option.

In a third option, shown in FIG. 13, the second PDR 302 is generated by the user plane entity 210 based on a PDR generation configuration 1301 received from the SMF 802. In case the application processing logic 213a deployed at the user plane entity 210 is able to mark the processed service data flow 214 using the local flow ID provided by the network layer 212, the second PDR 302 can be automatically generated by the user plane entity 210.

The SMF 802 may determine the PDR generation configuration 1301 per each user plane entity 210 based on the PDR generating instructions 1101 shown in the second option, and provides the PDR generation configuration 1301 to each user plane entity 210. The user plane entity 210 generates the second PDR 302 based on the PDR generation configuration 1301 received from the SMF 802.

The PDR generation configuration 1301 may configure the user plane entity to generate a local flow ID (e.g., mapped to a PDU session ID and global flow ID) for the service data flow requiring application layer processing, to provide the local flow ID to the application layer, and to generate the second PDR 302 using the local flow ID.

The PDR generation configuration 1301 can be used by the user plane entity 210, e.g., to generate a local flow ID (mapped to PDU session ID and flow ID) for traffic requiring local application layer processing and provide to the application layer, and to generate the second PDR 302 using the local flow ID.

In one implementation, the SMF 802 provides the configuration information 803 to the user plane entity 210 using a N4 procedure as shown in FIG. 14 The configuration information 803 may include enhanced user plane entity rules, such as the second PDR 302 or PDR generation configuration s) 1301, and may split the first FAR 303 and the second FAR 304. The first FAR 303 may comprise additional local application destination and conditional forwarding criteria (as listed above).

FIG. 15 shows a method 600 according to this disclosure. The method 600 is for a user plane entity 210 of a network that is capable of application layer processing of a service data flow 211, 214 transported through the network. The method 600 may be performed by any user plane entity 210, e.g., any cUPF, as described in this disclosure.

The method 600 comprises a step 601 of receiving, e.g. from a second user plane entity of the network, a network traffic flow comprising a service data flow 211 from an application. Further, the method 600 comprises a step 602 of determining whether the received service data flow 211 is intended for application layer processing at the user plane entity 210 and/or comprises a step 604 of determining whether a condition for performing the application layer processing is fulfilled.

If the received service data flow 211 is intended for application layer processing and/or if the condition for performing the application layer processing is fulfilled, the method 600 further comprises a step 604 of sending the received service data flow 211 from a network layer 212 to an application layer logic 213a at the application layer 213 of the user plane entity 210, and a step of performing 605 one or more first processing operations at the application layer logic 213a on the received service data flow 211.

FIG. 16 shows a method 1600 for a control plane entity 220 according to this disclosure. The method 1600 is for controlling an application layer processing of a service data flow 211, 214 transported through a user plane of a network. The method 1600 may be performed by any control plane entity 210, e.g., PCF 801 or SMF 802, described in this disclosure.

The method 1600 may comprise a step 1601 (optional step indicated by dashed box in FIG. 16) of determining, for each user plane entity of one or more selected user plane entities, one or more conditions regarding a first PDR 301 for the user plane entity. The one or more conditions may be used to control the selection of the user plane entities for the application layer processing. The method 1600 further comprises a step 1602 of determining, for each user plane entity 210 of the one or more selected user plane entities 210, configuration information 803. The configuration information 803 indicates the first PDR 301 for the user plane entity 210 to detect the service data flow 211 requiring application layer processing at the user plane entity 210, and a second PDR 302 to be associated by the user plane entity 210 with the application-layer processed service data flow 214.

According to this disclosure, a conditional local network traffic transfer to the application layer 213 at a user plane entity 210 is enabled. Flexible user plane entity 210 selection on a UE path based on, for example, processing load of the correspondent application at the user plane entity 210 or a configuration from the application is supported by the control plane entity 220.

Further traffic association before and/or after the application layer processing at the user plane entity 210 using the second PDR 302 is enabled. This also enables an e2e transport service assurance with INC performed by user plane entities 210 in the network. The second PDR 302 may be generated at the control plane entity like SMF 802 using the PDR generation instruction 1101 or may be generated at the user plane entity 210 using the PDR generation configuration 1301. This may reduce the complexity at the application layer 213, as well the message overhead between control plane entities like PCF 801 and SMF 802, and between the control plane entity 210 (e.g. including PCF 801 and/or SMF 802) and an application.

The solutions of this disclosure are backwards compatible to the current N3/N6 network functionalities at the user plane entities (UPFs).

The present disclosure has been described in conjunction with various embodiments as examples as well as implementations. However, other variations can be understood and effected by those persons skilled in the art and practicing the claimed matter, from the studies of the drawings, this disclosure and the independent claims. In the claims as well as in the description the word “comprising” does not exclude other elements or steps and the indefinite article “a” or “an” does not exclude a plurality. A single element or other unit may fulfill the functions of several entities or items recited in the claims. The mere fact that certain measures are recited in the mutual different dependent claims does not indicate that a combination of these measures cannot be used in an advantageous implementation.

Claims

1. A user plane entity (210) for a network capable of application layer processing of a service data flow transported through the network, the user plane entity (210) being configured to: receive, from a second user plane entity of the network, a network traffic flow comprising a service data flow (211) from an application; determine (601) whether the received service data flow (211) is intended for application layer processing at the user plane entity (210); and/or determine (602) whether a condition for performing the application layer processing is fulfilled; and if the received service data flow (211) is intended for application layer processing and/or the condition for performing the application layer processing is fulfilled:

- send (603) the received service data flow (211) from a network layer (212) to an application layer logic (213 a) at the application layer (213) of the user plane entity (210);

- perform (604) one or more first processing operations at the application layer logic (213a) on the received service data flow (211) to obtain an application-layer processed service data flow (214).

2. The user plane entity (210) according to claim 1, further configured to: send the application-layer processed service data flow (214) back to the network layer (212); and forward (606) the network traffic comprising the application-layer processed service data (214) flow.

3. The user plane entity (210) according to claim 1 or 2, configured to determine (102) whether the service data flow (211) is intended for application layer processing based on a first packet detection rule, PDR, (301) associated with the received service data flow (211).

4. The user plane entity (210) according to one of the claims 1 to 3, wherein the condition for performing the application layer processing includes one or more of:

- a processing load of the application logic (213a) at the application layer (213) of the user plane entity (210) is lower than a threshold;

- a queuing delay at the application layer (213) and/or application logic (213a) of the user plane entity (210) is lower than a threshold; - an availability status of the application logic (213a) at the application layer (213) of the user plane entity (210) indicates that the application logic (213a) is available;

- a weight factor and/or selection probability associated to the application logic (213a) at the user plane entity (210) is higher than a threshold.

5. The user plane entity (210) according to one of the claims 1 to 4, further configured to: associate the application-layer processed service data flow (214) with a second PDR

(302).

6. The user plane entity (210) according to one of the claims 1 to 5, further configured to: perform (605) one or more second processing operations at the network layer (212) on the network traffic flow comprising the application-layer processed service data flow (214) before forwarding (606) it.

7. The user plane entity (210) according to claim 6, further configured to, if the received service data flow (211) is not intended for application layer processing at the user plane entity (210) and/or the condition for performing the application layer processing is not fulfilled:

- perform (607) one or more third processing operations at the network layer (212) on the received network traffic flow comprising the service data flow (211); and

- forward (608) the processed network traffic flow.

8. The user plane entity (210) according to one of the claims 1 to 7, wherein the received network traffic flow comprising the service data flow (211) is associated with a first forwarding action rule, FAR, (303) and the user plane entity (210) is configured to associate the applicationlayer processed service data flow (214) with a second FAR (304).

9. The user plane entity (210) according to the claims 6 and 8, wherein the one or more second processing operations at the network layer (212) are performed according to the second FAR (304).

10. The user plane entity (210) according to one of the claims 5 to 9, further configured to generate the second PDR (302) based on a PDR generating configuration (1301) included in configuration information (803) received from a control plane entity (220).

11. The user plane entity (210) according to one of the claims 5 to 9, further configured to derive the second PDR (302) from configuration information (803) received from the control plane entity (220).

12. The user plane entity (210) according to claim 10 or 11, wherein the configuration information (803) comprises one or more of a flow description of the service data flow (211); an end-to-end flow description of the service data flow (211); a local source address; a local flow ID generated by the user plane entity (210).

13. The user plane entity (210) according to one of the claims 10 to 12, further configured to derive the second FAR (304) from the configuration information (803) received from the control plane entity (220).

14. A control plane entity (220) for controlling an application layer processing of a service data flow (211, 214) transported through a user plane of a network, the control plane entity (220) being configured to: determine, for each user plane entity (210) of one or more selected user plane entities (210), configuration information (803) indicating a first packet detection rule, PDR, (301) for the user plane entity (210) to detect a service data flow (211) requiring application layer processing at the user plane entity (210) and a second PDR (302) to be associated by the user plane entity (210) with an application-layer processed service data flow (214); and provide the determined configuration information to each user plane entity of the one or more selected user plane entities.

15. The control plane entity (220) according to claim 14, wherein the configuration information (803) provided to each user plane entity (210) further indicates a forwarding action rule, FAR, (304) to be associated by the user plane entity (210) to the application-layer processed service data flow (214).

16. The control plane entity (220) according to claim 14 or 15, further configured to: derive the first PDR (301) for each selected user plane entity (210) based on a condition for performing the application layer processing received from an application; wherein the condition for performing the application layer processing includes one or more of

- a processing load of an application logic (213a) at the application layer (213) of the user plane entity (210) is lower than a threshold;

- a queuing delay at the application layer (213) and/or application logic (213a) of the user plane entity (210) is lower than a threshold;

- an availability status of the application logic (213a) at the application layer (213) of the user plane entity (210) indicates that the application logic (213a) is available;

17. The control plane entity (220) according to claim 14 or 15, further configured to: derive the second PDR (302) for each selected user plane entity (210) from a list of service description of the service data flow (211) received from an application.

18. The control plane entity (220) according to claim 14 or 15, further configured to: generate the second PDR (302) or a PDR generation configuration (1301) for each selected user plane entity (210) based on a service description of the service data flow (211) and a PDR generating instruction (1101) received from an application or another control plane entity.

19. The control plane entity (220) according to claim 18, wherein the PDR generation instruction (1101) or the PDR generation configuration (1301) comprises one or more of: an end-to-end flow description of the service data flow (211); an instruction to use a new source and/or destination port; an instruction to use a local source IP address; an instruction to use a local flow ID generated by the user plane entity (210).

20. The control plane entity (220) according to claim 14 or 15, wherein the configuration information (803) provided to each selected user plane entity (210) includes a PDR generation configuration (1301) for configuring the user plane entity (210) to generate the second PDR (302) that is to be associated by the user plane entity (210) to the application-layer processed service data flow (214).

21. A method (600) for a user plane entity (210) of a network capable of application layer processing of a service data flow (211, 214) transported through the network, the method (600) comprising: receiving (601), from a second user plane entity of the network, a network traffic flow comprising a service data flow (211) from an application; determining (602) whether the received service data flow (211) is intended for application layer processing at the user plane entity (210); and/or determining (603) whether a condition for performing the application layer processing is fulfilled; and if the received service data flow (211) is intended for application layer processing and/or the condition for performing the application layer processing is fulfilled:

- sending (604) the received service data flow (211) from a network layer (212) to an application layer logic (213a) at the application layer (213) of the user plane entity (210);

- performing (605) one or more first processing operations at the application layer logic (213a) on the received service data flow (211).

22. A method (1600) for a control plane entity (220) for controlling an application layer processing of a service data flow (211, 214) transported through a user plane of a network, the method (1600) comprising: determining (1602), for each user plane entity (210) of one or more selected user plane entities (210), configuration information (803) indicating a first packet detection rule, PDR, (301) for the user plane entity (210) to detect a service data flow (211) requiring application layer processing at the user plane entity (210) and a second PDR (302) to be associated by the user plane entity (210) with an application-layer processed service data flow (214); and providing (1603) the determined configuration information to each user plane entity of the one or more selected user plane entities.

23. A computer program comprising instructions which, when the program is executed by a computer, cause the computer to perform the method (600, 1600) according to claim 21 or 22.