WO2023121665A1

WO2023121665A1 - Cross-domain resource coordination with ue episodic mobility

Info

Publication number: WO2023121665A1
Application number: PCT/US2021/064884
Authority: WO
Inventors: Bilgehan Erman; Ejder BASTUG; Bruce Cilli; Andrea Francini; Raymond Miller; Charles Payette; Sameerkumar Sharma
Original assignee: Nokia Technologies Oy; Nokia Of America Corporation
Priority date: 2021-12-22
Filing date: 2021-12-22
Publication date: 2023-06-29

Abstract

Method, apparatuses, and computer program products provide cross-domain resource coordination using predictions of user equipment (UE) episodic mobility routes. Various embodiments described herein can be applied to establish and/or migrate data sessions for low- latency applications with edge cloud resources. An example method performed by network functions of a network domain includes predicting an episodic mobility route for a UE and obtaining candidate edge application server instances identified based on their respective locations in relation to the episodic mobility route. The method further includes determining quality metrics for the candidate server instances in combination with anchor points, the quality metrics including data network latency. The method further includes selecting an edge server instance and/or an anchor point, and either establishing a new data session with the selected server instance via the selected anchor point, or migrating an existing data session to the selected server instance via the selected anchor point.

Description

CROSS-DOMAIN RESOURCE COORDINATION WITH UE EPISODIC MOBILITY

TECHNOLOGICAL FIELD

[0001] An example embodiment relates generally to techniques for resource coordination in telecommunication networks.

BACKGROUND

[0002] Among the broad set of new applications that are being enabled by modem telecommunication networks (e.g., Long Term Evolution or LTE, 5^th Generation (5G) New Radio), there are ones, such as cloud gaming and augmented/virtual reality (AR/VR), that demand ultra-low-latency communications. Proper operation of these applications requires specific resources, such as application servers and network gateways or functions, to be in proximity of the client endpoints with a latency distance typically lower than 10-13 milliseconds. [0003] The latency distance, which can be measured by the round-trip time between the application client and the server in some examples, heavily depends upon the type of access network used by the client and how it is configured, as the end-to-end connection traverses many network functions. While it may be relatively easy to configure fixed access networks to guarantee low client-server latency, many technical challenges emerge with mobile access networks. Due to user mobility, there are more network functions within a given data path. Additional challenges arise with the placement of those functions in different administrative domains that may be controlled by different entities.

BRIEF SUMMARY

[0004] Generally, various embodiments of the present disclosure address technical challenges relating to ultra-low-latency communications in mobile access networks across different administrative domains thereof. In particular, various embodiments described herein provide resource coordination between an application domain and a mobile network domain using predictions of user equipment (UE) episodic mobility. Examples of cross-domain resource coordination described herein cover edge cloud scenarios in which application architecture and signaling are independent of mobile network specification. In various example embodiments, example interface procedures enable an application to coordinate far-edge cloud resources with the mobile network domain so that the latency distance between an application client and its server is kept within a tight latency bound.

[0005] According to one aspect of the present disclosure, an apparatus is provided. The apparatus includes at least one processor and at least one memory including computer program code. In one example embodiment, the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to predict an episodic mobility route of a user equipment (UE) running one or more instances of an application. The at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus to obtain an indication of one or more edge server instances for the application. The one or more edge server instances are indicated based at least in part on a respective location of the one or more edge server instances in relation to at least a portion of the episodic mobility route of the UE. The at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus to determine quality metrics for the one or more edge server instances in combination with one or more network anchor points. The at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus to establish a data session between the UE and a particular edge server instance through a particular network anchor point. The particular edge server instance and the particular network anchor point are selected based at least in part on corresponding quality metrics.

[0006] In various embodiments, the indication of the one or more edge server instances is obtained from an application management device associated with the application in response to providing the episodic mobility route of the UE to the application management device.

[0007] In various embodiments, the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to obtain a second indication of one or more second edge server instances for the application responsive to an updated location of the UE and/or the episodic mobility rate, select a particular second edge server instance based at least in part on quality metrics determined for the one or more second edge server instances in combination with the one or more network anchor points, and migrate the data session to the particular second edge server instance.

[0008] In various embodiments, the episodic mobility route of the UE includes a plurality of waypoints and a plurality of dynamic probabilities corresponding the plurality of waypoints. In various embodiments, the one or more edge server instances are geographically located within an edge zone of one or more edge zones spanned by the episodic mobility route.

[0009] In various embodiments, the indication of the one or more edge server instances includes, for a respective edge server instance: a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes. In various embodiments, the quality metrics for the respective edge server instance are determined based at least in part on the sounding service node, the sounding service method, and the sounding service attributes. In various embodiments, the quality metrics for the respective edge server instance in combination with a respective network anchor point comprise a latency between the respective edge server instance and the respective network anchor point. In various embodiments, the particular edge server instance and the particular network anchor point are selected based at least in part on minimization of the latency.

[0010] According to one aspect of the present disclosure, an apparatus is provided. The apparatus includes at least one processor and at least one memory including computer program code. In one example embodiment, the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to obtain a prediction of an episodic mobility route of a UE on which one or more instances of an application reside. The at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus to identify one or more edge zones spanned by the episodic mobility route. The at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus to determine a set of edge server instances for the application and positioned within the one or more edge zones. The at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus to receive an indication of a particular edge server instance selected from the set of edge server instances. The at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus to enable establishment of a data session between the UE and the particular edge server instance.

[0011] In various embodiments, the indication of a particular edge server instance is received from the UE. In various embodiments, the indication of the particular edge server instance is received responsive to providing, for a respective edge server instance of the set of edge server instances, a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes.

[0012] In various embodiments, the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to periodically receive subsequent predictions of the episodic mobility route of the UE, and responsive to receiving a subsequent prediction of the episodic mobility route of the UE, determine a second set of edge server instances for the application and positioned within a second edge zone spanned by the subsequent prediction of the episodic mobility route. In various embodiments, the prediction of the episodic mobility route is obtained responsive to receiving an indication of session establishment from the UE.

[0013] According to another aspect of the present disclosure, an apparatus is provided, the apparatus including means for predicting an episodic mobility route of a UE on which one or more instances of an application reside. The apparatus further includes means for obtaining an indication of one or more edge server instances for the application. The one or more edge server instances are indicated based at least in part on a respective location of the one or more edge server instances in relation to at least a portion of the episodic mobility route of the UE. The apparatus further includes means for determining quality metrics for the one or more edge server instances in combination with one or more network anchor points. The apparatus further includes means for establishing a data session between the UE and a particular edge server instance through a particular network anchor point. The particular edge server instance and the particular network anchor point are selected based at least in part on corresponding quality metrics.

[0014] In various embodiments, the indication of the one or more edge server instances is obtained from an application management device associated with the application in response to providing the episodic mobility route of the UE to the application management device.

[0015] In various embodiments, the apparatus further includes means for obtaining a second indication of one or more second edge server instances for the application responsive to an updated location of the UE and/or the episodic mobility rate, means for selecting a particular second edge server instance based at least in part on quality metrics determined for the one or more second edge server instances in combination with the one or more network anchor points, and means for migrating the data session to the particular second edge server instance. [0016] In various embodiments, the episodic mobility route of the UE includes a plurality of waypoints and a plurality of dynamic probabilities corresponding the plurality of waypoints. In various embodiments, the one or more edge server instances are geographically located within an edge zone of one or more edge zones spanned by the episodic mobility route.

[0017] In various embodiments, the indication of the one or more edge server instances includes, for a respective edge server instance: a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes. In various embodiments, the quality metrics for the respective edge server instance are determined based at least in part on the sounding service node, the sounding service method, and the sounding service attributes. In various embodiments, the quality metrics for the respective edge server instance in combination with a respective network anchor point comprise a latency between the respective edge server instance and the respective network anchor point. In various embodiments, the particular edge server instance and the particular network anchor point are selected based at least in part on minimization of the latency.

[0018] According to another aspect of the present disclosure, an apparatus is provided, the apparatus including means for obtaining a prediction of an episodic mobility route of a UE on which one or more instances of an application reside. The apparatus further includes means for identifying one or more edge zones spanned by the episodic mobility route. The apparatus further includes means for determining a set of edge server instances for the application and positioned within the one or more edge zones. The apparatus further includes means for receiving an indication of a particular edge server instance selected from the set of edge server instances. The apparatus further includes means for enabling establishment of a data session between the UE and the particular edge server instance.

[0019] In various embodiments, the indication of a particular edge server instance is received from the UE. In various embodiments, the indication of the particular edge server instance is received responsive to providing, for a respective edge server instance of the set of edge server instances, a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes.

[0020] In various embodiments, the apparatus further includes means for periodically receiving subsequent predictions of the episodic mobility route of the UE and means for determining, responsive to receiving a subsequent prediction of the episodic mobility route of the UE, a second set of edge server instances for the application and positioned within a second edge zone spanned by the subsequent prediction of the episodic mobility route. In various embodiments, the prediction of the episodic mobility route is obtained responsive to receiving an indication of session establishment from the UE.

[0021] According to yet another aspect of the present disclosure, a method is provided, the method including predicting an episodic mobility route of a UE on which one or more instances of an application reside. The method further includes obtaining an indication of one or more edge server instances for the application. The one or more edge server instances are indicated based at least in part on a respective location of the one or more edge server instances in relation to at least a portion of the episodic mobility route of the UE. The method further includes determining quality metrics for the one or more edge server instances in combination with one or more network anchor points. The method further includes establishing a data session between the UE and a particular edge server instance through a particular network anchor point. The particular edge server instance and the particular network anchor point are selected based at least in part on corresponding quality metrics.

[0022] In various embodiments, the indication of the one or more edge server instances is obtained from an application management device associated with the application in response to providing the episodic mobility route of the UE to the application management device.

[0023] In various embodiments, the method further includes obtaining a second indication of one or more second edge server instances for the application responsive to an updated location of the UE and/or the episodic mobility rate, selecting a particular second edge server instance based at least in part on quality metrics determined for the one or more second edge server instances in combination with the one or more network anchor points, and migrating the data session to the particular second edge server instance.

[0024] In various embodiments, the episodic mobility route of the UE includes a plurality of waypoints and a plurality of dynamic probabilities corresponding the plurality of waypoints. In various embodiments, the one or more edge server instances are geographically located within an edge zone of one or more edge zones spanned by the episodic mobility route.

[0025] In various embodiments, the indication of the one or more edge server instances includes, for a respective edge server instance: a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes. In various embodiments, the quality metrics for the respective edge server instance are determined based at least in part on the sounding service node, the sounding service method, and the sounding service attributes. In various embodiments, the quality metrics for the respective edge server instance in combination with a respective network anchor point comprise a latency between the respective edge server instance and the respective network anchor point. In various embodiments, the particular edge server instance and the particular network anchor point are selected based at least in part on minimization of the latency.

[0026] According to yet another aspect of the present disclosure, a method is provided, the method including obtaining a prediction of an episodic mobility route of a user equipment (UE) on which one or more instances of an application reside. The method further includes identifying one or more edge zones spanned by the episodic mobility route. The method further includes determining a set of edge server instances for the application and positioned within the one or more edge zones. The method further includes receiving an indication of a particular edge server instance selected from the set of edge server instances. The method further includes enabling establishment of a data session between the UE and the particular edge server instance.

[0027] In various embodiments, the indication of a particular edge server instance is received from the UE. In various embodiments, the indication of the particular edge server instance is received responsive to providing, for a respective edge server instance of the set of edge server instances, a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes.

[0028] In various embodiments, the method further includes periodically receiving subsequent predictions of the episodic mobility route of the UE and determining, responsive to receiving a subsequent prediction of the episodic mobility route of the UE, a second set of edge server instances for the application and positioned within a second edge zone spanned by the subsequent prediction of the episodic mobility route. In various embodiments, the prediction of the episodic mobility route is obtained responsive to receiving an indication of session establishment from the UE.

[0029] According to yet another aspect of a present disclosure, a computer program product is provided. The computer program product includes at least one non-transitory computer readable storage medium having computer executable program code instructions stored therein. The computer executable program code instructions include program code instructions configured, upon execution, to predict an episodic mobility route of a UE on which one or more instances of an application reside. The computer executable program code instructions include program code instructions further configured, upon execution, to obtain an indication of one or more edge server instances for the application. The one or more edge server instances are indicated based at least in part on a respective location of the one or more edge server instances in relation to at least a portion of the episodic mobility route of the UE. The computer executable program code instructions include program code instructions further configured, upon execution, to determine quality metrics for the one or more edge server instances in combination with one or more network anchor points. The computer executable program code instructions include program code instructions further configured, upon execution, to establish a data session between the UE and a particular edge server instance through a particular network anchor point. The particular edge server instance and the particular network anchor point are selected based at least in part on corresponding quality metrics.

[0030] In various embodiments, the indication of the one or more edge server instances is obtained from an application management device associated with the application in response to providing the episodic mobility route of the UE to the application management device.

[0031] In various embodiments, the computer executable program code instructions include program code instructions further configured, upon execution, to obtain a second indication of one or more second edge server instances for the application responsive to an updated location of the UE and/or the episodic mobility rate, to select a particular second edge server instance based at least in part on quality metrics determined for the one or more second edge server instances in combination with the one or more network anchor points, and to migrate the data session to the particular second edge server instance.

[0032] In various embodiments, the episodic mobility route of the UE includes a plurality of waypoints and a plurality of dynamic probabilities corresponding the plurality of waypoints. In various embodiments, the one or more edge server instances are geographically located within an edge zone of one or more edge zones spanned by the episodic mobility route.

[0033] In various embodiments, the indication of the one or more edge server instances includes, for a respective edge server instance: a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes. In various embodiments, the quality metrics for the respective edge server instance are determined based at least in part on the sounding service node, the sounding service method, and the sounding service attributes. In various embodiments, the quality metrics for the respective edge server instance in combination with a respective network anchor point comprise a latency between the respective edge server instance and the respective network anchor point. In various embodiments, the particular edge server instance and the particular network anchor point are selected based at least in part on minimization of the latency.

[0034] According to yet another aspect of a present disclosure, a computer program product is provided. The computer program product includes at least one non-transitory computer readable storage medium having computer executable program code instructions stored therein. The computer executable program code instructions include program code instructions configured, upon execution, to obtain a prediction of an episodic mobility route of a UE on which one or more instances of an application reside. The computer executable program code instructions include program code instructions further configured, upon execution, to identify one or more edge zones spanned by the episodic mobility route. The computer executable program code instructions include program code instructions further configured, upon execution, to determine a set of edge server instances for the application and positioned within the one or more edge zones. The computer executable program code instructions include program code instructions further configured, upon execution, to receive an indication of a particular edge server instance selected from the set of edge server instances. The computer executable program code instructions include program code instructions further configured, upon execution, to enable establishment of a data session between the UE and the particular edge server instance.

[0035] In various embodiments, the indication of a particular edge server instance is received from the UE. In various embodiments, the indication of the particular edge server instance is received responsive to providing, for a respective edge server instance of the set of edge server instances, a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes.

[0036] In various embodiments, the computer executable program code instructions include program code instructions further configured, upon execution, to periodically receive subsequent predictions of the episodic mobility route of the UE and determine, responsive to receiving a subsequent prediction of the episodic mobility route of the UE, a second set of edge server instances for the application and positioned within a second edge zone spanned by the subsequent prediction of the episodic mobility route. In various embodiments, the prediction of the episodic mobility route is obtained responsive to receiving an indication of session establishment from the UE.

[0037] The above summary is provided merely for purposes of summarizing some example embodiments to provide a basic understanding of some aspects of the invention. Accordingly, it will be appreciated that the above-described embodiments are merely examples and should not be construed to narrow the scope or spirit of the invention in any way. It will be appreciated that the scope of the invention encompasses many potential embodiments in addition to those here summarized, some of which will be further described below. Other features, aspects, and advantages of the subject matter will become apparent from the description, the drawings, and the claims.

BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Having thus described certain example embodiments of the present disclosure in general terms above, non-limiting and non-exhaustive embodiments of the subject disclosure will now be described with reference to the accompanying drawings, which are not necessarily drawn to scale. The components illustrated in the accompanying drawings may or may not be present in certain embodiments described herein. Some embodiments may include fewer (or more) components than those shown in the drawings.

[0039] Figures 1 A and IB provide diagrams demonstrating different example regional arrangements of network resources, in accordance with various embodiments described herein; [0040] Figure 2 provides a diagram describing cross-domain latency, in accordance with various embodiments described herein;

[0041] Figure 3 provides an example diagram for cross-domain resource coordination using predicted UE episodic mobility, in accordance with various embodiments of the present disclosure;

[0042] Figure 4 provides a diagram illustrating example UE episodic mobility used at least in part for cross-domain resource coordination, in accordance with various embodiments;

[0043] Figure 5 illustrates an example cross-domain architecture for cross-domain resource coordination using predicted UE episodic mobility, in accordance with various embodiments of the present disclosure; [0044] Figure 6 provides a block diagram of an example apparatus configured for crossdomain resource coordination using predicted UE episodic mobility, in accordance with various embodiments of the present disclosure;

[0045] Figures 7A and 7B each provide a flowchart illustrating example operations performed for cross-domain resource coordination with UE episodic mobility predictions, in accordance with various embodiments of the present disclosure;

[0046] Figure 8 illustrates an example message structure used in cross-domain resource coordination, in accordance with various embodiments of the present disclosure;

[0047] Figures 9A and 9B describe various quality metrics and determination thereof during cross-domain resource coordination, in accordance with various embodiments of the present disclosure; and

[0048] Figures 10A and 10B illustrate sequence diagrams describing example operations and interactions for cross-domain resource coordination using UE episodic mobility predictions.

DETAILED DESCRIPTION

[0049] Some embodiments of the present invention will now be described more fully hereinafter with reference to the accompanying drawings, in which some, but not all, embodiments of the invention are shown. Indeed, various embodiments of the invention may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will satisfy applicable legal requirements. Like reference numerals refer to like elements throughout. As used herein, the terms “data,” “content,” “information,” “electronic information,” “signal,” “command,” and similar terms may be used interchangeably to refer to data capable of being captured, transmitted, received, and/or stored in accordance with various embodiments of the present disclosure. Thus, use of any such terms should not be taken to limit the spirit and scope of embodiments of the present disclosure. Further, where a first computing device is described herein to receive data from a second computing device, it will be appreciated that the data may be received directly from the second computing device or may be received indirectly via one or more intermediary computing devices, such as, for example, one or more servers, relays, routers, network access points, base stations, hosts, repeaters, and/or the like, sometimes referred to herein as a “network.” Similarly, where a first computing device is described herein as sending data to a second computing device, it will be appreciated that the data may be sent or transmitted directly to the second computing device or may be sent or transmitted indirectly via one or more intermediary computing devices, such as, for example, one or more servers, remote servers, cloud-based servers (e.g., cloud utilities), relays, routers, network access points, base stations, hosts, repeaters, and/or the like.

[0050] The term “comprising” means including but not limited to and should be interpreted in the manner it is typically used in the patent context. Use of broader terms such as comprises, includes, and having should be understood to provide support for narrower terms such as consisting of, consisting essentially of, and comprised substantially of. Furthermore, to the extent that the terms “includes” and “including,” and variants thereof are used in either the detailed description or the claims, these terms are intended to be inclusive in a manner similar to the term “comprising.”

[0051] The phrases “in one embodiment,” “according to one embodiment,” “in some embodiments,” “in various embodiments”, and the like generally refer to the fact that the particular feature, structure, or characteristic following the phrase may be included in at least one embodiment of the present disclosure, but not necessarily all embodiments of the present disclosure. Thus, the particular feature, structure, or characteristic may be included in more than one embodiment of the present disclosure such that these phrases do not necessarily refer to the same embodiment.

[0052] As used herein, the terms “example,” “exemplary,” and the like are used to mean “serving as an example, instance, or illustration.” Any implementation, aspect, or design described herein as “example” or “exemplary” is not necessarily to be construed as preferred or advantageous over other implementations, aspects, or designs. Rather, use of the terms “example,” “exemplary,” and the like are intended to present concepts in a concrete fashion. [0053] If the specification states a component or feature “may,” “can,” “could,” “should,”

“would,” “preferably,” “possibly,” “typically,” “optionally,” “for example,” “often,” or “might” (or other such language) be included or have a characteristic, that particular component or feature is not required to be included or to have the characteristic. Such component or feature may be optionally included in some embodiments, or it may be excluded.

[0054] As used herein, the term “computer-readable medium” refers to non-transitory storage hardware, non-transitory storage device or non-transitory computer system memory that may be accessed by a controller, a microcontroller, a computational system or a module of a computational system to encode thereon computer-executable instructions or software programs. A non-transitory “computer-readable medium” may be accessed by a computational system or a module of a computational system to retrieve and/or execute the computer-executable instructions or software programs encoded on the medium. Examples of non-transitory computer-readable media may include, but are not limited to, one or more types of hardware memory, non-transitory tangible media (for example, one or more magnetic storage disks, one or more optical disks, one or more USB flash drives), computer system memory or random-access memory (such as, DRAM, SRAM, EDO RAM), and the like.

[0055] Additionally, as used herein, the term ‘circuitry’ refers to (a) hardware-only circuit implementations (e.g., implementations in analog circuitry and/or digital circuitry); (b) combinations of circuits and computer program product(s) comprising software and/or firmware instructions stored on one or more computer readable memories that work together to cause an apparatus to perform one or more functions described herein; and (c) circuits, such as, for example, a microprocessor(s) or a portion of a microprocessor s), that require software or firmware for operation even if the software or firmware is not physically present. This definition of ‘circuitry’ applies to all uses of this term herein, including in any claims. As a further example, as used herein, the term ‘circuitry’ also includes an implementation comprising one or more processors and/or portion(s) thereof and accompanying software and/or firmware. As another example, the term ‘circuitry’ as used herein also includes, for example, a baseband integrated circuit or applications processor integrated circuit for a mobile phone or a similar integrated circuit in a server, a cellular network device, other network device (such as a core network apparatus), field programmable gate array, and/or other computing device.

[0056] As used herein, the term “far-edge cloud” may be associated with and/or may generally refer to network communications having an application client-server latency distance ranging from hundreds of microseconds to few milliseconds and a physical distance ranging between 5 and 40 km. That is, the far-edge cloud may be a portion of a cloud infrastructure spatially relative and specific to an application client-server combination. Other cloud regions of a cloud infrastructure may include a cell site, an edge cloud, and the central cloud. As used in the present disclosure, the term “edge cloud” may occasionally be used to generally refer to both the far-edge cloud and the edge cloud unless a distinction must be made. [0057] Various embodiments of the present disclosure provide cross-domain resource coordination to enable ultra-low-latency communications for a given application, and specifically, various embodiments provide resource coordination between an application domain and a network domain. As used herein, the network domain or mobile network domain may generally refer to an administrative domain of a mobile access network (e.g., a cellular network) controlled by a mobile network operator (MNO), while the application domain may generally refer to an administrative domain of the given application. As understood by those of skill in the field of the present disclosure, these two domains may be distinct and strictly isolate in many examples, and various embodiments described herein provide unique solutions to address technical challenges that arise in resource coordination across these two domains for ultra-low- latency communications.

[0058] As such, various embodiments described herein benefit applications that may have tight latency requirements and a broad range of throughput requirements. Some example applications may also be elastic in throughput, adjusting their data rates to the bandwidth conditions of their data paths. Generally, the term “Low-Latency Plus” (LL+) may refer to applications having a low-latency requirement and a throughput requirement that may range from very low to very high. In the example of 3^rd Generation Partnership Project (3GPP) 5^th Generation (5G) mobile networks, LL+ applications that are mission-critical and low-bandwidth (e.g., autonomous driving applications) may be supported by Ultra Reliable Low Latency Communications (URLLC) network services, while LL+ applications that are elastic and over- the-top (e.g., AR/VR applications, game streaming applications, video conference applications) may be running over enhanced Mobile Broad-Band (eMBB) network services. Generally then, URLLC network services and eMBB network services may both be considered to be invoked for the support of LL+ applications within the scope of the present disclosure.

[0059] Because various embodiments for cross-domain resource coordination can be applied to 5G-standardized mobile networks and applications communicating thereupon, various 5G- specific terminology may be occasionally used in the present disclosure. For instance, the 5G- specific term “user plane function” (UPF) may generally refer to a network domain’s anchor point for data flows to and from the application domain. However, it is to be understood that various embodiments described herein providing cross-domain resource coordination are not limited to 5G networks and environments and can be applied generally to any suitable network and environment. 5G-specific terms and entities can be understood in representation of generic terms and entities in the context of various technical specifications of 3GPP, including TS 23.748 Release 17, TS 23.501 Release 17, TS 23.502 Release 17, TS 23.548 Release 17, TS 23.288 Release 17, and TS 23.558 Release 17 — each of which is incorporated by reference herein in its entirety.

[0060] Example embodiments of cross-domain resource coordination as described herein may use and be based at least in part on prediction of user equipment (UE) episodic motion, or predicted episodic routes travelled by a user endpoint running instances of an application client. Specifically, UE episodic motion can be predicted by and acquired from the network domain and functions thereof, such as a Network Data Analytics Function (NWDAF) and associated Network Exposure Function (NEF) (e.g., an interaction point between the application domain and the network domain for control messaging) of a 5G network. A predicted episodic mobility route may include a full set of predicted waypoints for a complete trip (e.g., an episode), with the UE predicted to be positioned at and travel through said waypoints. As such, an episodic mobility route includes more complete information than plain mobility predictions that may only involve motion vectors to predict one or more upcoming waypoints (or direction and velocity only). For example, an episodic mobility route may be a routine trip or commute (e.g., via car or train) from a starting point to a destination taken by the UE or a user operating the UE.

Generally, an episodic mobility route can be predicted (given acceptance of location tracking with the application domain and/or the network domain by the user of the UE, in some examples) due to repeating mobility patterns of people’s lives, such as travelling to a workplace, a home, homes of close associates (e.g., family, friends), shopping or entertainment locations, a vacation home, and/or the like each along the same route. It may be understood that random trips to unusual locations along unusual routes typically represent only a small fraction of overall mobility of a given UE.

[0061] Therefore, for a majority of the UE motions, it is possible to predict with high likelihood which route a UE motion will involve until it ends (e.g., start, way points, destination, average velocity, time duration, and any combination thereof). Accordingly, various embodiments of cross-domain resource coordination can integrate and use highly predictable episodic (e.g., routine) mobility of UEs to thereby provide ultra-low-latency communications between a moving UE and an application server. In various embodiments, edge-cloud resources can be allocated in association with the predicted motion of a UE in order to support the connectivity of the UE and the applications ran by the UE. That is, if an LL+ application is activated during a UE motion that fits an episodic pattern, then the resources (in the network domain and/or the application domain) that will be needed to support the UE can be easily predicted. Compared to close range mobility predictions or time-based extended mobility predictions, the episodic mobility predictions used in various embodiments described herein are important and technically improved measures, as the wide time range of the episodic mobility predictions (e.g., predictions spanning from the start of a trip to the end of a trip) enables opportunities for using more advanced policies for optimizing resources and coordination thereof.

[0062] In providing resource coordination within and between an application domain and a network domain using predictions of UE episodic mobility, various embodiments of the present disclosure provide solutions to at least four technical challenges relating to low-latency communication and network operation in edge computing. These four technical challenges, as described herein, include deployment asymmetry of dependent resources, effective latency metrics, latency assessment methods, and edge cloud resource constraints.

[0063] Figures 1 A and IB provide diagrams illustrating technical challenges relating to deployment asymmetry of dependent resources and solutions thereof. Referring first to Figure 1 A, an application domain 110 and a network domain 120 are shown as separate and distinct domains; for example, the application domain 110 and the network domain 120 may be independently controlled by different entities. As shown in Figure 1 A, a cloud architecture or environment may include different cloud regions, including a cell site, a far-edge region, an edge region, and a central region. Generally, an application that requires ultra-low-latency communication (e.g., autonomous driving applications, AR/VR applications, game streaming applications, video conference applications) should be run using resources that reside in far-edge cloud platforms; that is, latency may be minimized using far-edge resources as opposed to communication between inefficiently-located resources.

[0064] Figure 1 A specifically illustrates a misaligned and inefficient, in some examples, placement of resources. In particular, the resources include application servers or application functions (e.g., edge application server 111) in the application domain 110 and network functions 124 (e.g., user plane functions in accordance with 5G specifications) in the network domain 120. With fixed networks, it can be easily assured that all network functions 124 are embedded in the data path between an edge application server 111 and a UE having an application client in a latency-efficient manner. However, with mobile networks, the positioning of the network functions 124 (relative to the UE and/or the edge application server 111) can vary or change over time. Figure 1 A, for example, shows a network function 124 being positioned in an edge region while an edge application server I l l is positioned in a far-edge region, thereby resulting in highly inefficient configuration of a data path 140. In various examples, detrimental application disruptions may result from highly inefficient configuration of a data path 140, as exemplified in Figure 1 A, even when such inefficient data paths are only temporary.

[0065] In contrast, Figure IB generally illustrates improved coordination of resources including network functions 124 and edge application servers 111. As shown, the network function 124 and the edge application server 111 are both within the far-edge region, resulting in a more efficient data path and lower latency in data communication between a UE and the edge application server 111.

[0066] Applications with functions that reside in the far-edge cloud may be deployed as integral part of the infrastructure of the network domain. Hardened applications with high development cost and critical functionality can benefit from this architectural choice if they do not change much over time. However, such strict dependency between the application and the network domains is highly restrictive to the evolution of the application, especially when the application is developed and deployed by an independent third party, for purposes such as entertainment. For example, UE-driven 3GPP procedures for application server discovery are also not always adequate for LL+ applications, because their sessions require scaling and migration of resources, which are typically controlled by a dedicated application manager based on multiple constraints specific to their domain.

[0067] In the operation of an LL+ application, the movement of a UE may require the network domain to migrate its anchor point (e.g., a user plane function in a 5G-specific context) and the application domain to migrate the edge application server (or instance thereof) that serves it. The execution of these migrations is not trivial because they involve the timely coordination of multiple connections and the preservation of their continuity and quality of service metrics. Thus, real-time scheduling of service sessions involving dynamic resource changes with minimal service quality interruption is a complex procedure, especially when a high volume of incoming transactions is considered. There are also the added cost and the performance optimization constraints induced by the special nature of the edge cloud (compared to the large-scale central cloud). The severity of the challenge increases with the speed at which the UE moves relative to the average cell coverage radius. For instance, assuming a 20 km separation between edge cloud sites, passengers playing immersive video games on a train that travels a 500 km distance will traverse 25 edge cloud sites. At a speed of 250 km/h, the microservices of the video game applications will switch cloud platforms every 4.8 minutes. Latency, reliability, and bandwidth requirements may become even more stringent over time. [0068] Referring next to Figure 2, a diagram is provided to demonstrate technical challenges related to effective latency metrics that are addressed through various embodiments of the present disclosure. Figure 2 illustrates an edge application server 111 and a network function 124. Figure 2 further illustrates a peering point 202 that a data path 140 between the network function 124 and the edge application server 111 traverses.

[0069] Specifically, Figure 2 demonstrates that geolocation information is not sufficient for coordination of cross-domain resources (e.g., the network function 124 and the edge application server 111) with respect to latency distance. As in the illustrated embodiment, a network function 124 and an edge application server 111 may be positioned near each other in a first geographic region 200A, but there may be a much larger latency between them than expected. For instance, one or more peering points along the data path 140 may be far apart from each other and/or from at least one of the network function 124 or the edge application server 111, thereby resulting in a large propagation latency. In the illustrated embodiment, for example, the peering point 202 is positioned in a second geographic region 200B that is separate and may be relatively distant from the first geographic region 200A. Additionally or alternatively, a large traffic load may exist on a network node that is interconnecting the two resources, resulting in a large queuing latency.

[0070] Further, “location” may also have different meanings and/or associations with respect to each of the application domain 110 and the network domain 120; for example, a location may relate to a target area, a cell and cell ID, an area of interest, a latency-zone group, referenced location coordinates (e.g., via a global positioning system, a global navigation satellite system, and/or the like), and/or the like. Since latency may be one of various key metrics used for crossdomain resource coordination, various embodiments normalize a cross-domain understanding of “location” with respect to a latency distance from the application’s perspective, or with respect to the application domain. Generally, various embodiments additionally normalize and homogenize understanding of other various quality metrics between the application domain and the network domain.

[0071] Another technical challenge identified above relates to latency assessment methods. As discussed previously, geo-location of application endpoints may not always be a good indication of the expected latency measures. Also, for a given application, assessment of the end- to-end latency may be different. For example, the expected maximum latency bound may be absolute, or only a quantile boundary may be expected. Generally then, the method for latency measurement may be different for different applications and different network communication protocols. For instance, for Internet Protocol (IP), pinging may be expected only if the edge application server 111 has pinging enabled. A given application may also have requirements for latency measures at a different protocol level.

[0072] As identified above, various embodiments also address technical challenges relating to edge cloud resource constraints. Because of their functional purpose, edge clouds (e.g., far- edge regions, edge regions) are highly distributed and positioned to be close to user endpoints (e.g., UEs). With dynamic load demands, this distributed configuration creates technical challenges in maintaining resource utilization at an optimum range — unlike central configurations in which statistical multiplexing can be used to smooth out resource utilization against load fluctuations because of their greater size.

[0073] In dynamic load demand situations for edge clouds, having the capability to predict future load as accurately as possible can significantly improve resource management. Example resource management policies may give priority to maintaining the quality of the active sessions rather than admitting new sessions if bad service quality is anticipated because of resource shortage. Example resource management policies may also prepare data caches ahead of time, or may prepare time-consuming platform reconfigurations. Accordingly, various embodiments of the present disclosure provide solutions to the technical challenges relating to edge cloud resource constraints in view of dynamic load demand situations, as various embodiments involve prediction of UE episodic mobility in order to inform such resource management policies and enable efficient and intelligent resource management/coordination. [0074] Having discussed technical challenges addressed by various embodiments described herein, various detailed aspects of cross-domain resource coordination using predictions of UE episodic mobility in accordance with various embodiments of the present disclosure are now described. As described, various embodiments provide resource coordination between an application domain 110 and a network domain 120, and may specifically be applied in edge clouds for LL+ applications with minimal dependencies between the two domains.

[0075] In various embodiments, network data analytic functions within the network domain 120 determine episodic mobility predictions of a UE, such predictions understood to be highly and reasonable accurate considering the relatively routine lives of UE users. In various embodiments, the network data analytic functions of the network domain 120 provide the episodic mobility predictions to the application domain 110, such as to an application management device. In general, the application domain 110 includes a key operator that orchestrates deployment of its functional components at the edge cloud.

[0076] In various embodiments, an ordered list of edge application servers 111 (and/or instances thereof) may be identified by the application domain 110 based at least in part on the episodic mobility predictions and various other factors, and the ordered list is recognized by the network domain 120. Specifically, the edge application servers 111 may identified by the application domain 110 based at least in part on proximity to the predicted episodic mobility route. With the list of edge application servers 111 and instances, a selection of a particular edge application server 111 and/or instance is made at the network domain 120 according to various quality metrics for each of the list of edge application servers 111 and instances. Specifically, in various embodiments, the network domain 120 determines quality metrics, such as latency, for different edge application servers 111 indicated by the application domain 110, and in particular, determines certain quality metrics in accordance with certain measurement methods/schemes that are specified by the application domain 110. For instance, the application domain 110 may provide a script for execution by the network domain 120 for determination of quality metrics for an edge application server 111.

[0077] Figure 3 depicts an example scenario within which example operations for crossdomain resource coordination can be performed. In the illustrated embodiment, edge application servers 111 A-D and network functions 124A-D are spatially and geographically distributed to provide edge clouds for a UE 300. Communication between the UE 300 and at least the network functions 124A-D can be specifically enabled via base stations 302X-Z, which are also positioned throughout the environment. Figure 3 further illustrates a latency boundary 310 for a LL+ application, and resources positioned within the latency boundary 310 may be used with the UE 300 for a data session that satisfies latency requirements for the LL+ application — at the UE’s current position.

[0078] In accordance with various embodiments, the application domain 110 is made aware of movement of a UE 300 towards an edge cloud set including edge application server 11 ID and network function 124D. In some example embodiments, the application domain 110 may be provided with a motion vector 320 (e.g., direction, speed) of the UE 300 according to a prediction of the UE’s episodic mobility or movement. With movement of the UE 300, the latency boundary 310 may not be necessarily accurate in identifying resources that can be used to satisfy latency requirements for the LL+ application.

[0079] With the knowledge of the UE’s episodic mobility in the context of various resources (e.g., UE 300 is moving towards edge application server 11 ID and away from edge application server 111 A), both the application domain 110 and the network domain 120 can better prepare resources at the edge cloud set including the edge application server 11 ID and the network function 124D, instead of at the edge cloud set including the edge application server 111 A and the network function 124 A.

[0080] Thus, knowledge of the motion vector 320 and/or an episodic mobility route of the UE 300 can influence related service admission policies. For example, if network function 124D is overloaded, the network domain 120 may be controlled to not accept new sessions that may use the network function 124D in order to maintain or guarantee the service quality of the incoming session of the UE 300. As a further example, the network domain 120 may be controlled to quickly reduce resources at the network function 124 A, such as by powering units down. Such resource coordination can be further based at least in part on multiple mobility routes of multiple UEs. For example, resource coordination may be prompted by hundreds of UEs 300 moving along an episodic mobility route (e.g., on mass transport such as a train, plane, boat).

[0081] While Figure 3 illustrates one motion vector 320, it is to be recognized that, in accordance with various embodiments, a prediction of an episodic mobility route extends the information provided by one motion vector 320 to then provide farther reaching (temporally) resource coordination. Figure 4 illustrates an example episodic mobility route 400 predicted for a UE 300. Generally, an episodic mobility route 400 may include various waypoints 402; in the illustrated embodiment, the episodic mobility route 400 includes five example waypoints 402A- E. In various embodiments, the episodic mobility route 400 comprehensively describes a movement pattern of the UE 300 from start to finish, and thus, the episodic mobility route 400 may include a starting waypoint (e.g., waypoint 402A) and a destination waypoint (e.g., waypoint 402E). In various embodiments, the episodic mobility route 400 effectively provides more information than one motion vector 320, and the episodic mobility route 400 may span or wind in different directions and different speeds at different points and times. Thus, an episodic mobility route 400 can be understood and/or represented as a plurality of individual discrete motion vectors 320, each of which may have a different direction and speed.

[0082] As described, various embodiments provide for resource coordination within and between an application domain 110 and a network domain 120. In Figure 5, a functional system architecture 500 for cross-domain resource coordination between different entities (or functions) associated with an application domain 110 and a network domain 120 is provided. It may be recognized that, while the functional system architecture 500 provided by Figure 5 is overlaid and supplements a 3GPP 5G system architecture, various concepts described herein relating to cross-domain resource coordination can be applied to various different network architectures having separate, distinct, and independent domains.

[0083] The functional system architecture 500 illustrates components and entities of each of the application domain 110 and the network domain 120, as well as the interactions within and across each domain in order to provide cross-domain resource coordination. As illustrated, the application domain 110 comprises and can be controlled at least in part by an application manager 501, or an application function (AF). Generally, the application manager 501 is in communication with one or more application client instances 502 that are being executed or run by UEs 300, and this communication may be referred to as the ail interface. Due to movement of a UE 300, a location associated with an application client instance 502 on the UE 300 can be variable. Meanwhile, the application manager 501 is in communication with a plurality of edge application server (EAS) instances 503 being executed or run by edge application servers 111, and the edge application servers 111 are spatially distributed throughout edge cloud regions. [0084] Generally, the application manager 501 may be configured to receive a prediction of an episodic mobility route of the UE 300 (and by extension, one or more application client instances 502 on the UE 300), and according to the prediction, identify a subset of the plurality of EAS instances 503 that can be used to serve the application client instances 502 according to various criteria. In various embodiments, the application manager 501 may provide the identified subset as an ordered list of EAS instances 503 to an application client instance 502 and/or the network domain 120 for selection, and a data session can be established with (or migrated to) a selected EAS instance 503. A particular EAS instance 503 may be specifically selected to satisfy various latency and/or quality thresholds for the data session; that is, it may be determined that the desired low latency exists between the application client instance 502 and the particular EAS instance 503.

[0085] As shown within the functional system architecture 500 in a 5G-specific context, the network domain 120 includes various network functions, including but not limited to a user plane function 511, a network exposure function (NEF) 512, an NWDAF 513, an access management function (AMF) 514, and a session management function (SMF) 515. Generally, various operations are performed within the network domain 120 to establish and migrate data sessions between an application client instance 502 and selected EAS instances 503 such that the data sessions have low-latency communications. For example, the NWDAF 513 may be configured to predict an episodic mobility route of the UE 300, and the prediction can then be provided by the network exposure function 512 to the application domain 110 (e.g., the application manager 501) via the N33 interface, in some example embodiments. While the functional system architecture 500 includes the network exposure function 512, which can enable interactions and communications between the application manager 501 and various network functions in the network domain 120, it will be understood that, in various example embodiments, the application manager 501 can be configured to interact and communicate directly with various network functions, such as the NWDAF 513. For example, various network functions can directly interact with the application manager 501 if the application manager 501 is trusted by the network domain 120, or if the application manager 501 is within a trust domain of a mobile network operator of the network domain 120. Also, generally, multiple NEFs 512 of different types can be included in the functional system architecture 500 for various purposes, such as monitoring, provisioning, policy and charging, analytics reporting, and/or the like. [0086] In various embodiments, various network functions of the network domain 120, such as the SMF 515 are configured to select a particular EAS instance 503 from an ordered list of EAS instances 503 provided by the application manager 501 (e.g., directly, via NEF 512) based at least in part on determining quality metrics including latency measurements for each of the ordered list of EAS instances 503. For example, the SMF 515 may select a particular EAS instance 503 having the best quality metrics and/or quality metrics that satisfy a threshold. As shown in Figure 5, an EAS instance 503 may be accessed via the network domain 120 via an anchor point 520 and a data network 530. Thus, in various embodiments, various network functions of the network domain 120 are configured to select a particular anchor point 520 in association with the selection of the particular EAS instance 503, and the network domain 120 may specifically select a particular anchor point 520 to minimize latency across the N6 interface (from anchor point 520 through data network 530 and/or to EAS instance 503). As such, in various embodiments, the network domain 120 and network functions thereof are configured to select EAS instances 503 and anchor points 520 in order to establish and migrate data sessions for an application client instance 502 that may be moving along an episodic mobility pattern or route.

[0087] Referring now to Figure 6, an example apparatus 600 is provided. In various embodiments, the apparatus 600 is configured for performing various operations for crossdomain resource coordination, such as the example operations described above. In some examples, the apparatus 600 may be embodied by various network functions within the network domain 120, including the network exposure function 512, the NWDAF 513, the AMF 514, the SMF 515, and/or the like. In some further examples, the apparatus 600 may be embodied by the application manager 501 within the application domain 110. In any regard, the apparatus 600 may be configured for various operations including predicting an episodic mobility route 400 of a UE 300, determining sets and ordered lists of EAS instances 503 according to the episodic mobility route 400, determining quality metrics for EAS instances 503, selecting particular EAS instances 503 and anchor points 520, establishing and/or migrating data sessions via selected anchor points 520 and EAS instances 503, and/or the like, in various embodiments.

[0088] The apparatus 600 may include processor 602, memory 604, and communications circuitry 606. The apparatus 600 may be configured to execute the operations described herein. Although these components are described with respect to functional limitations, it should be understood that the particular implementations necessarily include the use of particular hardware. It should also be understood that certain of these components may include similar or common hardware. For example, two sets of circuitries may both leverage use of the same processor, network interface, storage medium, or the like to perform their associated functions, such that duplicate hardware is not required for each set of circuitries.

[0089] In some embodiments, the processor 602 (and/or co-processor or any other processing circuitry assisting or otherwise associated with the processor) may be in communication with the memory 604 via a bus for passing information among components of the apparatus. The memory 604 is non-transitory and may include, for example, one or more volatile and/or non-volatile memories. In other words, for example, the memory 604 may be an electronic storage device (e.g., a computer-readable storage medium). The memory 604 may be configured to store information, data, content, applications, instructions, or the like for enabling the apparatus to carry out various functions in accordance with an example embodiment disclosed herein.

[0090] The processor 602 may be embodied in a number of different ways and may, for example, include one or more processing devices configured to perform independently. In some non-limiting embodiments, the processor 602 may include one or more processors configured in tandem via a bus to enable independent execution of instructions, pipelining, and/or multithreading. The use of the term “processor” may be understood to include a single core processor, a multi-core processor, multiple processors internal to the apparatus, and/or remote or “cloud” processors.

[0091] In some embodiments, the processor 602 may be configured to execute instructions stored in the memory 604 and/or circuitry otherwise accessible to the processor 602, such as but not limited to instructions for predicting UE episodic mobility routes and selecting EAS instances 503 and/or anchor points 520. In some embodiments, the processor 602 may be configured to execute hard-coded functionalities. As such, whether configured by hardware or software methods, or by a combination thereof, the processor 602 may represent an entity (e.g., physically embodied in circuitry) capable of performing operations according to an embodiment disclosed herein while configured accordingly. Alternatively, as another example, when the processor 602 is embodied as an executor of software instructions, the instructions may specifically configure the processor 602 to perform the algorithms and/or operations described herein when the instructions are executed. [0092] In some embodiments, the apparatus 600 may include input/output circuitry that may, in turn, be in communication with processor 602 to provide output to a user and/or other entity and, in some embodiments, to receive an indication of an input. The input/output circuitry may comprise a user interface and may include a display, and may comprise a web user interface, a mobile application, a query-initiating computing device, a kiosk, or the like. In some embodiments, the input/output circuitry may also include a keyboard, a mouse, a joystick, a touch screen, touch areas, soft keys, a microphone, a speaker, or other input/output mechanisms. The processor and/or user interface circuitry comprising the processor may be configured to control one or more functions of one or more user interface elements through computer program instructions (e.g., software and/or firmware) stored on a memory accessible to the processor (e.g., memory 604, and/or the like).

[0093] The communications circuitry 606 may be any means such as a device or circuitry embodied in either hardware or a combination of hardware and software that is configured to receive and/or transmit data from/to a network and/or any other device, circuitry, or module in communication with the apparatus 600. In this regard, the communications circuitry 606 may include, for example, a network interface for enabling communications with a wired or wireless communication network. For example, the communications circuitry 606 may include one or more network interface cards, antennae, buses, switches, routers, modems, and supporting hardware and/or software, or any other device suitable for enabling communications via a network. Additionally, or alternatively, the communications circuitry 606 may include the circuitry for interacting with the antenna/antennae to cause transmission of signals via the antenna/antennae or to handle receipt of signals received via the antenna/antennae.

[0094] It is also noted that all or some of the information discussed herein can be based on data that is received, generated and/or maintained by one or more components of apparatus 600. In some embodiments, one or more external systems (such as a remote cloud computing and/or data storage system) may also be leveraged to provide at least some of the functionality discussed herein.

[0095] Referring now to Figure 7A, a flowchart illustrating various example operations for cross-domain resource coordination using predicted episodic mobility routes is provided. In various embodiments, at least some of the example operations shown in Figure 7A can be performed in order to establish and initiate a new application data session and in order to migrate an existing application data session to different EAS instances 503. The example operations illustrated in Figure 7A can be performed by apparatus 600, and the apparatus 600 can specifically be embodied by the application manager 501, an application function, and/or other similar components within the application domain 110.

[0096] An example scenario in which a new data session is to be established is described first. At operation 701, the apparatus 600 includes means, such as processor 602, memory 604, communications circuitry 606, and/or the like, for receiving an indication of session establishment from a UE 300. The indication from the UE 300 essentially represents a request by the UE 300 to start an application data session, or a data session for a given application. As one non-limiting example, a user may request a movie for streaming via a streaming application, and may do so from a different server location with additional connection setups. That is, in some examples, the indication from the UE 300 at operation 701 may be a request for a new data session.

[0097] Specifically, the indication or request may originate from one or more application client instances 502 of the given application being run or residing on the UE 300. In various embodiments, the indication or request for session establishment from an application client instance 502 of the given application is communicated via standard network domain connection procedures such that the apparatus 600 (e.g., embodied by application manager 501) receives the indication/request. For instance, in 5G-specific contexts, the indication/request for a new application data session may involve network slicing entities and procedures in accordance with 3 GPP specifications.

[0098] At operation 702, the apparatus 600 includes means, such as processor 602, memory 604, communications circuitry 606, and/or the like, for obtaining a prediction of an episodic mobility route of the UE 300 from the network domain. In various embodiments, the apparatus 600, embodied by the application manager 501, may obtain the prediction via the NEF 512, directly from NWDAF 513, and/or the like. In various embodiments, the prediction of the episodic mobility route comprises a plurality of waypoints and/or one or more motion vectors, as previously described. Referring back to Figure 4, for example, an episodic mobility route 400 may wind through various waypoints in different orientations. In various example embodiments, the episodic mobility route 400 (and the prediction thereof) spans from a starting waypoint to a destination waypoint, thereby providing more complete information than one motion vector of the UE 300. In some example embodiments, the prediction of the episodic mobility route 400 may be probabilistic, and each waypoint may be associated with a probability of the UE 300 passing through it.

[0099] At operation 703, the apparatus 600 includes means, such as processor 602, memory 604, and/or the like, for identifying one or more edge zones spanned by the episodic mobility route. In various embodiments, an edge zone may refer to a geographic boundary within which application client instances 502 and EAS instances 503 can reside to satisfy quality constraints or requirements (e.g., latency requirements) of the application. Accordingly, if an application client instance 502 and an EAS instance 503 are within the same edge zone, it can be expected that quality requirements of the application will be met. In 5G-specific contexts, edge zones may be identified using data network access identifiers (DNAIs). In various embodiments, the apparatus 600 may at least identify an edge zone within which the UE 300 is presently positioned.

[00100] At operation 704, the apparatus 600 includes means, such as processor 602, memory 604, and/or the like, for determining a set of EAS instances 503 for each edge zone. In various embodiments, the set of EAS instances 503 may specifically be an ordered list of EAS instances 503, sorted with respect to various criteria. For instance, EAS instances 503 may be determined, identified, and ordered based at least in part on geographic location within the edge zone, latency requirements of the application, cost of edge cloud resources, availability of edge cloud resources, current load levels, and/or the like. In various embodiments, the apparatus 600 particularly determines existing EAS instances 503, instead of edge application servers 111 that are not yet instantiated. In doing so, an EAS instance 503 can be tested and verified with respect to quality metrics before making a decision for session establishment. Thus, for performance concerns, it can be understood that the EAS instances 503 determined and identified by the apparatus 600 are already active and serving other application client instances 502, or have been instantiated just before. Effectively then, the apparatus 600 is configured to, at operation 704, identify a prioritized list of candidate EAS instances 503 that can be used for session establishment.

[00101] At operation 705, the apparatus 600 includes means, such as processor 602, memory 604, communications circuitry 606, and/or the like, for providing one or more sets of EAS instances 503 to the UE 300 and/or the network domain 120. In this example scenario of session establishment, at least a set of EAS instances 503 determined based at least in part on the UE’s present location is provided to the UE 300. Other auxiliary information associated with the EAS instances 503 can also be provided to the UE 300 and/or the network domain 120, as will be described later in the present disclosure. For example, service quality attributes of the EAS instances 503 can be provided to the UE 300 in session establishment scenarios.

[00102] In session establishment scenarios, the UE 300 may identify the provided EAS instances 503 in a session establishment request to the network domain 120. For example, when establishing an LL+ application data session, the application client instance 502 includes descriptive information for each of the provided one or more sets of EAS instances 503 in a network domain data establishment request. In a 5G-specific context, such a request by the UE 300 may include a UE protocol data unit (PDU) session establishment request in accordance with 3GPP specifications. As a result, the network domain 120 obtains a prioritized list of candidate EAS instances 503 from the application domain 110 via the UE 300 during a session establishment request from the UE 300. In various embodiments, the network domain 120 is configured to store the prioritized list of EAS instances 503 in a policy control function (PCF) and may use the prioritized list as part of a UE route selection policy (URSP). In some alternative embodiments, the apparatus 600 may be configured to provide the one or more sets of EAS instances 503 (e.g., candidate EAS instances) directly to the network domain 120.

[00103] At operation 706, the apparatus 600 include means, such as processor 602, memory 604, communications circuitry 606, and/or the like, for enabling establishment of a data session between the UE 300 and a selected EAS instance. Apparatus 600 may do so in response to receiving an indication of the network domain’s selection of a particular EAS instance from the provided set(s) of candidate EAS instances. In various embodiments, the particular EAS instance may be selected generally according to its quality metrics, and selection of EAS instances 503 is described below in the context of example operations performed within and by the network domain 120. The selected EAS instance is then the EAS instance 503 that the network domain 120 has selected for establishment of a new data session. Accordingly, the apparatus 600 as embodied by the application manager 501 may be configured to prepare the selected EAS instance for establishment of a new data session. For instance, the apparatus 600 may identify the UE 300 and cause the selected EAS instance to allocate resources in preparation for the new data session to be established. [00104] At operation 707, the apparatus 600 includes means, such as processor 602, memory 604, communications circuitry 606, and/or the like, for initiating an automatic subscription to episodic mobility prediction updates for the UE 300 from the network domain 120. As the UE 300 continues to travel, whether along the initially predicted episodic mobility route or otherwise, the selected EAS instance may no longer be optimal for a data session for the UE 300 (and specifically the application client instance 502 residing on the UE 300). Accordingly, in various embodiments, the apparatus 600 subscribes to prediction updates for the UE’s episodic mobility route so that the data session can be migrated to more optimal EAS instances in accordance with the UE’s updated location.

[00105] Accordingly, following the session establishment scenario described hereto, at least some of the example operations illustrated by Figure 7A can be repeated for a session migration scenario. In various embodiments, following operation 702, the apparatus 600 may obtain a subsequent prediction of an episodic mobility route of the UE from the network domain 120 at operation 702. In accordance with operations 703 and 704, the apparatus 600 again determines and identifies a new list of candidate EAS instances based at least in part on the prediction of the episodic mobility route. At operation 705, the apparatus 600 may provide the new candidate EAS instances to the network domain 120 directly. That is, in various embodiments, session migration may require minimal involvement from the UE 300 directly, and may advantageously be handled by the application manager 501 and the network domain 120. Then, in accordance with operation 706, the apparatus 600 (again, embodied by application manager 501 in various examples) may receive an indication a particular EAS instance selected from the new candidate EAS instances by the network domain 120 and may prepare the particular EAS instance accordingly in preparation for the previously established data session to be migrated to the particular EAS instance.

[00106] As described, the example operations of Figure 7A may be performed by the application manager 501, an application function, and/or other similar entities within the application domain 110, in various embodiments in order to first enable establishment of a new application data session and to subsequently enable migration of the existing application data session to a different EAS instance and through a different anchor point if necessary. Referring next to Figure 7B, another flowchart is provided, the flowchart illustrating further example operations for cross-domain resource coordination using predicted episodic mobility routes. In various embodiments, similar to Figure 7A, at least some of the example operations of Figure 7B can be performed and initiate a new application data session and in order to migrate an existing application data session to a different EAS instance 503. The example operations illustrated in Figure 7B can be performed by an apparatus 600, and specifically can be performed by various network functions of the network domain 120 embodying the apparatus 600. For example, network functions including an NEF 512, an NWDAF 513, an AMF 514, an SMF 515, and/or the like may be configured to perform one or more of the example operations illustrated in Figure 7B.

[00107] As before, the example scenario in which a new data session is to be established is described first. At operation 711, the apparatus 600 includes means, such as processor 602, memory 604, communications circuitry 606, and/or the like, for predicting an episodic mobility route of a UE 300. In various embodiments, one or more network functions of the network domain 120 embodying the apparatus 600 are configured to predict the episodic mobility route of the UE 300 in response to a request from the application manager 501. For example, at operation 702, the application manager 501 may transmit a request to one or more network functions of the network domain 120 for an episodic mobility route prediction.

[00108] In various example embodiments in a 5G-specific context, the apparatus 600 may be embodied at least by the NWDAF 513, which is configured to predict an episodic mobility route of the UE 300. In various embodiments, the episodic mobility route of the UE 300 is predicted according to historical behavior and movements of the UE 300 that may have been recorded and stored (e.g., given explicit consent from a user of the UE 300, in various examples). Using detailed analytic methods, historical behavior, and other factors such as the UE’s present location, the day, the time of day, and/or the like, the apparatus 600 can then predict an episodic mobility route of the UE 300. In various embodiments, the predicted episodic mobility route comprises a plurality of waypoints and represents a UE’s trip in its entirety, in various examples. The plurality of waypoints may be associated with probabilities and can correspond to geographic locations where the network domain 120 has deployed resources (e.g., cell towers, edge cloud platforms) that may be used for support of LL+ applications. In various example embodiments, the probabilities associated with the waypoints may vary over time and may depend on accuracy of information. For example, the probability of the UE 300 passing by downstream or subsequent waypoints of the predicted episodic mobility route may dynamically increase according to whether the UE 300 continues along the predicted episodic mobility route. [00109] As discussed, since the edge cloud resources have many cost and availability constraints compared to central cloud locations, the prediction of the episodic mobility route of the UE 300 that describes which geographic regions an application client instance 502 will visit is valuable information to optimize resource scheduling. That is, episodic mobility predictions provide enhanced capability for resource optimization, and without the availability of episodic predictions, the resource scheduling may only rely upon short-sighted and limited information, such as a single vector of the UE’s current velocity.

[00110] At operation 712, the apparatus 600 includes means, such as processor 602, memory 604, communications circuitry 606, and/or the like, for providing the prediction of the episodic mobility route of the UE 300 to the application domain 110. As previously discussed, the apparatus 600 may be embodied by the NEF 512 and/or the NWDAF 513 in order to provide the prediction of the episodic mobility route to the application domain 110, or the application manager 501 specifically.

[00111] At operation 713, the apparatus 600 includes means, such as processor 602, memory 604, communications circuitry 606, and/or the like, for obtaining an indication of a plurality of EAS instances 503. Specifically, the indicated EAS instances are candidate EAS instances determined by the application domain 110 to be relevant and prioritized for establishment of a new data session. For example, one or more network functions of the network domain 120 embodying the apparatus 600 receive the prioritized list of candidate EAS instances in response to providing the episodic mobility route prediction to the application domain 110, and the candidate EAS instances are determined by the application domain 110 according to their respective locations in relation to the episodic mobility route prediction. For instance, the candidate EAS instances are positioned in edge zones spanned by the predicted episodic mobility route.

[00112] As previously described, this prioritized list of candidate EAS instances may be received from the UE 300 as part of a session establishment request from the UE 300. For example, the application domain 110 may have previously indicated to the UE 300 the candidate EAS instances so that the UE 300 may request (to the network domain 120) establishment of a new data session with any of the candidate EAS instances. Thus, at operation 713, one or more network functions embodying the apparatus 600 may obtain the indication of candidate EAS instances from the UE 300. In alternative example embodiments and/or example scenarios, the network domain 120 may obtain the prioritized list of candidate EAS instances from the application domain 110 directly. For instance, in session migration scenarios, the network domain 120 receives indications of candidate EAS instances from the application domain 110 to minimize involvement of the UE 300 in session migration.

[00113] In any regard, the indication of the candidate EAS instances received from the application domain 110 comprises further information associated with the candidate EAS instances. Figure 8 illustrates an example message structure for the indication of the candidate EAS instances as provided by the application manager 501; in various embodiments, the application manager 501 may generate the indication of the candidate EAS instances in accordance with the message structure shown in Figure 8 and subsequently provide said indication to the one or more network functions of the network domain 120.

[00114] In Figure 8, the message structure includes an EAS instance selection request 801, which represents a request from the application domain 110 (e.g., the application manager 501) for the network domain 120 (e.g., and network functions thereof) to select one of the candidate EAS instances described within the message structure. In the illustrated embodiment, the EAS instance selection request 801 includes various service attributes 802 and an EAS instance set 803. The EAS instance set 803 is configured to describe a plurality of EAS instances 503 and includes a sounding service description (SSD) 810 for each EAS instance 503 (determined to be a candidate EAS instance by the application manager 501 at a prior time).

[00115] In various embodiments, an SSD 810 for a respective candidate EAS instance conveys at least a sounding service node (SSN) 811, a sounding service method (SSM) 812, and sounding service attributes (SSA) 813 for the respective candidate EAS instance. The SSN 811 may serve as an identifier and/or name for the respective candidate EAS instance. For example, the SSN 811 can be a URL, an IP address, a specific identifier recognized by the network domain 120, and/or the like. Generally, the SSD 810 for a respective candidate EAS instance enables the application domain 110 to specify how quality metrics are determined for the respective candidate EAS instance. As a non-limiting example, an SSM 812 may be the HTTP protocol and SSA may identify port 80. As another non-limiting example, the SSM 812 may include a script, and the SSA 813 may include specific code of the script to be executed by the network domain 120. In yet another non-limiting example, the SSM 812 may be “ping” and the SSA 813 may be “not applicable” for the ping method. Therefore, using SSDs 810 for candidate EAS instances, determination of quality metrics for the candidate EAS instance by the network domain 120 is enabled as specified by the application domain 110.

[00116] Returning to Figure 7B, at operation 714, the apparatus 600 includes means, such as processor 602, memory 604, communications circuitry 606, and/or the like, for determining quality metrics for each EAS instance 503 in combination with one or more anchor points 520. In various embodiments, one or more network functions of the network domain 120 embodying the apparatus 600 are configured to determine quality metrics for an EAS instance 503 in accordance with the SSD 810 for the EAS instance 503 as indicated by the application domain 110.

[00117] In various embodiments, the quality metrics determined for an EAS instance 503 include various latency metrics, such that EAS instances 503 can be selected based at least in part on satisfying latency requirements of LL+ applications. Figure 9A demonstrates various latencies relevant to data sessions between application endpoints, specifically an application client instance 502 residing on a UE 300 and an EAS instance 503. An end-to-end application latency between the application client instance 502 and the EAS instance 503 may be referred to as an application target latency 902 and can comprise at least two latency components. First, as illustrated, a latency may exist between the application client instance 502 and an anchor point 520, and another latency may exist between the anchor point 520 and the EAS instance 503. The latency between the application client instance 502 and the anchor point 520 may be referred to as a network domain latency 904, while the latency between the anchor point 520 and the EAS instance 503 may be referred to as a data network latency 906. As a result, the application target latency 902 may be the sum of the network domain latency 904 and the data network latency 906.

[00118] Each of the network domain latency 904 and the data network latency 906 may be based at least in part on various factors, including geographic distance (or the cumulative length of the network links that cover that distance), traffic load on each network link, a number of network functions traversed by the end-to-end connection and the logical topology that they form, and the traffic load handled by each network function. [00119] In order to satisfy quality and latency requirements of an LL+ application, the network domain 120 must guarantee that a set percentile of the end-to-end latency samples collected for application packets do not exceed the application target latency 902. In various embodiments, the network domain 120 can determine the network domain latency 904 according to the placement of the anchor point 520, while determining the value of the data network latency 906 may include more technical challenges as the data path is not within the network domain 120.

[00120] Figure 9B provides a diagram generally demonstrating determination of quality metrics for an EAS instance 503, including determination of a data network latency 906. As discussed, determination of quality metrics for an EAS instance 503 is performed in accordance with an SSD 810 for the EAS instance 503 as specified and provided by the application manager 501. In the illustrated embodiment, the SSD 810 is received at an NEF 512 and may be distributed to various network functions throughout the network domain 120, such as a policy control function (PCF) and/or an SMF 515. Then, as illustrated in Figure 9B, the network domain 120 may determine quality metrics through executing an SSM 812 with the anchor point 520 and SSAs 813 of the EAS instance 503 identified by the SSN 811 of the SSD 810. In various embodiments, execution of the SSM 812 can result in determination of latency metrics including data network latencies 906 for an EAS instance 503. That is, in some examples, definition of an SSD 810 and/or at least an SSM 812 enables determination of data network latencies 906.

[00121] At operation 715, the apparatus 600 includes means, such as processor 602, memory 604, communications circuitry 606, and/or the like, for selecting a particular EAS instance and a particular anchor point. Specifically, network functions of the network domain 120 select a particular EAS instance from the candidate EAS instances and an anchor point 520 that result in quality metrics that satisfy various requirements or thresholds. In various embodiments, data network latency 906 may be used as at least one quality metric by which the particular EAS instance and anchor point are selected. In various embodiments, other factors can be additionally or alternatively considered, including past measurements, network management configurations, policy rules, and/or the like. In some example instances, the network domain 120 may determine that none of the candidate EAS instances are satisfactory with respect to their quality metrics. In such instances, the network domain 120 may be configured to provide an error describing a service violation to the application client instance 502 and/or the application manager 501, in various embodiments.

[00122] At operation 716, the apparatus 600 includes means, such as processor 602, memory 604, communications circuitry 606, and/or the like, for establishing a new data session between the UE 300 and the selected EAS instance via the selected anchor point. In various embodiments, network functions of the network domain 120 embodying the apparatus 600 may first indicate at least the selection of the particular EAS instance to the application domain 110, and the application domain 110 may enable a data session to be established with the particular EAS instance (e.g., at operation 706). Accordingly, with selection of a particular EAS instance and a particular anchor point with respect to quality metrics (e.g., a data network latency 906), the established data session may have improved low latency.

[00123] As previously described, the application domain 110 may have subscribed to prediction updates. Accordingly, in various embodiments, network functions of the network domain 120 may continue to predict episodic mobility routes of the UE at operation 711. In various embodiments, subsequent predictions may be made on a periodic, intermittent, and/or continuous basis and/or made in response to significant changes in location of the UE 300. Given that the application domain 110 has subscribed to prediction updates, these subsequent predictions may be provided by the network domain 120 to the application domain 110 after a new data session has been established.

[00124] In session migration scenarios occurring after session establishment, the network domain 120 may again obtain new candidate EAS instances that are determined by the application domain 110 based at least in part on subsequent predictions of the UE’s episodic mobility route, in accordance with operation 713. For instance, the new candidate EAS instances may be positioned within an edge zone of a subsequent waypoint along a subsequent prediction of the UE’s episodic mobility route. Here, with a data session being established, an indication of the new candidate EAS instances may be obtained directly from the application domain 110 (e.g., in accordance with the message structure of Figure 8), rather than via the UE 300. As such, involvement of the UE 300 in session migration is advantageously minimized. With the new candidate EAS instance, a new selection of a particular EAS instance may be performed according to operations 714 and 715. In some example scenarios, a new anchor point may be additionally or alternatively selected according to the quality metrics (e.g., a data network latency 906). Then, at operation 716, the established data session may be migrated to the newly selected EAS instance, migrated to be via the newly selected anchor point, or both. In doing so, the previously established data flow may be ended and released, in various embodiments.

[00125] Thus, example operations that may be performed within the application domain 110 by the application manager 501, for example, have been described (and illustrated in Fig. 7A) along with other example operations that may be performed within the network domain 120 (illustrated in Fig. 7B) by various network functions, such as NEF 512, NWDAF 513, AMF 514, SMF 515, PCFs, UPFs, and/or the like. Figures 10A and 10B provide sequence diagrams that describe various example operations performed by a UE 300 and within an application domain 110 and a network domain 120.

[00126] First, Figure 10A describes various example operations and interactions for establishing a new data session for a given application for a UE 300. To initiate establishment of a data session for an LL+ application, the UE 300 and an application client instance 502 first provide a session establishment request to an application manager 501. In response, the application manager 501 may request a prediction of an episodic mobility route of the UE 300 from the network domain 120 (or functions thereof such as an NWDAF 513), and the network domain 120 provides a prediction of UE’s episodic mobility route to the application manager 501.

[00127] According to the predicted episodic mobility route, the application manager 501 may determine or identify various candidate EAS instances, for example, EAS instances that are located within one or more edge zones spanned by the predicted episodic mobility route. As discussed, EAS instances may be identified according to factors beyond a geographical location, such as factors relating to network topology and load. In a similar manner, edge zones can be defined based at least in part on such factors beyond geographical locations, such that candidate EAS instances can be easily identified when positioned within an edge zone. Upon determination of candidate EAS instances, the application manager 501 may prepare the EAS instances 503 for potential session establishment. Additionally, the application manager 501 may provide an indication of the candidate EAS instances to the UE 300 along with further information such as an SSD 810 for each candidate EAS instance. In some example embodiments, the application manager 501 additionally or alternatively provides the indication of candidate EAS instances to the network domain 120 during session establishment (not explicitly illustrated). [00128] With the indication of the candidate EAS instances, the UE 300 may request a data session initiation to the network domain 120, and in the request, the UE 300 may include the indication of the candidate EAS instances. As a result, the network domain 120 obtains the set of candidate EAS instances, whether via the UE 300 (as in the illustrated embodiment) or directly from the application manager 501 in some examples (not explicitly illustrated).

[00129] To satisfy the request for data session initiation by the UE 300, the network domain 120 determines which of the candidate EAS instances to use and further, a particular anchor point 520. Accordingly, the network domain 120 determines quality metrics for the candidate EAS instances and for combinations of the candidate EAS instances with anchor points 520. In particular, a data network latency 906 is determined for each combination of candidate EAS instance and anchor point 520.

[00130] With the quality metrics (e.g., a data network latency 906), the network domain 120 selects a particular EAS instance and a particular anchor point for the data session to be established. It may be recognized here that, as the application manager 501 has already prepared each candidate EAS instance at an earlier point, the network domain 120 may not need to communicate the selection of the particular EAS instance to the application manager 501. In alternative embodiments, however, the application manager 501 may prepare one particular and selected EAS instance responsive to being informed of the selection by the network domain 120. [00131] Following establishment of the data session with the selected EAS instance, selected anchor point, and UE 300, the application manager 501 may subscribe to prediction updates from the network domain 120, which may enable data sessions to be migrated upon changes in predicted episodic mobility routes of the UE 300. In this regard, Figure 10B describes example operations and interactions for session migration.

[00132] As illustrated in Figure 10B, a data session has been established and exists between the UE 300 and a first EAS instance 503 A via a first anchor point 520A. With the application manager 501 being subscribed to prediction updates, the network domain 120 may provide an updated prediction of the UE’s episodic mobility route. In various embodiments, the updated prediction may be provided on a periodic basis, responsive to discrete behaviors of the UE 100, responsive to user input, responsive to deviations of the UE 100 from a previously predicted episodic mobility route that exceed thresholds, and/or the like. [00133] With an updated prediction of the UE’s episodic mobility route, the application manager 501 may determine a new set of candidate EAS instances and indicate the new set to the network domain with sounding service information (e.g., in accordance with the message structure of Figure 8). As previously described, the new candidate EAS instances may be indicated directly to the network domain 120 by the application manager 501 and may not involve the UE 300. As such, communication by the UE 300 is appropriately minimized, and session migration can be efficiently handled.

[00134] With the indication of the new candidate EAS instances, the network domain can determine quality metrics of the new candidate EAS instances in combination with anchor points 520. The new candidate EAS instances may or may not include the first EAS instance 503 A with which a data session has already been established. Similarly, new and/or previously evaluated anchor points can again be evaluated in combination with the new candidate EAS instances. With determination of quality metrics (e.g., data network latency 906), the network domain 120 is configured to either select a new EAS instance 503B to use with the first anchor point 520A, a new anchor point 520B to use with the first EAS instance 503 A, or a new EAS instance 503B to use with a new anchor point 520B. The network domain 120 can also determine that none of the candidate EAS instances would provide a significantly improved data session over the existing data session.

[00135] The illustrated embodiment in particular illustrates the network domain 120 selecting a new EAS instance 503B and a new anchor point 520B. With this selection, the network domain 120 then migrates the data session to the new EAS instance 503B through the new anchor point 520B. Meanwhile, the data flow for the existing data session can be ended and released. Accordingly, with the session migration, satisfaction of low-latency requirements and/or other quality constraints can continue with UE episodic mobility.

[00136] As described within the present disclosure, various embodiments can be applied to example scenarios in which components of the application domain 110 (e.g., application client instances 502, application manager 501, EAS instances 503) are not part of a mobile network operator’s administrative domain, or the network domain 120. However, in various embodiments, a sufficient trust level is established between the application domain 110 and the network domain 120 such that the application manager 501 is authorized to execute various NEF application programming interface (API) procedures in performing various example operations described herein. In some alternative example embodiments, the application manager 501 may directly communicate and interact with various network functions of the network domain 120, such as an NWDAF 513, a PCF, and an SMF 515.

[00137] In general, there may be multiple application client instances 502 that are running or reside on one UE 300. There may also be application client instances 502 that may not explicitly be running on the UE 300 but on another platform connected to the UE 300. In this case, UE episodic mobility still remains important to cross-domain resource coordination because the UE remains in the data path between an application client instance 502 and an EAS instance 503. [00138] Various embodiments described herein for resource coordination and example interactions between one network domain 120 and an application domain 110 can be extended to situations in which an application domain 110 interacts with multiple network domains 120, such as roaming situations.

[00139] While the present disclosure discusses various methods, processes, and operations in a procedural form, PCFs of the network domain 120 may be heavily involved for execution of such methods, processes, and operations in a more declarative form. With policy involvement, information can be stored at and interpreted by a policy manager function, or a PCF in a 5G- specific context. For example, a selection of a particular candidate EAS instance by an SMF 515 can be stored at a PCF, and the selection can be later used as part of UE route selection policies for PDU session establishment.

[00140] Overall, various embodiments described herein provide cross-domain resource coordination using predictions of UE episodic mobility routes, and in doing so, provide various technical advantages. Various embodiments enable the realization of applications with low latency and with strictly independent lifecycle control, in various examples, despite their strong dependency on the network domain for performance. Further, various embodiments enable reduction of edge cloud operation cost through more efficient orchestration of edge cloud resources.

[00141] Figures 7A and 7B illustrate flowcharts depicting operations according to an example embodiment of the present disclosure, and Figures 10A and 10B illustrate sequence diagrams depicting operations and interactions. It will be understood that each block of the flowcharts and combination of blocks in the flowcharts may be implemented by various means, such as hardware, firmware, processor, circuitry, and/or other communication devices associated with execution of software including one or more computer program instructions. For example, one or more of the procedures or operations described above may be embodied by computer program instructions. In this regard, the computer program instructions which embody the procedures or operations described above may be stored by a memory 604 of an apparatus (e.g., apparatus 600,) employing an embodiment of the present invention and executed by a processor 602. As will be appreciated, any such computer program instructions may be loaded onto a computer or other programmable apparatus (for example, hardware) to produce a machine, such that the resulting computer or other programmable apparatus implements the functions specified in the flowchart blocks. These computer program instructions may also be stored in a computer- readable memory that may direct a computer or other programmable apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture the execution of which implements the function specified in the flowchart blocks. The computer program instructions may also be loaded onto a computer or other programmable apparatus to cause a series of operations to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide operations for implementing the functions specified in the flowchart blocks.

[00142] Accordingly, blocks of the flowcharts support combinations of means for performing the specified functions and combinations of operations for performing the specified functions for performing the specified functions. It will also be understood that one or more blocks of the flowcharts, and combinations of blocks in the flowcharts, can be implemented by special purpose hardware-based computer systems which perform the specified functions, or combinations of special purpose hardware and computer instructions.

[00143] Many modifications and other embodiments of the inventions set forth herein will come to mind to one skilled in the art to which these inventions pertain having the benefit of the teachings presented in the foregoing descriptions and the associated drawings. Therefore, it is to be understood that the inventions are not to be limited to the specific embodiments disclosed and that modifications and other embodiments are intended to be included within the scope of the appended claims.

[00144] Moreover, although the foregoing descriptions and the associated drawings describe example embodiments in the context of certain example combinations of elements and/or functions, it should be appreciated that different combinations of elements and/or functions may be provided by alternative embodiments without departing from the scope of the appended claims. In this regard, for example, different combinations of elements and/or functions than those explicitly described above are also contemplated as may be set forth in some of the appended claims. Although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.

Claims

THAT WHICH IS CLAIMED:

1. An apparatus comprising: at least one processor; and at least one memory comprising computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to: predict an episodic mobility route of a user equipment (UE) on which one or more instances of an application reside; obtain an indication of one or more edge server instances for the application, the one or more edge server instances being indicated based at least in part on a respective location of the one or more edge server instances in relation to at least a portion of the episodic mobility route of the UE; determine quality metrics for the one or more edge server instances in combination with one or more network anchor points; and establish a data session between the UE and a particular edge server instance through a particular network anchor point, wherein the particular edge server instance and the particular network anchor point are selected based at least in part on corresponding quality metrics.

2. The apparatus of claim 1, wherein the indication of the one or more edge server instances is obtained from an application management device associated with the application in response to providing the episodic mobility route of the UE to the application management device.

3. The apparatus of any of the preceding claims, wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to: responsive to an updated location of the UE and/or the episodic mobility rate, obtain a second indication of one or more second edge server instances for the application; select a particular second edge server instance based at least in part on quality metrics determined for the one or more second edge server instances in combination with the one or more network anchor points; and

- 43 - migrate the data session to the particular second edge server instance.

4. The apparatus of any of the preceding claims, wherein the episodic mobility route of the UE comprises a plurality of waypoints and a plurality of dynamic probabilities corresponding to the plurality of waypoints.

5. The apparatus of any of the preceding claims, wherein the one or more edge server instances are geographically located within one edge zone of one or more edge zones spanned by the episodic mobility route.

6. The apparatus of any of the preceding claims, wherein the indication of the one or more edge server instances comprises, for a respective edge server instance: a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes.

7. The apparatus of claim 6, wherein the quality metrics for the respective edge server instance are determined based at least in part on the sounding service node, the sounding service method, and the sounding service attributes.

8. The apparatus of any of the preceding claims, wherein the quality metrics for the respective edge server instance in combination with a respective network anchor point comprise a latency between the respective edge server instance and the respective network anchor point.

9. The apparatus of claim 8, wherein the particular edge server instance and the particular network anchor point are selected based at least in part on minimization of the latency.

10. An apparatus comprising: at least one processor; and at least one memory including computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to:

- 44 - obtain a prediction of an episodic mobility route of a user equipment (UE) on which one or more instances of an application reside; identify one or more edge zones spanned by the episodic mobility route; determine a set of edge server instances for the application and positioned within the one or more edge zones; receive an indication of a particular edge server instance selected from the set of edge server instances; and enable establishment of a data session between the UE and the particular edge server instance.

11. The apparatus of claim 10, wherein the indication of a particular edge server instance is received from the UE.

12. The apparatus of any of claims 10-11, wherein the indication of the particular edge server instance is received responsive to providing, for a respective edge server instance of the set of edge server instances, a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes.

13. The apparatus of any of claims 10-12, wherein the at least one memory and the computer program code are further configured to, with the at least one processor, cause the apparatus at least to: periodically receive subsequent predictions of the episodic mobility route of the UE; and responsive to receiving a subsequent prediction of the episodic mobility route of the UE, determine a second set of edge server instances for the application and positioned within a second edge zone spanned by the subsequent prediction of the episodic mobility route.

14. The apparatus of any of claims 10-13, wherein the prediction of the episodic mobility route is obtained responsive to receiving an indication of session establishment from the UE.

15. An apparatus comprising:

- 45 - means for predicting an episodic mobility route of a user equipment (UE) on which one or more instances of an application reside; means for obtaining an indication of one or more edge server instances for the application, the one or more edge server instances being indicated based at least in part on a respective location of the one or more edge server instances in relation to at least a portion of the episodic mobility route of the UE; means for determining quality metrics for the one or more edge server instances in combination with one or more network anchor points; and means for establishing a data session between the UE and a particular edge server instance through a particular network anchor point, wherein the particular edge server instance and the particular network anchor point are selected based at least in part on corresponding quality metrics.

16. The apparatus of claim 15, wherein the indication of the one or more edge server instances is obtained from an application management device associated with the application in response to providing the episodic mobility route of the UE to the application management device.

17. The apparatus of any of claims 15-16, further comprising: means for obtaining, responsive to an updated location of the UE and/or the episodic mobility rate, a second indication of one or more second edge server instances for the application; means for selecting a particular second edge server instance based at least in part on quality metrics determined for the one or more second edge server instances in combination with the one or more network anchor points; and means for migrating the data session to the particular second edge server instance.

18. The apparatus of any of claims 15-17, wherein the episodic mobility route of the UE comprises a plurality of waypoints and a plurality of dynamic probabilities corresponding to the plurality of waypoints.

19. The apparatus of any of claims 15-18, wherein the one or more edge server instances are geographically located within one edge zone of one or more edge zones spanned by the episodic mobility route.

20. The apparatus of any of claims 15-19, wherein the indication of the one or more edge server instances comprises, for a respective edge server instance: a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes.

21. The apparatus of claim 20, wherein the quality metrics for the respective edge server instance are determined based at least in part on the sounding service node, the sounding service method, and the sounding service attributes.

22. The apparatus of any of claims 15-21, wherein the quality metrics for the respective edge server instance in combination with a respective network anchor point comprise a latency between the respective edge server instance and the respective network anchor point.

23. The apparatus of claim 22, wherein the particular edge server instance and the particular network anchor point are selected based at least in part on minimization of the latency.

24. An apparatus comprising: means for obtaining a prediction of an episodic mobility route of a user equipment (UE) on which one or more instances of an application reside; means for identifying one or more edge zones spanned by the episodic mobility route; means for determining a set of edge server instances for the application and positioned within the one or more edge zones; means for receiving an indication of a particular edge server instance selected from the set of edge server instances; and means for enabling establishment of a data session between the UE and the particular edge server instance.

25. The apparatus of claim 24, wherein the indication of a particular edge server instance is received from the UE.

26. The apparatus of any of claims 24-25, wherein the indication of the particular edge server instance is received responsive to providing, for a respective edge server instance of the set of edge server instances, a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes.

27. The apparatus of any of claims 24-26, further comprising: means for periodically receiving subsequent predictions of the episodic mobility route of the UE; and means for determining, responsive to receiving a subsequent prediction of the episodic mobility route of the UE, a second set of edge server instances for the application and positioned within a second edge zone spanned by the subsequent prediction of the episodic mobility route.

28. The apparatus of any of claims 24-27, wherein the prediction of the episodic mobility route is obtained responsive to receiving an indication of session establishment from the UE.

29. A method comprising: predicting an episodic mobility route of a user equipment (UE) on which one or more instances of an application reside; obtaining an indication of one or more edge server instances for the application, the one or more edge server instances being indicated based at least in part on a respective location of the one or more edge server instances in relation to at least a portion of the episodic mobility route of the UE; determining quality metrics for the one or more edge server instances in combination with one or more network anchor points; and establishing a data session between the UE and a particular edge server instance through a particular network anchor point, wherein the particular edge server instance and the particular network anchor point are selected based at least in part on corresponding quality metrics.

- 48 -

30. The method of claim 29, wherein the indication of the one or more edge server instances is obtained from an application management device associated with the application in response to providing the episodic mobility route of the UE to the application management device.

31. The method of any of claims 29-30, further comprising: obtaining, responsive to an updated location of the UE and/or the episodic mobility rate, a second indication of one or more second edge server instances for the application; selecting a particular second edge server instance based at least in part on quality metrics determined for the one or more second edge server instances in combination with the one or more network anchor points; and migrating the data session to the particular second edge server instance.

32. The method of any of claims 29-31, wherein the episodic mobility route of the UE comprises a plurality of waypoints and a plurality of dynamic probabilities corresponding to the plurality of waypoints.

33. The method of any of claims 29-32, wherein the one or more edge server instances are geographically located within one edge zone of one or more edge zones spanned by the episodic mobility route.

34. The method of any of claims 29-33, wherein the indication of the one or more edge server instances comprises, for a respective edge server instance: a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes.

35. The method of claim 34, wherein the quality metrics for the respective edge server instance are determined based at least in part on the sounding service node, the sounding service method, and the sounding service attributes.

- 49 -

36. The method of any of claims 29-35, wherein the quality metrics for the respective edge server instance in combination with a respective network anchor point comprise a latency between the respective edge server instance and the respective network anchor point.

37. The method of claim 36, wherein the particular edge server instance and the particular network anchor point are selected based at least in part on minimization of the latency.

38. A method compri sing : obtaining a prediction of an episodic mobility route of a user equipment (UE) on which one or more instances of an application reside; identifying one or more edge zones spanned by the episodic mobility route; determining a set of edge server instances for the application and positioned within the one or more edge zones; receiving an indication of a particular edge server instance selected from the set of edge server instances; and enabling establishment of a data session between the UE and the particular edge server instance.

39. The method of claim 38, wherein the indication of a particular edge server instance is received from the UE.

40. The method of any of claims 38-39, wherein the indication of the particular edge server instance is received responsive to providing, for a respective edge server instance of the set of edge server instances, a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes.

41. The method of any of claims 38-40, further comprising: periodically receiving subsequent predictions of the episodic mobility route of the UE; and

- 50 - determining, responsive to receiving a subsequent prediction of the episodic mobility route of the UE, a second set of edge server instances for the application and positioned within a second edge zone spanned by the subsequent prediction of the episodic mobility route.

42. The method of any of claims 38-41, wherein the prediction of the episodic mobility route is obtained responsive to receiving an indication of session establishment from the UE.

43. A computer program product comprising at least one non-transitory computer readable storage medium having computer executable program code instructions stored therein, the computer executable program code instructions comprising program code instructions configured, upon execution, to: predict an episodic mobility route of a user equipment (UE) on which one or more instances of an application reside; obtain an indication of one or more edge server instances for the application, the one or more edge server instances being indicated based at least in part on a respective location of the one or more edge server instances in relation to at least a portion of the episodic mobility route of the UE; determine quality metrics for the one or more edge server instances in combination with one or more network anchor points; and establish a data session between the UE and a particular edge server instance through a particular network anchor point, wherein the particular edge server instance and the particular network anchor point are selected based at least in part on corresponding quality metrics.

44. The computer program product of claim 43, wherein the indication of the one or more edge server instances is obtained from an application management device associated with the application in response to providing the episodic mobility route of the UE to the application management device.

45. The computer program product of any of claims 43-44, wherein the computer executable program code instructions comprise program code instructions further configured, upon execution, to:

- 51 - responsive to an updated location of the UE and/or the episodic mobility rate, obtain a second indication of one or more second edge server instances for the application; select a particular second edge server instance based at least in part on quality metrics determined for the one or more second edge server instances in combination with the one or more network anchor points; and migrate the data session to the particular second edge server instance.

46. The computer program product of any of claims 43-45, wherein the episodic mobility route of the UE comprises a plurality of waypoints and a plurality of dynamic probabilities corresponding to the plurality of waypoints.

47. The computer program product of any of claims 43-46, wherein the one or more edge server instances are geographically located within one edge zone of one or more edge zones spanned by the episodic mobility route.

48. The computer program product of any of claims 43-47, wherein the indication of the one or more edge server instances comprises, for a respective edge server instance: a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes.

49. The computer program product of claim 48, wherein the quality metrics for the respective edge server instance are determined based at least in part on the sounding service node, the sounding service computer program product, and the sounding service attributes.

50. The computer program product of any of claims 43-49, wherein the quality metrics for the respective edge server instance in combination with a respective network anchor point comprise a latency between the respective edge server instance and the respective network anchor point.

- 52 -

51. The computer program product of claim 50, wherein the particular edge server instance and the particular network anchor point are selected based at least in part on minimization of the latency.

52. A computer program product comprising at least one non-transitory computer readable storage medium having computer executable program code instructions stored therein, the computer executable program code instructions comprising program code instructions configured, upon execution, to: obtain a prediction of an episodic mobility route of a user equipment (UE) on which one or more instances of an application reside; identify one or more edge zones spanned by the episodic mobility route; determine a set of edge server instances for the application and positioned within the one or more edge zones; receive an indication of a particular edge server instance selected from the set of edge server instances; and enable establishment of a data session between the UE and the particular edge server instance.

53. The computer program product of claim 52, wherein the indication of a particular edge server instance is received from the UE.

54. The computer program product of any of claims 52-53, wherein the indication of the particular edge server instance is received responsive to providing, for a respective edge server instance of the set of edge server instances, a sounding service node identifying the respective edge server instance, a sounding service method, and sounding service attributes.

55. The computer program product of any of claims 52-54, wherein the computer executable program code instructions comprise program code instructions further configured, upon execution, to: periodically receive subsequent predictions of the episodic mobility route of the UE; and

- 53 - responsive to receiving a subsequent prediction of the episodic mobility route of the UE, determine a second set of edge server instances for the application and positioned within a second edge zone spanned by the subsequent prediction of the episodic mobility route.

56. The computer program product of any of claims 52-55, wherein the prediction of the episodic mobility route is obtained responsive to receiving an indication of session establishment from the UE.

- 54 -