EP4677819A1

EP4677819A1 - Methods and apparatuses for learning the delay of impact of a network function

Info

Publication number: EP4677819A1
Application number: EP23765253.2A
Authority: EP
Inventors: Sabrine AROUA; Mathieu Leconte; Abdulrahman ALABBASI; Athanasios KARAPANTELAKIS; Danesh DAROUI; Lackis ELEFTHERIADIS; Marin ORLIC; Oleg GORBATOV
Original assignee: Telefonaktiebolaget LM Ericsson AB
Current assignee: Telefonaktiebolaget LM Ericsson AB
Priority date: 2023-03-03
Filing date: 2023-09-05
Publication date: 2026-01-14
Also published as: WO2024183931A1

Abstract

Embodiments described herein relate to methods and apparatuses for learning the delay of impact of network functions in a network. A method in an analytics network function comprises inputting a first time series into a machine learning, ML, model, wherein the first time series comprises: analytics information, L _t→T, relating to whether an analytics report was transmitted to each of the one or more NFs between an initial time t and a first time t + T;, and state information, X _t→T„ related to one or more performance metrics in the network between the initial time and the first time, t + T; and outputting, from the ML model, for each of the one or more NFs, f _i, a plurality of probability values formula (I) (t ₁ < t ₂, t ₁ ϵ {t,..., t + T}, t ₂ ϵ {t,..., t + H}), where the first time is earlier than a third time, t + H, wherein the plurality of probability values are each representative of a probability that the delay of the impact of the NF f _i is equal to t2 - t1.

Description

METHODS AND APPARATUSES FOR THE DELAY OF IMPACT OF A NETWORK FUNCTION Technical Field Embodiments described herein relate to the learning of the delay of impact of network function in a network. Background Generally, all terms used herein are to be interpreted according to their ordinary meaning in the relevant technical field, unless a different meaning is clearly given and/or is implied from the context in which it is used. All references to a/an/the element, apparatus, component, means, step, etc. are to be interpreted openly as referring to at least one instance of the element, apparatus, component, means, step, etc., unless explicitly stated otherwise. The steps of any methods disclosed herein do not have to be performed in the exact order disclosed, unless a step is explicitly described as following or preceding another step and/or where it is implicit that a step must follow or precede another step. Any feature of any of the embodiments disclosed herein may be applied to any other embodiment, wherever appropriate. Likewise, any advantage of any of the embodiments may apply to any other embodiments, and vice versa. Other objectives, features and advantages of the enclosed embodiments will be apparent from the following description. A Network Data Analytics Function (NWDAF) is a 3GPP network function in the 5G Core Networks 3GPP TS 23.288 v 18.0.0. The NWDAF is designed to collect data from diverse data sources such as User Equipments (UE), Network Functions (NF), and the Operation, Administration and Maintenance (OAM) located in the 5G Core, Cloud, and Edge networks. The NWDAF can generate different analytic reports (statistics/predictions) about the past and future of the system's states. The NWDAF may then provide analytic reports to 5G Core NFs to advise them about the changes in the network's behavior and eventual events that may happen. Therefore, once an NF receives an analytic report, it may interact with its environment through a specific action in the pursuit of a goal. The NWDAF can generate different reports to provide each NF with information that it has requested. The NFs may also be referred to as service consumers since they benefit from the NWDAF's services, such as Slice load level related network data analytics, UE-related analytics, and user data congestion analytics. Generally, an NF subscribes to NWDAF to receive predictions on the part of the system's state they are interested in. For example, an NF may subscribe to NWDAF to receive: an event-based, periodic, threshold, or (on-demand) one-time notification. In the one-time subscription, it will be appreciated that the NF may be automatically unsubscribed once it receives an analytic report from NWDAF. Once one NF receives the requested analytic report(s), the NF may decide whether to interact with the surrounding environment, and if so how to interact in order to improve one or more Key Performance Indicators (KPls) such as delay and traffic throughput. In current existing systems, the NFs' actions are not revealed to NWDAF. An action performed by an NF may be considered as the mechanism one NF can make to produce a transition from one network state to another network state. A network state may comprise the values of one or more KP ls of the network at a particular time. In the current design of 5GC (Rel.18), the NWDAF is responsible for delivering analytic reports to NFs upon received requests or subscriptions, whilst the actions which are taken by the NFs in response to the analytic reports are unknown to the NWDAF. Generally, the existing NFs receive analytic reports from the NWDAF at different points in the time. For example, two NFs that have subscribed to receive periodic notifications might receive their requested predictions at different periods. For example, as depicted in Figure 1, one NF (e.g. NF1) may receive analytic reports every 30 ms, and then the second NF (e.g. NF2) may receive analytic reports every 45 ms. Accordingly, every NF has its own time scale over which to learn about the network's evolution and to react, i.e., trigger actions. In such a system, the NFs' requests and decision-making time scales might be very different, and therefore may not be synchronized. Moreover, the impact of the NF's actions on the surrounding environment might not be instantaneous, meaning it might take time for the action of the NF to have an impact on the performance metrics in the network. Figure 2 illustrates how the time for a impact on the performance metrics to occur depends on the NFs. Also, what impact actually occurs to the performance metrics may also depend on the NF and the action taken by the NF. In the graph 200, a first NF receives an analytic report at time t1. The NF then makes an action at time t2. However, the overall impact on the performance metrics does not occur until time t3. The difference between t3 and t1 may be referred to as the delay of the impact of the first NF. In the graph 201, a second NF receives an analytic report at time t1. The NF then makes an action at time t2'. It will be appreciated that t2' may be different to t2. In this case, the overall impact on the performance metrics does not occur until time t3' (which again may be different to t3). The difference between t3' and t1 may be referred to as the delay of the impact of the second NF. It can also be seen that the impact of the second NF lowers a performance metric whereas the impact of the first NF increases the performance metric. Summary Most existing studies focus on proposing new solutions to learn the set of performance metrics that might interest a specific NF, in other words, the set of performance metrics that one NF wants to improve. Accordingly, while dealing with protected NF actions, the focus of existing research studies is on how to quantify the impacts of the NFs actions while observing the network's state transition. None of the existing studies tackle the delayed effects of the NFs' actions. Most existing studies simplify the network model. They assume that when one NF receives an analytic report, the time, i.e., delay, required to make the transitions in the network state is known by NWDAF. Embodiments described herein address the problem that the NWDAF has no information about the properties of the action(s), e.g., timing properties of the actions, taken by a NF. The NWDAF can currently only provide information about the network's state and the impact of the triggered actions. Hence a lack of such knowledge on the NWDAF's side can lead to a possibly inaccurate analysis of the performance variations. According to some embodiments there is provided A computer-implemented method for transmitting analytic reports to one or more network functions, NFs, in a network. The method comprises at a first time, t + T , receiving a first request for an analytic report from a first NF, f ₁, of one or more NFs, a first analytic report to the first NF responsive to the first request, using time-series forecasting to determine a first probability value «l_+T,t_+T+T, representative of a probability that a delay of an impact of the NF, f1, is equal to a first delay, T_i ; at a second time, t + t', receiving a second request for an analytic report from a second NF, fa, of the one or more NFs, where the second time, t + t', is earlier than a fourth time, t + T + Ti ; and responsive to the first probability value meeting a first criteria, transmitting a second analytic report to the second NF with an indication that a state of one or more performance metrics will be changed at a future time. According to some embodiments there is provided a computer-implemented method for learning a delay of impact of one or more network functions' interactions with a network. The method comprises inputting a first time series into a machine learning, ML, model, wherein the first time series comprises: analytics information, L_t➔T • relating to whether an analytics report was transmitted to each of the one or more NFs between an initial time t and a first time t+ T, and state information, X_t➔T,, related to one or more performance metrics in the network between the initial time and the first time, t + T; and outputting, from the ML model, for each of the one or more NFs, /;,, a plurality of probability values a!₁.i_i (t₁� t₂ , t₁ e {t, ... , t + T}, t₂ e {t, ... , t + H}), where the first time is earlier than a t + H, wherein the plurality of probability values are each representative of a probability that the delay of the impact of the NF/;, is equal to t2 - t1. According to some embodiments there is provided a method, in a second network function, NF for utilising analytic reports in a network. The method comprising: at a second time, t + t', transmitting a second request for an analytic report to an analytics NF, where the second time, t + t', is earlier than a fourth time, t + T + T_i receiving a second analytic report from the analytics NF comprising an indication that a state of one or more performance metrics in the network will be changed at the fourth time t + T + T_i ; and determining whether to implement an action in the network based on the indication. According to some embodiments there is provided an analytics network function for transmitting analytic reports to one or more NFs in a network. The analytics network function comprises processing circuitry configured to cause the analytics network function to: at a first time, t + T , receive a first request for an analytic report from a first NF, f_t , of one or more NFs, transmit a analytic report to the first NF responsive to the first request; use time-series forecasting to determine a first probability value _«l+T,t+T+Ti representative of a probability that a delay of an impact of the NF, f1, is equal to a first delay, T_i; at a second time, t + t', receive a second request for an analytic report from a NF, f2, of the one or more NFs, where the second time, t + t', is ^{earlier than a fourth time, t} + ^T + _h ^{and responsive to the first probability value meeting} a first criteria, transmit report to the second NF with an indication that a state of one or more performance metrics will be changed at a future time. According to some embodiments there is provided an analytics network function for learning a delay of impact of one or more network functions' interactions with a network. The analytics network function comprises processing circuitry configured to cause the analytics network function to: input a first time series into a machine learning, ML, model, wherein the first time series comprises: analytics information, L_t➔T • relating to whether an analytics report was transmitted to each of the one or more NFs between an initial time t and a first time t+ T, and state information, X_t➔T " related to one or more performance metrics in the network between the initial time and the first time, t + T; and output, from the ML model, for each of the one or more NFs, Ji,, a plurality of probability values a�_1.t/t₁� t₂ , t₁ e {t, ... , t + T}, t₂ e {t, ... , t + H}), where the first time is earlier than a third time, t wherein the plurality of probability values are each representative of a probability that the delay of the impact of the NF f_i is equal to t₂ - t₁. According to some embodiments there is provided a second network function, NF, for utilising analytic reports in a network. The second network function comprises processing circuitry configured to cause the second network function to: at a second time, t + t', transmit a second request for an analytic report to a analytics NF, where the second time, t ₊ t', is earlier than a fourth time, t ₊ T ₊ T_i receive a second analytic report from the analytics NF comprising an indication that a state of one or more performance metrics in the network will be changed at the fourth time t + T +Ti; and determine whether to implement an action in the network based on the Aspects and examples of the present disclosure thus enable an analytics network function to understand how and when the network's state will change. For the purposes of the present the term "ML model" encompasses within its scope the following concepts: Machine Learning algorithms, comprising processes or instructions through which data may be used in a training process to generate a model artefact for performing a given task, or for representing a real world process or system; the model artefact that is created by such a training process, and which comprises the computational architecture that performs the task; and the process performed by the model artefact in order to complete the task. References to "ML model", "model", model parameters", "model information", etc., may thus be understood as relating to any one or more of the above concepts encompassed within the scope of "ML model". Brief Description of the Drawings For a better understanding of the embodiments of the present disclosure, and to show how it may be put into effect, reference will now be made, by way of example only, to the accompanying drawings, in which: Figure 1 is a time line illustrating the receipt of requests for analytic reports from two network functions; Figure 2 illustrates how the time for a recordable impact on the performance metrics to occur depends on the NFs; Figure 3 illustrates an example in which network state transitions to a new network state due to the interaction of just one NF; Figure 4 illustrates how the performance metrics may change when there is an overlap between two or more NFs' interactions with the environment; Figure 5 illustrates a computer-implemented method for learning a delay of impact of one or more network functions' interactions with a network; Figure 6 is a signalling diagram illustrating how a NF subscribes to receive analytic reports; Figure 7 illustrates an example of the ML model; Figure 8 illustrates a computer-implemented method for transmitting analytic reports to one or more NFs in a network; Figure 9 illustrates a method, in a second network function, for utilising analytic reports; Figure 10 is a signalling diagram illustrating an example implementation of the methods of Figures 8 and 9; Figure 11 illustrates an example of the inputs and outputs of the ML model according to some embodiments; Figure 12 illustrates a network function comprising processing circuitry (or logic); Figure 13 is a block diagram illustrating an analytics network function according to some embodiments; Figure 14 is a block diagram illustrating an analytics network function according to some embodiments; Figure 15 is a block diagram illustrating a second network function according to some embodiments. Description The following sets forth specific details, such as particular embodiments or examples for purposes of explanation and not limitation. It will be appreciated by one skilled in the art that other examples may be employed apart from these specific details. In some instances, detailed descriptions of well-known methods, nodes, interfaces, circuits, and devices are omitted so as not to obscure the description with unnecessary detail. Those skilled in the art will appreciate that the functions described may be implemented in one or more nodes using hardware circuitry (e.g., analog and/or discrete logic gates interconnected to perform a specialized function, ASICs, PLAs, etc.) and/or using software programs and data in conjunction with one or more digital microprocessors or general purpose computers. Nodes that communicate using the air interface also have suitable radio communications circuitry. Moreover, where appropriate the technology can additionally be considered to be entirely within any form of computer readable memory, such as solid-state memory, magnetic disk, or optical disk containing an appropriate set of computer instructions that would cause a processor to carry out the techniques described herein. Hardware implementation may include or encompass, without limitation, digital signal processor (DSP) hardware, a reduced instruction set processor, hardware (e.g., digital or analogue) circuitry including but not limited to application specific integrated circuit(s) (ASIC) and/or field programmable gate array(s) (FPGA(s)), and (where appropriate) state machines capable of performing such functions. As described above, Figure 2 illustrates the concept of the delay of the impact of a particular NFs interaction with the network. The delay of the impact of an NFs' interaction with the network may be considered as the time required by the NF to transition the network's state (defined by a set of performance metrics (e.g. KPls)) to a new state, once the NF receives an analytic report from the NWDAF. Since it may be desirable to keep the actions of the NFs well-protected (e.g. hidden from the NWDAF). The delay of the impact may be considered to start when the NF receives the analytic report. However, as the NFs in a network may not be synchronized regarding request time distribution in a real network scenario. For example, one NF interested in traffic delay might send a request for (or be subscribed to receive) an analytic report every 2 seconds. A second NF, e.g. focusing on congestion control, may send a request for or be subscribed to receive a prediction of network congestion every 30 s, and a third NF, e.g. interested in physical resource block (PRB) allocation, may send a request for or be subscribed to receive an analytic report on PRB allocation every 1 min. These three NF might impact the same set of performance metrics, for example, they all may impact the network to reduce network delay. Thus, their impacts on the system may overlap. Consequently, the NWDAF may not be able to easily understand which NF is causing any particular variation in the delay. Thus, the network transitioning to a new network state (e.g. a lower delay performance metric) may be a cumulative result of the actions of different NFs during multiple time instants. Figures 3 and 4 depict two examples of diagrams that illustrate the variation in the network state (KPls) based on the interaction of one or more NFs with the environment. Figure 3 illustrates an example in which network state transitions to a new network state due to the interaction of just one NF, e.g. NF1. In step 301, the NF1 subscribes to receive analytic reports from the NWDAF. For example the NF1 may subscribe to receive analytic reports about predicted changes to a particular performance metric. In step 302, the NF2 subscribed to receive analytic reports from the NWDAF. For example, the NF2 may subscribe to receive the same analytic reports as the NF1, or may subscribe to receive different analytic reports. In some examples, after subscribing to receive analytic reports the NF1 may request an analytic report, as illustrated in step 303. It will be appreciated that in some cases the subscription may indicate particular circumstances in which analytic reports should be sent to an NF. In these cases, the NF may not need to send individual requests for analytic reports. In step 304, the NWDAF transmits an analytic report to NF1. In step 305, the NF1 performs an action in the environment. At a later time, the network state changes from an initial state X₁ to a new state X₂. In this scenario, calculating the delay of the impact might easily be learned since the NWDAF does not send an analytic report to NF2, (or they are transmitted sufficiently earlier than request of step 303 or occur after the state change). Alternatively, or additionally, in this example it may be assumed that one NF can't receive a report from the NWDAF until another NF achieves its goal in terms of performance metric improvement. Therefore, in Figure 3, the new state X₂ may be determined to be the impact of NF1 's interaction with the network and the transition from X₁ to X₂ is thus the result of the NF1 's interaction in the environment. The impact for NF1 may therefore be the time between the receipt of step 304 and the change to the new state X₂. Figure 4 illustrates how the performance metrics may change when there is an overlap between two or more NFs' interactions with the environment, i.e., NFs impacting non disjoint sets of performance metrics. Steps 401 to 405 correspond to steps 301 to 305 of Figure 3. However, in Figure 4, in step 406, the NF2 requests an analytics report. In step 407, the NWDAF transmits an analytic report to the NF2. It will be appreciated that in some circumstances the request of step 406 is omitted as the NF2 has subscribed to receive relevant analytic reports. In step 408 the NF2 performs an action in the environment. In this example, at a later time than both of steps 405 and 408, the network state changes from an initial state X ₁ to a new state X₂. The state of the network may then evolve further, for example into states X ₃ and X₄. In this example, therefore, the first variation in the network state appears after both NF1 and NF2 have interacted with the environment. Thus, it is challenging for the NWDAF to understand which NF is impacting the network state to cause the transitions from X ₁ to ^X ₂ ^. In order to overcome this issue, embodiments described herein enable the NWDAF to learn and understand how and when the network state will vary over the course of time, without any knowledge about the various NF's actions. Long and short-term dependency patterns between the NFs' requested analytics reports and the transition in specific performance metrics may be used to introduce a framework that enables the NWDAF to learn and understand the delay required by one NF to make observable transitions in the network state. Finally, using the learned delay of the impact, the NWDAF may be able to provide NFs with additional information about the expected network state variations over time. By receiving knowledge about expected network state variations due to actions from other NFs, a NF may be able to trigger (or refrain from triggering) proper action(s) to reach a desired state of the network. Embodiments described herein enable NWDAF to learn the delay of the impact for each NF. For example, by utilizing the time series forecasting methods and/or temporal attention mechanisms, the NWDAF may learn how the environment will change over time. Once a NF receives an analytic report, the NWDAF may then predict when the network state will transition to a new one. Moreover, this functionality, may deal with the overlap between different NFs' interactions with the network state, i.e., the NFs that impact the same set of performance metrics during the same period. As a result, NWDAF may quantify one NF's impact on the network state. Based on the learned delay of the impact, in some embodiments, the NWDAF may provide the NFs with additional information about the network state transition when an analytic report is requested. This further information may indicate how the NFs interacting with the network during a time period relevant to the requesting NF will impact the network states. Accordingly, some NFs may receive this additional information along with their requested analytic reports. It will be appreciated that subscribed NFs may transmit individual requests to the NWDAF to receive predictions relating to performance metrics within the network states they are interested in. It will also be appreciated that an NF may analyse any received analytic report, and may or may not interact with its surrounding environment to make some changes in the network state in response to receiving the analytic report. To enable NWDAF to learn the delay of the impact of the different NFs' interactions, one or more of the following may be assumed: 1) For each NF, the NWDAF already knows: - The set of performance metrics, e.g., delay, throughput, etc., that may be impacted by the NF interacting with the network. Thus, NWDAF knows the relevant dimensions of the state space on which every NF acts. - The request time distribution, i.e., frequency of submitting a request for notifications. - The maximum delay of the impact of every NF interaction with the systems. The maximum delay of the impact may be estimated based on the time scale of the NF's requested analytic reports. o For simplicity, it may be considered that the maximum delay of the impact is limited to the periodicity of receiving a request for an analytic report. For example, if one NF a request for an analytic report every 2 s, then its delay of the impact may be assumed not to exceed 2 s. o For non-periodic requests, it may be assumed that the NWDAF is configured with a maximum threshold to limit the delay of the impact. 2) The NWDAF has limited observations about each NF's decided action every time it receives an analytic report. The NWDAF may instead observe variations in performance metrics based on the actions taken in response to analytic reports transmitted to one or more NFs. For example, when NF1 receives an analytical report, NWDAF may observe a variation in traffic load. 3) Different eventual possible actions may be considered by the NFs, for example, the management of the Physical Resource Blocks (PRBs), the traffic scheduling, etc. Herein NFs are described as being able to make actions in the network. It will be appreciated that to make an action in the network the NF may adjust one or more network parameters which may have an effect on one or more performance metrics (e.g. KPls). However, the variation in these performance metrics may not be instantaneous. The following comprises examples of possible performance metrics that may be altered by the actions of NFs in a network: - Number of retransmissions Data rates - Active power-saving features for radio units - Active RAN features in a specific network, such as Microsleep TX, Less, MINO, etc Power consumption For each performance metric, one or more NFs may be able to take an action in the network in order to adjust the performance metrics. The following comprises examples of possible actions in the network: - Offload to other cells - Offload to other cells, or use Delay of a power amplifier (PA) switching ON and OFF - Schedule/or reschedule re-coordination of RAN features to be active. For example, reducing the delay may result in lower power consumption, or by adding a new band to reduce the interference may improve the quality of service. Figure 5 illustrates a computer-implemented method for learning a delay of impact of one or more network functions' interactions with a network. The one or more NFs may be grouped together according to which performance metrics the one or more NFs may change in the network. To learn the delay of the impact of the NF interactions with the network, a ML model may be used. The main objectives of the proposed model are: - Learning/predicting how the network state will change during every time slot. - Quantifying the participation of every NF in the network's state transitions. This quantification may be presented in terms of probabilities, as will be described below. The method 500 may be performed by a network function, which may comprise a physical or virtual node, and may be implemented in a computing device or server apparatus and/or in a virtualized environment, for example in a cloud, edge cloud or fog deployment. It will be appreciated that the method 500 may be performed by a NWDAF. To enable NWDAF performing the method of Figure 5 to follow the variation in the network's state, it may be assumed that the time axis can be divided into several time slots with equal lengths�t. The duration of one slot�t may be set equal to the shortest duration that separates two consecutive requests for analytical reports from the same NF. It may be understood that F = {f₁ , ... , f_n} comprises a set of one or more NFs. By t and r_i (t) are denoted the time instant t and a request sent by/;, at t to NWDAF, respectively. As depicted in Figure 6, once /;, transmits the request r_i (t), the NWDAF may answer with the requested analytic that is denoted as l_i (t). 6 illustrates an example in which{;_ subscribes to analytic reports in step 601. In step 602 {;_ submits a request r_i (t). In steps 603 and 604 the NWDAF collects data for the analytic report from a data source. In step 605, the NWDAF transmits the analytic report l_i (t) to _{{;_} ^. In step 606 f_i performs an action in the network in response to the information in the analytic report l_i (t). However, as depicted in Figure 4, different NFs might interact with the surrounding environment during the same period. As a result, they might impact correlated or the same performance metrics. For example, in Figure 4, the first variation in the system state appears after /₁ and /₂ have both received their requested analytic reports, l₁ and l₂ ^, respectively. As a result, they might have a joint impact on the system state transition. The notation x_t, denotes the state information related to one or more performance metrics in the network at time t. x(t) e�^m . me N where m may represent the number of performance metrics that are the network state. By x_i is denoted a set of performance metrics that f_i is interested in. In this example, it may be assumed that: E^{very pair of N Fs act on a non-empty set of joint features: v {;_, fj e F, x}i ^{n x} _i * ^0; and In one time slot, only one NF may transmit a request for an analytic report, and only one NF may start to interact with the network environment. In step 501 the method comprises inputting a first time series into a machine learning, ML, model. It will be appreciated that the ML model may be trained and used by the NWDAFs for the one or more NFs. The NWDAF may produce and train other ML models for other groups of NFs which are, for example, able to change other types of performance metrics in the network. The first time series comprises: analytics information, L_t➔t+T • relating to whether an analytics report was transmitted to each of the one or more NFs between an initial time t and a first time t + T. For example, the analytics information may comprise, for example of the one or more NFs, f_i , an analytic report value, l_i (t) for each time t ➔ t + T . For example: NFs. li₍t) is the analytic reporVprediction _{sent to Ii} ^at _{t. li(t) e [X, Y] if the NF Ii} ^{was transmitted an analytic report at time} _t. ^{If f;,} was not transmitted an analytic at t, then li(t) equals z ^EJ= [X, Y]. In examples X = 0, Y = 1 and Z= -1. The first time series further comprises: state information, x(t), related to one or more performance metrics in the network between the initial time t and the first time, t + T. The state information comprises a time series _x(t) e IR!.^m that represents the network state (e.g. the one or more performance metrics in the network). More formally, the first time series Y_t➔r may be expressed as: Y_t ^➔ _t+T = {y(t)^, y(t + 1)^, ... ,^y(^t + ^T)^} +_{- (}t₎, l , + , l + , ... , + _, l₍t + ₎} In one or ^{more NFs, I}i ^{a plurality of probability values} _a!1.t_z ^(t1� ^t2 ^{, t}1 ^{e {t, ... , t + T}, t}2 ^e _{t, ... , t + H}₎, where the first time _t + T is earlier than a third time, t + H, wherein the plurality of probability values are each representative of a probability that the delay of the impact of the NF fi is equal to _t2- _t1. In other words, f;, is partipating in the system state's transition from x₍t₁ ⁾ to x₍t₂ ⁾ with a probability equal to a^� ₁ ^,t ₂ ^. Also, a^� _1,i2 indicates the conditional probability that the _{predicted system from x(t1} ⁾ _{1 ( 2} ⁾ ₂ ^{given that f;,} _li ^{at t to x t at t received at} ^t ₁ ^{, i.e.,} _ai1.tz ^{= P(x(t} ₁ ^{) ➔ x(t} ₂ ^)/l _i ^(t ₁ ^)). In 5 may further comprise deriving a second time ^{series comprising predicted values, x(t), of the state information between the first time,} _{t, and a third time, t + H. The second time series may be defined as:} x_t ^➔ _t+H = {^{x(t),x(t +} 1)^{, ...} ,x(t + T), ... ,x(t + H₎J It will be appreciated that the second time series is derived from the plurality of probability values for each of the one or more NFs (for example, as illustrated later with reference to Figure 7). The second time series X_t➔t+H a concatenation of two lists X_t➔t+r + _{x(t),, ... ,_x(t + T)} and X_t+r_{+l➔t+H +- {x}(t + T + 1), ... ,_x(t + _H)_}. _xt➔t+r presents a prediction of a part of first time series Y_t ^➔T· In other words, X_t ^➔ _t+r is a prediction of ^X _t➔_t+ ^r = (^x(t), ^{... , x}(t + T)). X_t➔_t+ ^{r is predicted in the output X} _t➔_t+H ^{in order to learn the} ^{impact on the network states} _{{x(t' + 1} ^), _{... ,x(t + T)} of every NF's interactions} performed during the times t' e {t, ... , T} given the received analytic reports l_i (t'). It will be appreciated that the ML model may be trained such that the prediction X_t➔t+r ^{converges to the actual values of X} _t➔_t+ ^r = ^(x(t), ... ,^{x(t + T)).} The method of Figure 5 may therefore comprise training the ML model by updating one or more parameters of the ML model to minimise a loss function derived from the state information, x(t), and the second time series, x'(t). The minimisation of the loss function may be expressed as: mJⁿ L ^{lx(t) - x(t)I;} where 0 denotes the !l_rrain comprises a set of time ^{stamps used for training (e.g. t tot+ T), and is the Frobenius norm.} The plurality of probability values output in step 502 may be utilised to transmit an indication of NFs that state information will change from current state information at some point in the future (e.g. as will be described in more detail with reference to Figure 8). However, the probability values may be used in one or more other ways. For example, if a particular probability value is higher than a predetermined threshold, then NWDAF may use it to update future predictions about the network state, which may, in turn, have an effect on future analytic reports. It will also be appreciated that, based on the output probabilities, the NWDAF may delay processing of other requests for analytic reports until the (predicted) time indicates that the network has been impacted, in order to avoid conflict. The ML model may comprise any form of suitable ML model, for example an artificial neural network (ANN). In some examples, the ML model comprise one or more of: a Convolutional Neural Network layer, a recurrent layer, and a temporal attention layer. Figure 7 illustrates an example of the ML model 700. The ML model 700 comprises a CNN 701 (e.g. a CNN layer without pooling). The CNN layer 701 may for example aim to extract short-term patterns in the time dimension as well as local dependencies between the NF's requested analytic reports, and thereby the potential impact on the network state, i.e., the performance metrics. The ML model 700 also comprises a recurrent layer 702. The recurrent layer 702 may be used to memorize historical information and therefore to be aware of relatively long term dependencies in input data. The ML model also comprises a temporal attention layer 703. The temporal attention layer 703 may output the probabilities, at,t₂ ^• The second time series X_t ^➔H may be derived from the probabilities a_t ⁱ 1_{, t 2}. For example, the state x(t be derived from all probabilities where t ₂ = t + i. Figure 8 illustrates a computer-implemented method for transmitting analytic reports to one or more NFs in a network. The method 800 may be performed by a network function, which may comprise a physical or virtual node, and may be implemented in a computing device or server apparatus and/or in a virtualized environment, for example in a cloud, edge cloud or fog deployment. It will be appreciated that the method 800 may be performed by an analytics network function, e.g. a NWDAF. In step 801, the method comprises at a first time, t + T, receiving a first request for an analytic report from a first NF, f₁, of one or more NFs. In step 802, the method comprises transmitting a first analytic report to the first NF responsive to the first request. In step 803, the method comprises using time-series forecasting to determine a first probability value al_+r,t₊T₊Tc Representative of a probability that a delay of an impact of the NF, f, is 803 may comprise utilising the method as described with reference to Figure 5 to determine the probability value al_+r,_t+T₊Tc It will be appreciated that other methods, which may or may utilise machine learning, may be used to determine the first probability value. In step 804 the method comprises at a second time, t + t', receiving a second request for an analytic report from a second NF, f2, of the one or more NFs, where the second time, t + t', is earlier than a fourth time, t + T + h In step 805, responsive to the first probability value meeting a first criteria, transmitting a second analytic report to the second NF with an indication that a state of one or more performance metrics will be changed at a future time. The first criteria may comprise comparing the first probability value to a predetermined threshold. For example, if the first probability value is greater or equal to the predetermined threshold this may indicate that the state of the one or more performance metrics is likely to change as predicted, and that therefore it would be useful to transmit this indication to the second NF. In other words, the first probability value may be considered to meet the first criteria if it is greater than or equal to a first predetermined threshold value. However, if the first probability value does not meet the first criteria, the method may comprise refraining from transmitting, to the second NF, the indication that the state of the one or more performance metrics will be changed at the future time. In this example, the method may simply comprise transmitting the second analytic report to the second NF without the indication. In some examples, the indication that a state of one or more performance metrics will be changed at a future time may only be transmitted to the second NF is a subscription request has been received from the second NF subscribing to receive predicted values of state information. Figure 9 illustrates a method, in a second network function, for utilising analytic reports. The method 900 may be performed by a network function, which may comprise a physical or virtual node, and may be implemented in a computing device or server apparatus and/or in a virtualized environment, for example in a cloud, edge cloud or fog deployment. It will be appreciated that the method 800 may be a NF that is consuming the services of an NWDAF. In step 901, at a second time, t + t', the comprises transmitting a second request for an analytic report to an analytics network function, where the second time, t + t', is earlier than a fourth time, t + T + h- In step 902, the method comprises receiving a second analytic report from the analytics network function comprising an indication that a state of one or more performance metrics in the network will be changed at the fourth time t + T + h In step 903, the method comprises determining whether to implement an action in the network based on the indication. For example, the NF may decide to refrain from performing any action in the network that it would otherwise have performed in response to the second analytic report as the prediction of the state at the fourth time means that the action is not going to be necessary. In other words, based on the previously learned information, i.e., the predicted future system state and the probability that one NF is impacting the variation in this network state, the methods of Figures 8 and 9 look to provide the NFs with additional information (e.g. the indication of step 902). This additional information indicates how the network state will change based on the other NFs interactions with the network. Figure 10 is a signalling diagram illustrating an example implementation of the methods of Figures 8 and 9. In step 1001 and 1002 the NFs f_i and [j subscribe to predicted values of state information. In step 1003 at a first time, t + T, the NWDAF receives a first request for an analytic report from a first NF, f1, of one or more NFs. Step 1003 comprises an example implementation of step 801 of Figure 8 The NWDAF uses time-series forecasting (for example, as described above) to determine a first probability value a't_+r,t_+r+ri representative of a probability that a delay of an impact of the NF, f1, is equal to Ti. In step 1004, the NWDAF transmits a first analytic report to the first NF responsive to the first request. Step 1004 corresponds to step 802 In step 1005, the first NF performs an action in the network in response to the first analytic report. In step 1006, at a second time, t + t', the NWDAF receives a second request for an analytic report from a second NF, fa, of the one or more NFs, where the second time, t + t', is earlier than a fourth time, t + T + Ti. Step 1006 corresponds to 804 and step 901. ^{In step 1007, responsive to af+r ' t+T+T·} ' ^{;;:: E, transmitting a second analytic report to the} second NF with a prediction of the network state x(t + T + Ti). In this example, E is a threshold that may be designed to ensure the reliability of the predictions. Step 1007 corresponds to steps 505 and 902. However, if a^f+r_,t+T+T_; < E, then the NWDAF only transmits l_j (t') to {j in step 1008. In both cases, upon receipt of either step 1007 or 1008 {j a decision about any actions to perform in the network (e.g. step 1009) based on the received information. Therefore, if the {j receives the indication of the state prediction x(t + T + Ti), it obtains additional and richer information about how the network state will change during the time {j may interact with the network. Figure 11 illustrates an example of the inputs and outputs of the ML model 1100 according to some embodiments. In this example, the network comprises 4 NFs, e.g. NF1, NF2, NF3, and NF4. The following indicates what these example NFs control and the performance metrics that may be changes by their actions in the network. In this example NF1: o Controls the PRB allocation per traffic slice; and o Impacts the traffic delay and the traffic load In this example NF2: o Is responsible for the traffic scheduling; and o Impacts the traffic delay and the traffic load In this example NF3: o Switches ON/OFF deployed cells/ antennas; and o Impacts the traffic load, delay, and energy consumption In this example NF4: o Manages the number of re-transmissions; and o Impacts link reliability In this example NFS: o Manages the antenna power transmission; and o Impacts link reliability and energy consumption. Based on the set of performance metrics that the various NFs' actions may impact, the NWDAF may group these example NFs into three groups. It will be appreciated that the NFs may be grouped such that a group of NFs affects the same or non-disjoint sets of performance metrics. For example, the 4 example NFs may be grouped as follows: - Group 1 = {NF1, NF2, NF3, NFS}. This group focus on the system traffic delay and load. - Group 2 = {NF3, NF 5}. This group is interested in energy consumption - Group 3 = {NF4, NFS}. This group is interested in link reliability. Accordingly, for every group of NFs, the NWDAF may produce specific trained ML model (e.g. as described with reference to Figure 5). The ML model for group 1 learns the network state variation during the time. Here, the network state comprises the traffic load and delay. Thus, the ML model will learn how the traffic delay and load will change based on analytic reports sent to NF1, NF2, NF3, and NFS. Also, the NWDAF may analyse how much, e.g. with what probability, every NFs is impacting the network state at different times. The network state for group 2 may the energy consumption during every time slot. The network state for group 3 may comprise link reliability at every time slot. Figure 11 in particular depicts the ML model 1100 for group 3 which comprises NF4 and NFS. The link reliability values for the times t to t+ T are input into the ML model as the network state information. The analytics information comprises the analytic reports transmits to the NF4 and NFS between the times t and t+ T. The ML model then outputs the probabilities a{t, where j e {4,5}, and these probabilities are used to determine predications of the link reliability from the times t to t+H. Returning to the description of Figure 10, consider if fi = NFS and fj = NF4. The analytic report transmitted in step 1004 may indicate that the link reliability will drop to 0.3 at t+4. This may be considered very low. The ML model may then predict that NFS will improve the link reliability (due to performing an action to increase antenna power transmission in step 1005) at t+S, to 0.9. After that, at t+1, in step 1006 NF4 asks for the same analytic report, i.e., the link reliability at t+4. In this example it is assumed that the probability of the prediction that NFS will improve the link reliability at t+S is greater than the predetermined threshold. Therefore, in this example, NWDAF will transmit (e.g. in step 1008) the default analytic report, i.e., the link reliability equals 0.3 at t+4 to NF4 along with the indication that the link reliability at t+S will increase to 0.9. In this case, therefore NF4 might not increase the number of re-transmissions in response to the received analytic report because NFS has already acted to improve the link reliability, and, in some examples, NFS has a shorter delay of the impacts. Figure 12 illustrates a network function 1200 comprising processing circuitry (or logic) 1201. The processing circuitry 1201 controls the operation of the network function 1200 and can implement the method described herein in relation to an network function 1200. The processing circuitry 1201 can comprise one or more processors, processing units, multi-core processors or modules that are configured or programmed to control the network function 1200 in the manner described herein. In particular implementations, the processing circuitry 1201 can a plurality of software and/or hardware modules that are each configured to perform, or are for performing, individual or multiple steps of the method described herein in relation to the network function 1200. It will be appreciated that the network function 1200 may comprise one or more virtual machines running different software and/or processes. The network function 1200 may therefore comprise, or be implemented in or as one or more servers, switches and/or storage devices and/or may comprise cloud computing infrastructure that runs the software and/or processes. Briefly, the processing circuitry 1201 of the network function 1200 is configured to perform the method as described herein with respect to an analytics network function (e.g. an NWDAF) or a second network function. In some embodiments, the network function 1200 may optionally comprise a communications interface 1202. The communications interface 1202 of the network function 1200 can be for use in communicating with other nodes, such as other virtual nodes. For example, the communications interface 1202 of the network function 1200 can be configured to transmit to and/or receive from other nodes requests, resources, information, data, signals, or similar. The processing circuitry 1201 of network function 1200 may be configured to control the communications interface 1202 of the network function 1200 to transmit to and/or receive from other nodes requests, resources, information, data, signals, or similar. The communications interface 1202 can use any suitable communication technology. Optionally, the network function 1200 may comprise a memory 1203. In some embodiments, the memory 1203 of the network function 1200 can be configured to store program code that can be executed by the processing circuitry 1201 of the network function 1200 to perform the method described herein in relation to the network function 1200. Alternatively or in addition, the memory 1203 of the network function 1200, can be configured to store any requests, resources, information, data, signals, or similar that are described herein. The processing circuitry 1201 of the network function 1200 may be configured to control the memory 1203 of the network function 1200 to store any requests, resources, information, data, signals, or similar that are described herein. The network function 1200 may be configured operate in the manner described herein in respect of an network function. Figure 13 is a block diagram illustrating an analytics network function 1300 according to some embodiments. The analytics network function 1300 transmit analytic reports to one or more network functions, NFs. The analytics network function 1300 comprises a receiving module 1302 configured to at a first time, t + T, receive a first request for an analytic report from a first NF, f_t , of one or more NFs. The analytics network function 1300 comprises a transmitting module 1304 configured to transmit a first analytic report to the first NF responsive to the first request. The analytics network function 1300 further comprises a using module 1306 configured to use time-series forecasting to determine a first probability value «^l _+T,t_+T+T; representative of a probability that a delay of an impact of the NF, f1, is equal to a first delay, Ti. The receiving module 1302 is further configured ^{to at a second time, t} + ^{t', receive a second request for an analytic report from a second} NF, f2, of the one or more NFs, where the second time, t + t', is earlier than a fourth time, t + T + h The transmitting module 1304 is further configured to responsive to the first probability value meeting a first criteria, transmitting a second analytic report to the second NF with an indication that a state of one or more performance metrics will be changed at a future time. The analytics network function 1300 may operate in the manner described herein in respect of an analytics network function (e.g. an NWDAF). Figure 13 is a block diagram illustrating an analytics network function 1300 according to some embodiments. The analytics network function 1300 can transmit analytic reports to one or more network functions, NFs. The analytics network function 1300 comprises a receiving module 1302 configured to at a first time, t + T, receive a first request for an analytic report from a first NF, f_t , of one or more NFs. The analytics network function 1300 comprises a transmitting module 1304 configured to transmit a first analytic report to the first NF responsive to the first request. The analytics network function 1300 further comprises a using module 1306 configured to use time-series forecasting to determine a first probability value «^l _+T,t_+T+T; representative of a probability that a delay of an impact of the NF, f1, is equal to a first delay, Ti. The receiving module 1302 is further configured to at a second time, t + t', receive a second request for an analytic report from a second ^{NF, f2, of the one or more NFs, where the second time, t} + ^{t', is earlier than a fourth} ^{time, t} + ^T + ^{h The transmitting module 1304 is further configured to responsive to the} first probability value meeting a first criteria, transmitting a second analytic report to the second NF with an indication that a state of one or more performance metrics will be changed at a future time. The analytics function 1300 may operate in the manner described herein in respect of an analytics network function (e.g. an NWDAF). Figure 14 is a block diagram illustrating an analytics network function 1400 according to some embodiments. The analytics network function 1400 can learn a delay of impact of one or more network functions' interactions with a network. The analytics network function 1400 comprises an input module 1402 configured to input a first time series into a machine learning, ML, model, wherein the first time series comprises: analytics information, L_t➔T • relating to whether an analytics report was transmitted to each of the one or more NFs between an initial time t and a first time t+ T, and state information, X_t➔T " related to one or more performance metrics in the network between the initial time ^{and the first time, t} + ^{T. The analytics network function 1400 comprises a output module} 1404 configured to output, from the ML model, for each of the one or more NFs, {;_, a plurality of probability values _af1h (t₁ :5 t₂ , t₁ e {t, ... ,t+T},t₂ e {t, ... ,t+H}), where the first time is earlier than a third time, t the plurality of probability values are each representative of a probability that the delay of the impact of the NF f_i is equal to t₂ - t₁ • The analytics network function 1400 may operate in the manner described herein in respect of an analytics network function (e.g. an NWDAF). Figure 15 is a block diagram illustrating a second network function 1500 according to some embodiments. The second network function 1500 can utilise analytic reports in a network. The second network function 1500 comprises a transmitting module 1502 configured to at a second time, t + t', transmit a second request for an analytic report to a analytics NF, where the second time, t + t', is earlier than a fourth time, t + T + T_i . The second network function 1500 comprises a receiving module 1504 configured to receive a second analytic report from the analytics NF comprising an indication that a state of one or more performance metrics in the network will be changed at the fourth time t + T + T_i . The second network function 1500 further comprises a determining module 1506 configured to determine whether to implement an action in the network based on the indication. The second network function 1500 may operate in the manner described herein in respect of a second network function. There is also provided a computer program comprising instructions which, when executed by processing circuitry (such as the processing circuitry 1201 of the network function 1200 described earlier), cause the processing circuitry to perform at least part of the method described herein. There is provided a computer program product, embodied on a non-transitory machine- medium, comprising instructions which are executable by processing circuitry to cause the processing circuitry to perform at least part of the method described herein. There is provided a computer program product comprising a carrier containing instructions for causing processing circuitry to perform at least part of the method described herein. In some embodiments, the carrier can be any one of an electronic signal, an optical signal, an electromagnetic signal, an electrical signal, a radio signal, a microwave signal, or a computer-readable storage medium. Embodiments described herein enable the NWDAF to understand how and when the network's state will change. It may also be able to detect which NFs are impacting the network's state. Despite the overlap between the NFs' interaction with the network, the NWDAF according to embodiments described herein may be able quantify the participation of every NF in the network's state transition, for example in terms of percentage or probability. For example, if two NFs simultaneously impact the traffic delay, NWDAF may be able to understand that the first affects the traffic delay with 70% and the second with 30%. When providing an NF with additional information about the network's state transition in the future, the NF may perform more efficiently, thereby improving the NFs' experience in the network. For example, if NF1 receives an analytical report about the traffic delay in 30 ms and NWDAF learns that the NF1 's delay of the impact is 2 s, that means that in 2 s, NF1 might improve the traffic delay, thereby getting better delay in 2 s. After that, NF2 asks for the same analytical report, i.e., traffic delay in 30 ms, the NWDAF may transmit additional information about the newly expected network state, i.e., the traffic delay in 2 s. As a result, NF2 may improve its interaction with the network. It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims. The word "comprising" does not exclude the presence of elements or steps other than those listed in a claim, "a" or "an" does not exclude a plurality, and a single processor or other unit may fulfil the functions of several units recited in the claims. Any reference signs in the claims shall not be construed so as to limit their scope.

Claims

CLAIMS 1. A computer-implemented method for transmitting analytic reports to one or more network functions, NFs, in a network, the method comprising: at a first time, t + T , receiving (801) a first request for an analytic report from a first NF, f_t , of one or more NFs, transmitting (802) a first analytic report to the first NF responsive to the first request, using (803) time-series forecasting to determine a first probability value af_+T,t_+T+Tj representative of a probability that a delay of an impact of the NF, f1, is equal to a first delay, Ti^; at a second time, t + t', receiving (804) a second request for an analytic report from a second NF, fa, of the one or more NFs, where the second time, t + t', is earlier than a fourth time, t + T + T_i ; and responsive to the first probability value meeting a first criteria, transmitting (805) a second analytic report to the second NF with an indication that a state of one or more performance metrics will be changed at a future time.

2. The method as claimed in claim 1 wherein the first probability value meets the first criteria if it is greater than or equal to a first predetermined threshold value.

3. The method as claimed in claim 1 or 2 wherein the step of using time-series forecasting to determine a first probability value af_+T,t_+T+Tj comprises: inputting a first time series into a machine learning, ML, model, wherein the first time series comprises: analytics information relating to whether an analytics report was transmitted to each of the one or more NFs between an initial time t and the first time t + T, and state information related to one or more performance metrics in the network between the initial time and the first time, t + T^; and outputting, from the ML model, for each of the one or more NFs, [;_, a p^{lurality of probability values} ai₁ ^,t ₂ (t1 $; t₂ , t₁ E {t, ... , t ⁺ T^}, t₂ E { t, ... , t + H}), where the first time, t + T, than a third time, t + H, representative of a probability that the delay of the impact of the NF _fi is equal to t₂ - t₁ ^.

4. The method, as claimed in claim 3, further comprising deriving a second time series comprising predicted values of the state information between the first time, t, and a third time, t + H, wherein the second time series is derived from the plurality of probability values for each of the one or more NFs.

5. The method, as claimed in claim 4, further comprising: training the ML model by updating one or more parameters of the ML model to minimise a loss function derived from the state information and the second time series.

6. A computer-implemented method for learning a delay of impact of one or more network functions' interactions with a network, the method comprising: Inputting (501) a first time series into a machine learning, ML, model, wherein the first time series comprises: analytics information, L_t➔T , relating to whether an analytics report was transmitted to each of the one or more NFs between an initial time t and a first time t + T, and state information, x_t➔T • related to one or more performance metrics in the network between the initial time and the first time, t + T; and outputting (502), from the ML model, for each of the one or more NFs, [;_, a p^{lurality of probability values} af₁.i_z (^t1 :5 ^t2 ^{, t}1 ^{e {t, ... , t +} T^}, t₂ e {t, ... , t ⁺ H}), where the first time is earlier than a third time_, t + H, wherein the plurality of probability values are each representative of a probability that the delay of the impact of the NF f_i is equal to _t2 - _t1.

7. The method as claimed in claim 6 further comprising: deriving a second time series, x_t➔H comprising predicted values, of the state information between the first time, t, and the third time, t + H, wherein, the second time series is derived from the plurality of probability values for each of the one or more NFs.

8. The method as claimed in claim 7, further comprising: training the ML model by one or more parameters of the ML model to minimise a loss function derived from the state information and the second time series.

9. The method as claimed in any one of claims 6 to 8 wherein the analytics information comprises: for each of the one or more NFs, an analytic report value for each time t ➔ t + T, wherein the analytics time value has a first value between X and Y inclusive, where X and Y are integer values, if the NF was transmitted an analytic report at a particular time.

10. The method as claimed in claim 9 wherein analytics value time value has a second value if the NF was not transmitted an analytics report at a particular time, where the second value is not between X and Y inclusive.

11. The method as claimed in any one of claims 6 to 10 wherein the ML model comprises one or more of: a CNN layer, a recurrent layer, and a temporal attention layer.

12. The method as claimed in any one of claims 6 to 11 further comprising: at the first time, t + T, receiving (801) a first request for an analytic report from a first NF, f1, of the one or more NFs, transmitting (802) a first analytics report to the first NF responsive to the first request, at a second time, t + t', receiving (803) a second request for an analytics report from a second NF, f2, of the one or more NFs,

13. The method, as claimed in claim 12 further comprising: responsive to a first probability value «l_+T,t_+T+T; meeting a first criteria, where a fourth time, t + T + T_i is later than the second time, t + t', and responsive to receiving the second request, transmitting an indication to the second NF that the state information will change from current state information.

14. The method as claimed in claim 13 wherein the first probability value meets the first criteria if it is greater than or equal to a first predetermined threshold value.

15. The method as claimed in claim 12 or 13 wherein the step of transmitting an indication comprises transmitting a predicted value of the state information at the fourth time to the second NF.

16. The method as claimed in any one of claims 13 to 15 further comprising receiving a subscription request from the second NF subscribing to receive predicted values of state information.

17. The method as claimed in any one of claims 6 to 16 wherein the one or more network functions' are grouped together according to which performance metrics the one or more network functions may change in the network.

18. A method, in a second network function, NF for utilising analytic reports in a network, the method comprising: at a second time, t + t', transmitting (901) a second request for an analytic report to an analytics NF, where the second time, t + t', is earlier than a fourth time, t + T + T_i receiving (902) a second analytic report from the analytics NF comprising an indication that a state of one or more performance metrics in the network will be changed at the fourth time t + T +T_i ; and determining (903) whether to implement an action in the network based on the indication.

19. An analytics network function (1200) for transmitting analytic reports to one or more NFs in a network, the analytics network function comprising processing circuitry (1201) configured to cause the analytics network function to: at a first time, t + T , receive (801) a first request for an analytic report from a first NF, /₁ , of one or more NFs, transmit (802) a first analytic report to the first NF responsive to the first request; use (803) time-series to determine a first probability value af_+r,t₊T₊T; representative of a probability that a delay of an impact of the NF, f1, is equal to a first delay, T_i ; at a second time, t + t', receive (804) a second request for an analytic report from a second NF, f2, of the one or more NFs, where the second time, t + t', is earlier than a fourth time, t + T + h and responsive to the first probability value meeting a first criteria, transmit (805) a second analytic report to the second NF with an indication that a state of one or more performance metrics will be changed at a future time_.

20. An analytics network function as claimed in claim 19 wherein the processing circuitry is further configured to cause the analytics network function to perform the method as claimed in any one of claims 2 to 5.

21. An analytics network function (1200) for learning a delay of impact of one or more network functions' interactions with a network, the analytics network function comprising processing circuitry (1201) configured to cause the analytics network function to: input (501) a first time series into a machine learning, ML, model, wherein the first time series comprises: analytics information, L_t➔T • relating to whether an analytics report was transmitted to each of the one or more NFs between an initial time t and a first time t + T, and state information, X_t➔r .. related to one or more performance metrics in the network between the initial time and the first time, t + T; and output (502), from the ML model, for each of the one or more NFs, Ii,. a p^{lurality of probability values} ai₁,t₂ (t1 :5; t2 , t1 ^{e {}t, ... , t ^{+ T}}, t₂ e {t, ... , t ⁺ H}), where the first time is third time, t + H, wherein the plurality of probability values are each representative of a probability that the delay of the impact of the NF f_i is equal to t₂ - t₁ ^.

22. The analytics network function as claimed in claim 21 wherein the processing circuitry is further configured to cause the analytics network function to perform the method as claimed in any one of claims 7 to 17.

23. A second network function, NF, for utilising analytic reports in a network, the second network function comprising processing circuitry ( 1201) configured to cause the second network function to: at a second time, t + t', transmit (901) a second request for an analytic report to a analytics NF, where the second time, t + t', is earlier than a fourth time, t + T + T_i receive (902) a second analytic report from the analytics NF comprising an indication that a state of one or more performance metrics in the network will be changed at the fourth time t + T +T_i ; and determine (903) whether to implement an action in the network based on the indication.

24. A computer program comprising instructions which, when executed on at least one processor, cause the at least one processor to carry out a method according to any of claims 1 to 18.

25. A computer program product comprising non transitory computer readable media having stored there on a computer program according to claim 24.