First node, communications system and methods performed thereby for handling one or more settings

Abstract

A computer-implemented method, performed by a first node (111) operating in a communications system (100). The first node (111) determines (402), jointly, one or more settings for a sleep mode of a radio network node (110) serving one or more groups of wireless devices (130) and a DL transmission power of the radio network node (110). The determining (402) is performed iteratively using machine learning, so that based on a respective target, one or more key performance indicators and energy saving, are improved over a first period of time. The first node (111) also outputs (403) one or more first indications indicating a result of the determination.

Description

[0001] FIRST NODE, COMMUNICATIONS SYSTEM AND METHODS PERFORMED THEREBY FOR HANDLING ONE OR MORE SETTINGS

[0002] TECHNICAL FIELD

[0003] The present disclosure relates generally to a first node and methods performed thereby for handling one or more settings. The present disclosure also relates generally to a communications system, and methods performed thereby for handling one or more settings. The present disclosure also relates generally to computer programs and computer-readable storage mediums, having stored thereon the computer programs to carry out these methods.

[0004] BACKGROUND

[0005] Computer systems in a communications network or communications system may comprise one or more nodes. A node may comprise a processing circuitry which, together with computer program code may perform different functions and actions, a memory, a receiving port, and a sending port. A node may be, for example, a server. Nodes may perform their functions entirely on the cloud.

[0006] The communications system may cover a geographical area which may be divided into cell areas, each cell area being served by a type of node, a network node in the Radio Access Network (RAN), radio network node or Transmission Point (TP), for example, an access node such as a Base Station (BS), e.g., a Radio Base Station (RBS), which sometimes may be referred to as e.g., gNB, evolved Node B (“eNB”), “eNodeB”, “NodeB”, “B node”, or Base Transceiver Station (BTS), depending on the technology and terminology used. The base stations may be of different classes such as e.g., Wide Area Base Stations, Medium Range Base Stations, Local Area Base Stations, and Home Base Stations, based on transmission power and thereby also cell size. A cell may be understood to be the geographical area where radio coverage may be provided by the base station at a base station site. One base station, situated on the base station site, may serve one or several cells. Further, each base station may support one or several communication technologies. The telecommunications network may also comprise network nodes which may serve receiving nodes, such as user equipments, with serving beams.

[0007] The standardization organization Third Generation Partnership Project (3GPP) is currently in the process of specifying a New Radio Interface called Next Generation Radio or New Radio (NR) or 5G-Universal Terrestrial Radio Access (UTRA), as well as a Fifth Generation (5G) Packet Core Network, which may be referred to as 5G Core Network (5GC), abbreviated as 5GC.

[0008] DRX The discontinuous reception mechanism (DRX) may be understood to be an approach aimed at reducing the energy consumption of a User Equipment (UE) by controlling the periods during which the UE radio may be active. In Figure 1, a schematic process of DRX and its associated power consumption levels are illustrated. At the commencement of a DRX cycle or a specified offset from it, a UE may awaken to monitor a channel, e.g., the Physical Downlink Control Channel, during an ON period. At the start of this ON period, the UE may initiate a timer with the length specified by an ON duration setting, and upon its expiration, it may transition into a dormancy phase, as depicted in Figure 1. If the UE receives data during an ON period, it may consume additional energy, shown by the solid black bar in Figure 1. After the end of data reception phase, the UE may start the inactivity timer and may await its expiration before entering the dormancy phase again. The dormancy phase may involve significantly lower energy consumption for the radio components.

[0009] Some existing methods [1] have enabled to enhance UE energy saving though optimizing DRX configurations for video streaming services. The core concept of such an approach may be understood to revolve around predicting channel quality to activate a UE exclusively during favorable channel conditions, allowing it to enter dormancy otherwise. This strategy may be understood to ensure that the UE may remain active only when optimal channel conditions may prevail, enabling quicker traffic serving and maximizing opportunities for UE sleep.

[0010] BS Sleeping

[0011] The concept of BS sleeping may be understood to be a widely employed strategy to mitigate energy consumption in a RAN. In this approach, BSs may have the capability to temporarily deactivate certain radio hardware components, such as power amplifiers, mixers, digital-to-analog converters, or even an entire Radio Frequency (RF) chain, resulting in energy savings. BS sleeping may be characterized by different modes, each distinguished by the extent of dormancy, the resultant energy conservation, and the latency required to reactivate the hardware components. Additionally, each sleep mode (SM) may be associated with a ramp-down / ramp-up energy. In Tables 5.1-3, 5.1-4, and 5.1-5 of 3GPP TR 38.864, v. 18.1.0 [2], 3GPP provides details of various BS SMs. The Tables 5.1-3, 5.1-4, and 5.1-5 of 3GPP, mentioned above highlight a fundamental tradeoff inherent in BS sleeping functionality. The greater the depth of the SM, the less energy may be understood to be consumed by the RAN. “Depth” may be understood to refer to the number of slept hardware components and for some components, it may be understood to refer to the magnitude of sleeping. However, this comes at the expense of an increased delay in switching back on the components that were put to sleep earlier, potentially leading to additional latency for UEs. Other existing methods [3] have employed a Machine Learning (ML) approach to determine optimal BS switching and SM design based on predictions of network traffic in specific time slots.

[0012] Most SM features may be defined at near-real-time, meaning that the decisions related to these modes may be commonly made every tens to hundreds of milliseconds. However, due to the unpredictive nature of cell traffic, there may be understood to often be an inadequacy of idle times to trigger deeper SMs, leading to marginal energy saving in practice.

[0013] A relevant aspect of BS sleeping may be understood to be the great impact of the deep_sleep mode, where an entire radio chain of the BS may be turned off, resulting in substantial energy savings. To gain a clearer understanding, the energy consumption of the BS may be divided into two components: load-dependent and load-independent. While other sleeping modes may primarily reduce the load-dependent part, often having limited impact on the load-independent component, deep_sleep stands out.

[0014] In the case of deep_sleep, not only may the load-dependent component energy consumption be understood to become negligible, reducing to 0, but the load-independent energy consumption component may also experience a significant reduction compared to other modes, approaching almost zero.

[0015] Machine Learning

[0016] Machine learning (ML) may be understood as the study of computer algorithms that may improve automatically through experience. It is seen as a part of Artificial Intelligence (Al). ML algorithms may build a model based on sample data, known as "training data", in order to make predictions or decisions without being explicitly programmed to do so. ML algorithms may be used in a wide variety of applications, such as email filtering and computer vision, where it may be difficult or unfeasible to develop conventional algorithms to perform the needed tasks.

[0017] There may be basically three types of ML Algorithms: Supervised Learning, Unsupervised Learning, and Reinforcement Learning (RL).

[0018] Supervised Learning algorithms may comprise a target / outcome variable, or dependent variable, which may have to be predicted from a given set of predictors, that is, independent variables. Using this set of variables, a function may be generated that may map inputs to desired outputs. The training process may continue until the model may achieve a desired level of accuracy on the training data. Once an ML model may have been trained, an inference process may begin, whereby new data may be run through the ML model to calculate an output. Examples of Supervised Learning may be Regression, Decision Tree, Random Forest, KNN, Logistic Regression etc.

[0019] In Unsupervised Learning algorithms, there may be no target or outcome variable to predict / estimate. It may be used for clustering a population into different groups, which may be widely used for segmenting customers in different groups for specific intervention. Examples of Unsupervised Learning may be K-means, mean-shift clustering, Density-Based Spatial Clustering of Applications with Noise (DBSCAN), Expectation-Maximization (EM) Clustering using Gaussian Mixture Models (GMM), Agglomerative Hierarchical Clustering, etc....

[0020] Cluster analysis or clustering may be understood as an ML technique which may comprise grouping a set of objects in such a way that objects in the same group, which may be called a cluster, may be understood to be more similar, in some sense, to each other than to those in other groups, that is, other clusters. It may be understood as a main task of exploratory data mining, and a common technique for statistical data analysis, used in many fields, including pattern recognition, image analysis, information retrieval, bioinformatics, data compression, computer graphics and ML.

[0021] Reinforcement learning (RL), which is schematically depicted in Figure 2, may be understood to be a type of ML where an agent 201 may learn to make decisions by taking actions 202 in an environment 203 to achieve some goal. The agent 201 may receive feedback in the form of rewards 204, which it may use to learn the best strategy, or policy, to accumulate the most reward 204 over time.

[0022] Reinforcement learning may involve the following. An agent 201 may be understood to be a learner or decision maker that may interact with the environment 203. The environment 203 may be understood to refer to the world that the agent 201 may interact with and learn from. A state (s) 205 may be understood to refer to a representation of the current situation that the agent 201 may be in. It may be understood to be the context within which the agent 201 may make decisions. An action (a) 202 may be understood to refer to a choice made by the agent 201 that may affect the state. A reward (r) : may be understood to refer to feedback from the environment 203 in response to the actions 202 taken by the agent. It may be a scalar signal that may indicate how well the agent 201 is doing at a given moment. A Policy may be understood to be a strategy used by the agent 201, which may map states 205 to actions 202. The policy may be deterministic, that is, always the same action 202 for a given state 205, or stochastic, that is, probabilistic actions 202 for a given state 205. A value function may be understood to refer to a function that may estimate how good it may be for the agent 201 to be in a given state 205, or how good it may be to perform a certain action 202 in a given state 205. The "goodness" may be typically measured as the expected return, e.g., the cumulative reward 204, that may be achieved. A Q-function, an Action-Value Function, may be understood to refer to a function that may estimate the value of taking a certain action 202 in a given state 205, and then following the current policy thereafter. A model of the environment 203 may be understood to refer to the fact that some RL approaches may involve learning a model that may predict the next state (s’) 206 and the reward 204 for the current state 205 and action 202. This may be understood to allow for planning and reasoning about the future without needing to actually take the action 202.

[0023] RL may involve making decisions in regard to exploration vs. exploitation. In reinforcement learning, the agent may need to balance exploration, that is, trying new things to discover better rewards, with exploitation, that is, using known information to maximize rewards. This may be understood to be a trade-off in RL.

[0024] In the reinforcement learning problem, the state 205 may change every time the agent 201 may apply a new action 202. The problem may be represented in the following way: The agent 201 may receive the state 205 of the environment 203 at a certain time (s). Then the agent 201 may select on action (a) 202 and apply it in the environment 203. When this action 205 is applied, the environment 205 may provide a reward (r) 204 and change to a new state (s’) 206, the reward 204 and state 206 may be provided finally by an interpreter. In reinforcement learning, the term "interpreter" may be used to describe a component of a reinforcement learning system that may interpret the state of the environment and the actions of an agent. It may be understood to be the part of the system that may bridge the agent 201 with the environment 203 it may be trying to learn from.

[0025] This cyclic procedure may be understood to bring a sequence of states, actions and rewards: s1, a1 ,r1 ;sT,aT,rT. The agent 201 may use different learning algorithms to learn the most appropriate action 202 to take on every different state 205 of the NW, e.g., policylearning based, such as actor-critic approaches, or value-based learning, such as deep-q networks.

[0026] SUMMARY

[0027] Certain aspects of the present disclosure and their embodiments address one or more of the challenges identified with the existing methods and provide solutions to these challenges or other challenges.

[0028] According to a first aspect of embodiments herein, the object is achieved by a computer-implemented method, performed by a first node. The first node operates in a communications system. The first node determines, jointly, one or more settings for a sleep mode of a radio network node serving one or more groups of wireless devices and a DL transmission power of the radio network node. The determining is performed iteratively using machine learning, so that based on a respective target, one or more key performance indicators and energy saving, are improved over a first period of time. The first node then outputs one or more first indications indicating a result of the determination.

[0029] According to a second aspect of embodiments herein, the object is achieved by a computer-implemented method, performed by the communications system. The communications system comprises the first node and a second node. The communications system determines, by the first node, jointly, the one or more settings for the sleep mode of the radio network node serving the one or more groups of wireless devices and the DL transmission power of the radio network node. The determining is performed iteratively with a first periodicity using machine learning, so that based on the respective target, the one or more key performance indicators and energy saving, are improved over the first period of time. The communications system outputs, by the first node, the one or more first indications indicating the result of the determination. The communications system determines, by the second node and based on the one or more first indications for a first number of first iterations: i) respective DRX configurations for the one or more groups of wireless devices, and ii) updated respective target one or more key performance indicators and energy saving. The determining by the second node is performed iteratively with a second periodicity and using machine learning, so that based on the respective target, the one or more key performance indicators and energy saving, are improved over a second period of time. The communications system obtains, by the first node, following the first number of first iterations, the updated respective target one or more key performance indicators and energy saving from the second node to be used in a subsequent second number of first iterations. The communications system then provides, by the second node, the determined respective DRX configurations for the one or more groups of wireless devices to the radio network node

[0030] According to a third aspect of embodiments herein, the object is achieved by the first node, configured to perform the method. The first node is configured to operate in the communications system. The first node is configured to determine, jointly, the one or more settings for the sleep mode of the radio network node configured to serve the one or more groups of wireless devices and the DL transmission power of the radio network node. The determining is configured to be performed iteratively using machine learning, so that based on the respective target, the one or more key performance indicators and energy saving, are improved over the first period of time. The first node is also configured to output the one or more first indications configured to indicate the result of the determination.

[0031] According to a fourth aspect of embodiments herein, the object is achieved by the communications system, configured to perform the method. The communications system is configured to comprise the first node and the second node. The communications system is configured to determine, by the first node, jointly, the one or more settings for the sleep mode of the radio network node configured to serve the one or more groups of wireless devices and the DL transmission power of the radio network node. The determining is configured to be performed iteratively with the first periodicity and using machine learning, so that based on the respective target, the one or more key performance indicators and energy saving configured to be obtained from the second node, are improved over the first period of time. The communications system is also configured to output, by the first node, the one or more first indications configured to indicate the result of the determination. The communications system is also configured to determine, by the second node and based on the one or more first indications for the first number of first iterations: i) the respective DRX configurations for the one or more groups of wireless devices, and ii) the updated respective target one or more key performance indicators and energy saving. The determining by the second node is configured to be performed iteratively with the second periodicity and using machine learning, so that based on the respective target, the one or more key performance indicators and energy saving, are improved over the second period of time. The communications system is further configured to obtain, by the first node, following the first number of first iterations, the respective target one or more key performance indicators and energy saving configured to be updated from the second node to be used in the subsequent second number of first iterations. The communications system is additionally configured to provide, by the second node, the respective DRX configurations for the one or more groups of wireless devices configured to be determined to the radio network node.

[0032] According to a fifth aspect of embodiments herein, the object is achieved by a computer program, comprising instructions which, when executed on at least one processing circuitry, cause the at least one processing circuitry to carry out the method performed by the first node.

[0033] According to a sixth aspect of embodiments herein, the object is achieved by a computer-readable storage medium, having stored thereon the computer program, comprising instructions which, when executed on at least one processing circuitry, cause the at least one processing circuitry to carry out the method performed by the first node.

[0034] According to a seventh aspect of embodiments herein, the object is achieved by a computer program, comprising instructions which, when executed on at least one processing circuitry, cause the at least one processing circuitry to carry out the method performed by the communications system.

[0035] According to an eighth aspect of embodiments herein, the object is achieved by a computer-readable storage medium, having stored thereon the computer program, comprising instructions which, when executed on at least one processing circuitry, cause the at least one processing circuitry to carry out the method performed by the communications system.

[0036] In a first aspect, embodiments herein may be understood to enable reducing energy consumption. Embodiments herein may be understood to introduce a new feature which may enable the provision of extra-low energy consumption for a selected group of wireless devices, achieved, for example, by allocating additional downlink power to them, with marginal or no extra overall energy consumption for the RAN. Services relying on wireless devices that may require low energy consumption may be understood to stand to gain significantly from embodiments herein, such as in the case of industrial loT devices.

[0037] Embodiments herein may be understood to introduce a novel application of Al in RAN to effectively reduce overall RAN energy consumption. Embodiments herein may use a new joint DRX coordination mechanism designed to minimize the energy consumption of wireless devices. Embodiments herein may offer energy saving targets at the second node as well as dynamic tuning of energy-saving parameters at first node cycles, guided by the target being set at the second node.

[0038] In a second aspect, embodiments herein may be understood to enable maintenance of service KPIs. Embodiments herein may ensure, in conjunction with energy savings, the preservation of service KPIs for wireless devices. This flexibility may be particularly beneficial in scenarios where wireless devices may experience temporary spikes in downlink traffic load, necessitating adjustments to parameters, such as ON_DU RATION or INACTIVITY_TIMER. These adaptations may lead to lower energy savings for the UE and fewer sleeping opportunities for the radio network node 110.

[0039] In a further aspect, embodiments herein may be understood to enable to support multiple service types, DRX Groups, and UE Groups. Embodiments herein may be understood to facilitate the management and optimization of diverse service types, DRX groups, and UE groups, providing a versatile and adaptable approach for various network scenarios and configurations.

[0040] BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Examples of embodiments herein are described in more detail with reference to the accompanying drawings, according to the following description.

[0042] Figure 1 is a schematic diagram illustrating an example of a process of DRX and its associated power consumption levels, according to existing methods.

[0043] Figure 2 is a schematic diagram illustrating RL, according to existing methods.

[0044] Figure 3 is a schematic diagram illustrating a non-limiting example of a communications system, according to embodiments herein.

[0045] Figure 4 is a flowchart depicting embodiments of a first method in a first node, according to embodiments herein.

[0046] Figure 5 is a flowchart depicting embodiments of a first method in a communications system, according to embodiments herein.

[0047] Figure 6 is a schematic diagram illustrating aspects of control cycles, according to embodiments herein.

[0048] Figure 7 is a signalling diagram illustrating a non-limiting example of a non-restrictive architecture of a multi-level optimization framework, according to embodiments herein.

[0049] Figure 8 is a schematic diagram illustrating a non-limiting example of a use case, according to embodiments herein. Figure 9 is a schematic diagram illustrating RAN energy saving which may be achieved according to embodiments herein.

[0050] Figure 10 is a signalling diagram depicting a non-limiting example of a method in a communications system, according to embodiments herein.

[0051] Figure 11 is a schematic block diagram illustrating a first node, according to embodiments herein.

[0052] Figure 12 is a schematic block diagram illustrating a second node, according to embodiments herein.

[0053] DETAILED DESCRIPTION

[0054] As part of the development of a embodiments herein, one or more problems with the existing technology will first be identified and discussed.

[0055] Existing methods have not addressed a simultaneous optimization of DRX configurations and RAN parameters.

[0056] For example, the approach mentioned in the Background revolving around predicting channel quality to activate a UE exclusively during favorable channel conditions, allowing it to enter dormancy otherwise may be understood to concentrate primarily on optimizing DRX configurations for UEs individually, and may not consider joint optimization of the UEs DRX and Radio Access Network (RAN) parameters. In fact, a BS may compensate for the performance drop due to the use of DRX by, among other approaches, increasing the downlink transmit power. Additionally, the existing methods referred to above [1] do not address the coordinated DRX settings for multiple UEs, potentially limiting its applicability in scenarios where joint optimization of DRX configurations for multiple UEs may be especially needed, e.g., in a scenario where RAN energy saving may be a relevant concern.

[0057] Proper design of DRX in UEs may be understood to allow a BS to enter deeper SMs and stay in sleeping modes for a longer period, resulting in extra energy savings. The surplus energy, thus saved, may be utilized to enhance downlink data rate for UEs experiencing temporary Key Performance Indicator (KPI) drops or unexpected increases in downlink traffic. This novel feature sets the approach of embodiments herein apart from existing studies on UE DRX design or BS sleeping mode design. In fact, the existing DRX studies mostly consider the RAN parameters as a predefined and fixed set of inputs and not part of the same design loop as DRX configurations. This leads to a substantial performance drop compared to cases where we jointly design these loops.

[0058] The existing BS sleeping studies take incoming cell traffic as an input and employ various models, including ML models, to predict near-future traffic and apply suitable BS sleeping strategies. However, when incoming traffic is random and uncontrollable, opportunities for sleeping, especially deep_sleep, become scarce. Introducing DRX into this design loop may be understood to enable the BS to shape cell traffic and create conditions favorable for deeper sleeping modes.

[0059] Existing products in RAN may provide features, e.g., a low-energy scheduler, which may be designed to reduce energy consumption by increasing the number of blank subframes where no traffic data may be transmitted. It has been demonstrated that blank subframes consume less energy, and therefore, more blank subframes may be understood to result in greater energy savings. Furthermore, such features may introduce a delay in downlink scheduling until the queued traffic data may meet specific requirements, offering an effective approach to reducing energy consumption. However, if DRX is enabled, such feature may encounter challenges.

[0060] A low-energy scheduler may be understood to aim to maximize the number of blank slots by delaying traffic data transmission until the traffic data conditions may meet specific thresholds. Such an approach may involve two thresholds: i) a timer, and ii) a buffer threshold. A low-energy scheduler may force the slot to be blank unless 1) the length of consecutive blank slots surpasses the timer, or 2) the number of bits in the downlink buffer surpasses the buffer threshold, or 3) there may be high-priority signals to be transmitted in the current slot, e.g., MSG2, Channel State Information- Reference Signal (CSI-RS), Synchronization Signal Block (SSB), Broadcast Control Channel (BCCH).

[0061] One challenge may be the feasibility of deeper sleeping modes. Shallow sleeping modes become the only viable option since the ON_PERIODs of uncoordinated UEs may extend across the majority of subframes. This leaves very few subframes available for the low energy scheduler approach to optimize. Notably, during the ON_PERIOD, UEs may be understood to be expected to receive data, and missing this opportunity may introduce an extra DRX_CYCLE delay to the downlink of the respective UE, which is not desirable.

[0062] Another challenge may be the exclusion of other RAN parameters and DRX configurations from the design loop. The design limitation may arise from the decision to keep other RAN parameters and DRX configurations out of the design loop. As pointed out earlier, the joint optimization of these variables may potentially yield substantial gains, a feature that is currently absent from existing products, particularly those using the low energy scheduler approach. The exclusion of these factors may limit the overall efficiency and performance optimization achievable through the existing methods.

[0063] It may be noted that if DRX is not enabled, e.g., ON_PERIOD is equal to DRX_CYCLE, devices may not be able to enter deeper SMs, thus resulting in high overall network energy consumption, regardless of how optimal existing products comprising the low energy scheduler approach may be. Certain aspects of the present disclosure and their embodiments address one or more of the challenges identified with the existing methods and provide solutions to the challenges discussed.

[0064] If a BS enters deep_sleep for a duration, which may be more than the required time for the involved hardware components to go to sleep and wake-up, it may subsequently transmit with considerably higher power in the following subframes when it may be active and still consume significantly less total energy overall. This observation forms the basis of the design of embodiments herein, leveraging the advantages of deep_sleep to optimize energy efficiency in BS operations.

[0065] Some embodiments herein may relate to a multi-level optimization of DRX configuration and RAN sleeping. More particularly, embodiments herein may relate to a comprehensive multi-layered decision-making system aimed at reducing energy consumption for both UEs and BSs, while simultaneously keeping service KPIs within their requirement thresholds.

[0066] As a summarized overview, embodiments herein may incorporate both higher-level (HL) and lower-level (LL) cycles, where a HL cycle may span multiple LL cycles, with each LL cycle potentially aligning with the duration of a single Physical Random Access Channel (PRACH) period.

[0067] At the higher level, a node, referred to herein as a “second node”, which may run a HL Orchestrator (HLO), and which may be hosted at the BS or edge cloud, may be responsible for jointly designing the DRX configurations for UEs, setting energy-saving targets for sleeping modes, and defining long-term service KPI objectives. Specifically, the second node may optimize DRX configurations to engineer more "consecutive" traffic idle time, allowing the BS to utilize sleeping functions more frequently and efficiently.

[0068] These decisions may then be communicated to another node, referred to herein as a first node, which may run a LL Orchestrator (LLO), and which, on a cycle- by- cycle basis, may assess a) the saved RAN energy due to sleeping modes and b) service KPIs on shorter time scales. The first node may dynamically adjust the downlink transmit power of individual UEs across various subframes to meet the targets specified by the second node within a single HL cycle. Coordinated DRX design may enable energy savings in LL cycles, which may be utilized in subsequent cycles where service KPIs for certain UEs may temporarily fall below the target.

[0069] Furthermore, the surplus energy may be strategically allocated to increase power for specific UEs while reducing their ON time via decreasing DRX ON_PERIOD and INACTIVITY_TIMER. This feature may enhance energy savings for these UEs without compromising their service KPIs, an aspect not currently feasible with existing

[0070] implementations. If the DRX parameters are found to be suboptimal and energy and service KPI budgets are not met, the second node may leverage feedback from LL cycles to adjust UE DRX parameters and establish new budget allocations for energy and KPIs in the subsequent LL cycles. This iterative and adaptive approach may ensure continuous optimization and improved performance in response to changing network conditions.

[0071] The present disclosure may assume that a communications system may comprise multiple services, each being served by a distinct set of wireless devices with identical DRX configuration, enabling targeted and effective connectivity for diverse applications.

[0072] The embodiments will now be described more fully hereinafter with reference to the accompanying drawings, in which examples are shown. In this section, embodiments herein are illustrated by exemplary embodiments. It should be noted that these embodiments are not mutually exclusive. Components from one embodiment or example may be tacitly assumed to be present in another embodiment or example and it will be obvious to a person skilled in the art how those components may be used in the other exemplary embodiments. All possible combinations are not described to simplify the description.

[0073] Figure 3 depicts two non-limiting examples, in panels “a” and “b” respectively, of a communications system 100, in which embodiments herein may be implemented. In some example implementations, such as that depicted in the non-limiting example of Figure 3 a) a, the communications system 100 may be a computer network. In other example implementations, such as that depicted in the non-limiting example of Figure 3 b), the communications system 100 may be implemented in a telecommunications system, sometimes also referred to as a telecommunications network, cellular radio system, cellular network, or wireless communications system. In some examples, the telecommunications system may comprise network nodes which may serve receiving nodes, such as wireless devices. The communications system 100 may for example be a network such as a 5G system, or a newer system supporting similar functionality. The telecommunications system may alternatively or additionally support other technologies such as, for example, Long-Term Evolution (LTE), e.g., LTE Frequency Division Duplex (FDD), LTE Time Division Duplex (TDD), LTE Half-Duplex Frequency Division Duplex (HD-FDD), or LTE operating in an unlicensed band. The telecommunications system may also support yet other technologies, such as Wideband Code Division Multiple Access (WCDMA), Universal Mobile Telecommunications System Terrestrial Radio Access (UTRA) TDD, Global System for Mobile communications (GSM) network, GSM / Enhanced Data Rate for GSM Evolution (EDGE) Radio Access Network (GERAN) network, Ultra-Mobile Broadband (UMB), EDGE, any combination of Radio Access Technologies (RATs) such as e.g. Multi-Standard Radio (MSR) base stations, multi-RAT base stations etc., any 3rd Generation Partnership Project (3GPP) cellular network, Wireless Local Area Network / s (WLAN) or WiFi network / s, Worldwide Interoperability for Microwave Access (WiMax), IEEE 802.15.4-based low-power short-range networks such as IPv6 over Low-Power Wireless Personal Area Networks (6LowPAN), Zigbee, Z-Wave, Bluetooth Low Energy (BLE), or any cellular network or system. The telecommunications system may for example support a Low Power Wide Area Network (LPWAN). LPWAN technologies may comprise Long Range physical layer protocol (LoRa), Haystack, SigFox, LTE for Machines (LTE-M), and Narrow-Band loT (NB-loT).

[0074] The communications system 100 comprises nodes, whereof a first node 111 and a second node 112 are depicted in Figure 3. The communications system 100 may comprise additional nodes. In some examples, the communications system 100 may comprise a third node 113, and / or a fourth node 114 which are not depicted in Figure 3 to simplify the Figure.

[0075] Any of the first node 111, the second node 112, the third node 113 and the fourth node 114 may be understood, respectively, as a first computer system or server, a second computer system or server, a third computer system or server and a fourth computer system or server. Any of the first node 111, the second node 112, the third node 113 and the fourth node 114 may be implemented as a standalone server in e.g., a host computer in the cloud 115, as depicted in the non-limiting example of Figure 3b), for the first node 111 and the second node 112. In other examples, any of the first node 111, the second node 112, the third node 113 and the fourth node 114 may be a distributed node or distributed server, such as a virtual node in the cloud 115, and may perform some of its respective functions locally, e.g., by a client manager, and some of its functions in the cloud 115, by e.g., a server manager. In other examples, any of the first node 111, the second node 112, the third node 113 and the fourth node 114 may perform its functions entirely on the cloud 115, or partially, in collaboration or collocated with a radio network node. Yet in other examples, any of the first node 111 , the second node 112, the third node 113 and the fourth node 114 may also be implemented as processing resources in a server farm.

[0076] Yet in other examples, any of the first node 111 , the second node 112, the third node 113 and the fourth node 114 may also be implemented as virtual network functions, e.g., according to a Network Functions Virtualization (NFV) Architecture.

[0077] Any of the first node 111, the second node 112, the third node 113 and the fourth node 114 may be under the ownership or control of a service provider or may be operated by the service provider, or on behalf of the service provider.

[0078] Both of the first node 111 and the second node 112 may have a capability to perform machine-implemented learning procedures, which may be also referred to as “machine learning” (ML). In some examples, any of the first node 111, the second node 112, the third node 113 and the fourth node 114 may have a capability to perform machine-implemented learning procedures. The communications system 100 may in some examples, comprise one or more radio network nodes, such as radio network node 110, depicted in Figure 3. The radio network node 110 may be, e.g., comprised in a Radio Access Network of the telecommunications system. That is, the radio network node 110 may be a transmission point such as a radio base station, for example a gNB, an eNB, or any other network node with similar features capable of serving a wireless device, such as a user equipment or a machine type communication device, in the communications system 100. In typical examples, the radio network node 110 may be a base station, such as a gNB or an eNB. In other examples, the radio network node 110 may be a distributed node, such as a virtual node in the cloud 115, and may perform its functions entirely on the cloud 115, or partially, in collaboration with a radio network node.

[0079] The telecommunications system may cover a geographical area, which in some embodiments may be divided into cell areas, wherein each cell area may be served by a radio network node 110, although, one radio network node 110 may serve one or several cells. In the example of Figure 3 b), the radio network node 110 serves a cell 121. The radio network node 110 may be of different classes, such as, e.g., macro eNodeB, home eNodeB or pico base station, based on transmission power and thereby also cell size. In some examples, the radio network node 110 may serve receiving nodes with serving beams. The radio network node 110 may be directly connected to one or more core networks.

[0080] Any of the first node 111, the second node 112 and the radio network node 110, and / or any of the other nodes comprised in the communications system 100 may support one or several communication technologies, and its name may depend on the technology and terminology used.

[0081] One or more wireless devices may be comprised in the telecommunication network, which may comprise one or more groups of wireless devices 130, e.g., a set of M UEs, clustered in G groups, comprising a first wireless device 131 as depicted in the non-limiting example of Figure 3 b). Any of the wireless devices comprised in the one or more groups of wireless devices 130, such as the first wireless device 131 may be a wireless communication device such as a 5G UE, or a UE, which may also be known as e.g., mobile terminal, wireless terminal and / or mobile station, a Customer Premises Equipment (CPE) a mobile telephone, cellular telephone, or laptop with wireless capability, just to mention some further examples. Any of the wireless devices comprised in the one or more groups of wireless devices 130, such as the first wireless device 131 comprised in the telecommunications system may be, for example, portable, pocket-storable, hand-held, computer-comprised, or a vehicle-mounted mobile device, enabled to communicate voice and / or data, via the RAN, with another entity, such as a server, a laptop, a Personal Digital Assistant (PDA), or a tablet, Machine-to-Machine (M2M) device, device equipped with a wireless interface, such as a printer or a file storage device, modem, sensor, loT device, or any other radio network unit capable of communicating over a radio link in a communications system. Any of the wireless devices comprised in the one or more groups of wireless devices 130, such as the first wireless device 131 comprised in the telecommunications system may be enabled to communicate wirelessly in the telecommunications system. The communication may be performed e.g., via a RAN, and possibly the one or more core networks, which may be comprised within the telecommunications system. Each of the wireless devices comprised in the one or more groups of wireless devices 130, such as the first wireless device 131 may be equipped with DRX functionality.

[0082] It may be understood that the telecommunications network may comprise additional radio network nodes and / or additional wireless devices than those depicted in Figure 3. In Figure 3 b, the one or more groups of wireless devices 130 are schematically depicted as comprising two groups of three wireless devices each. This may be understood to be a nonlimiting example. In other examples, there may be more or less groups of wireless devices 130 and each may comprise fewer or additional wireless devices than those depicted in Figure 3b).

[0083] In particular examples of embodiments herein, the communications system 100 may comprise the following components.

[0084] The first node 111 and the second node 112 may be each run different parts of a Hierarchical Reinforcement Learning (RL) Decision-Maker.

[0085] The second node 112 may run a HLO which may take actions, such as, e.g., setting a) ON_PERIOD, b) DRX_CYCLE time, c) INACTIVITY-TIMER, and d) OFFSET. Additionally, e) target energy-saving goals and f) target service KPIs for each service, served by a group of the one or more groups of wireless devices 130, may be determined for a cycle of the second node 112. It may be noted that these actions may be defined for one of the one or more groups of wireless devices 130 serving a specific service and with a designated DRX configuration. Nevertheless, in a general form, these actions may be tailored for individual wireless devices as well.

[0086] The first node 111 may run an LLO which may take actions related to downlink transmit power adjustments toward each of the one or more groups of wireless devices 130. Moreover, the first node 111 may tune hyper-parameters of a low-energy-scheduler or parameters of a sleeping algorithm of the radio network node 110, based on the long-term targets set by second node 112 for a cycle of the second node 112.

[0087] In some examples, the third node 113 may run another component of the communications system 100, which may be an Intent Management Function (IMF). The second node 112 and the first node 111 may utilize a set of intents provided by the third node 113, guiding their decision-making processes for the action space. The radio network node 110 and the one or more groups of wireless devices 130 may be comprised in yet another component of the communications system 100, which may be an Environment of the RL. The RL environment may take the new actions set by the second node 112 and the first node 111 as input and may provide statistics on energy savings at both the radio network node 110 and the wireless devices 130 on LL and HL levels. Short-term service KPIs may also be returned for every new action at cycles of the first node 111.

[0088] In some examples, the fourth node 114 may run a further component of the communications system 100, which may be a Two-level Data Processing Unit. The Data Processing Unit (DPU) may operate at two levels. At cycles of the first node 111 , the fourth node 114 may collect statistics from the environment to generate rewards and states, feeding them back to the first node 111 for subsequent actions. At the end of the cycle of the second node 112, it may process statistics collected over past cycles of the first node 111 within that cycle of the second node 112, to generate a higher-level reward and state. These higher-level parameters may assist the second node 112 in determining the next set of actions. The fourth node 114 may provide short-term and long-term predictions for the cell traffic loads to be used for both the second node 112 and the first node 111 decision making systems.

[0089] The first node 111 may be configured to communicate within the communications system 100 with the second node 112 over a first link 151 , e.g., a radio link, or a wired link. The first node 111 may be configured to communicate within the communications system 100 with the radio network node 110 over a second link 152, e.g., a radio link, or a wired link. The second node 112 may be configured to communicate within the communications system 100 with the radio network node 110 over a third link 153, e.g., a radio link, or a wired link. The radio network node 110 may be configured to communicate within the communications system 100 with any of the wireless devices in the one or more groups of wireless devices 130 over a respective fourth link 154, e.g., a radio link. Only the respective fourth link 154 between the radio network node 110 and the first wireless device 131 is depicted in Figure 3 b) in order to simplify the Figure. The first node 111 may be configured to communicate within the communications system 100 with the fourth node 114 over a fifth link, e.g., a radio link, or a wired link, which is not depicted in Figure 3. The radio network node 110 may be configured to communicate within the communications system 100 with the fourth node 114 over a respective sixth link, e.g., a radio link, which is not depicted in Figure 3.

[0090] Any of the first link 151, the second link 152, the third link 153 and the respective fourth link 154 may be a direct link or may be comprised of a plurality of individual links, wherein it may go via one or more computer systems or one or more core networks in the communications system 100, which are not depicted in Figure 3, or it may go via an optional intermediate network. The intermediate network may be one of, or a combination of more than one of, a public, private or hosted network; the intermediate network, if any, may be a backbone network or the Internet; in particular, the intermediate network may comprise two or more sub-networks, which is not shown in Figure 3.

[0091] In general, the usage of “first”, “second”, “third” and / or “fourth” herein may be understood to be an arbitrary way to denote different elements or entities, and may be understood to not confer a cumulative or chronological character to the nouns they modify.

[0092] Some of the embodiments contemplated herein will now be described more fully with reference to the accompanying drawings. Other embodiments, however, are contained within the scope of the subject matter disclosed herein, the disclosed subject matter should not be construed as limited to only the embodiments set forth herein; rather, these embodiments are provided by way of example to convey the scope of the subject matter to those skilled in the art.

[0093] Some embodiments herein may relate to a method performed by the first node 111. Some embodiments herein may relate to a method performed by the communications system 100.

[0094] Embodiments of a computer-implemented method, performed by the first node 111, will now be described with reference to the flowchart depicted in Figure 4. The first method may be understood to be for handling one or more settings. The first node 111 operates in the communications system 100.

[0095] In some examples, the communications system 100 may be a 5G network.

[0096] Several embodiments are comprised herein. In some embodiments, all the actions may be performed. In some embodiments, one or more of the actions may be performed. It should be noted that the examples herein are not mutually exclusive. One or more embodiments may be combined, where applicable. All possible combinations are not described to simplify the description. Components from one embodiment may be tacitly assumed to be present in another embodiment and it will be obvious to a person skilled in the art how those components may be used in the other exemplary embodiments. A non-limiting example of the first method performed by the first node 111 is depicted in Figure 4. In a particular non-limiting example of the first method, Action 402 and Action 403 may be performed.

[0097] In Figure 4, optional actions are represented with dashed lines.

[0098] Action 401

[0099] In this Action 401 , the first node 111 may obtain one or more first indicators of one or more key performance indicators (KPIs). The one or more key performance indicators may be of one or more of: at least one of the one or more groups of wireless devices 130, the first wireless device 131 of the one or more groups of wireless devices 130, and a first service. The one or more groups of wireless devices 130 may be the set of M UEs, clustered in G groups. Each wireless device in the or more groups of wireless devices 130 may be equipped with DRX functionality. The wireless devices in the or more groups of wireless devices 130 may be operating in RRC_CONNECTED mode. This may be understood to oblige them to adhere to a timing of a Physical Random-Access Channel (PRACH) configuration period of the cell 121, e.g., defined as a maximum of 160 ms in Table 8.1-1 of 3GPP TS 38.213, 18.4.0.

[0100] The first wireless device 131 may be referred to herein as UE m.

[0101] The one or more KPIs may be understood to indicate performance. The one or more KPIs may be, for example, latency, cell throughout, network energy consumption, UEs energy consumption, etc.

[0102] The one or more KPIs may be obtained from the one or more groups of wireless devices 130, e.g., indirectly, via the radio network node 110 and, in some examples, the fourth node 114.

[0103] The service may be understood to be a communication service. That is, a specific type of network service designed to fulfill particular communication requirements in terms of bandwidth, latency, reliability, and other KPIs.

[0104] As a non-limiting example, the first node 111 may obtain, in this Action 401 , statistics of the KPIs, e.g., latency, average bit rate, etc. of the wireless devices, including but not limited to the average and standard deviation, for each of the one or more groups of wireless devices 130 g, i.e. , Vg e [C],

[0105] In this Action 401 , the first node 111 may also obtain one or more second indicators of a traffic in the cell 121 served by the radio network node 110, a single BS, where the one or more groups of wireless devices 130 may be operating.

[0106] The one or more second indicators of the traffic may be defined in terms of a rate of incoming objects or packets in a cell, size of a downlink buffer of the radio network node 110 averaged over a certain time window, etc. An example of the one or more indicators of the traffic may be average traffic at the cell 121.

[0107] The one or more second indicators of the traffic may be obtained from the radio network node 110, e.g., indirectly, in some examples, via the fourth node 114.

[0108] In this Action 401 , the first node 111 may further obtain one or more third indicators of energy usage by one or more of: the radio network node 110, the at least one of the one or more groups of wireless devices 130, and the first wireless device 131 of the one or more groups of wireless devices 130.

[0109] The one or more third indicators of the energy usage may be, e.g., for the radio network node 110: an energy consumption, or saving, measurement, prediction or indicator.

[0110] There may be two scenarios for the one or more groups of wireless devices 130, and / or the wireless device 131, i) if they share energy consumption (or saving) measurements, predictions or indicators, or otherwise ii) a wireless device energy model or ML energy predictor may be used instead.

[0111] The one or more third indicators of the energy usage by the radio network node 110 may be, for example, total energy saved at the radio network node 110 due to the sleeping modes, and / or extra energy consumption at the radio network node 110 due to surplus power allocation to some of the one or more groups of wireless devices 130.

[0112] The one or more third indicators of the energy usage by the at least one of the one or more groups of wireless devices 130, and the first wireless device 131 of the one or more groups of wireless devices 130 may be, for example, total energy saving of wireless device(s) due to DRX for each UE group g, that is, fg e [C],

[0113] The one or more third indicators of the energy usage may be obtained from the at least one of the one or more groups of wireless devices 130, and the first wireless device 131 of the one or more groups of wireless devices 130 e.g., indirectly, and the radio network node 110, e.g., indirectly, via the radio network node 110 and, in some examples, the fourth node 114.

[0114] The one or more third indicators of the energy usage by the at least one of the one or more groups of wireless devices 130, and the first wireless device 131 of the one or more groups of wireless devices 130 may be obtained from the at least one of the one or more groups of wireless devices 130, and the first wireless device 131 of the one or more groups of wireless devices 130 according to multiple approaches, such as that described e.g., in WO2022229233A1, based on various levels of privacy at the wireless devices.

[0115] At any time, and of the wireless devices in the one or more groups of wireless devices 130, e.g., the first wireless device 131, may be in one of the following states shown in Table 1.

[0116] Table 1 is based on Tables 18 and 19 of 3GPP TR 38.840, version 16.0.0, and shows UE states and relative power consumptions.

[0117] The first wireless device 131 may be taken as an example of any of the wireless devices in the one or more groups of wireless devices 130. Any statement herein made in reference to the first wireless device 131, orto a / the wireless device, or to wireless devices, unless otherwise noted, may be understood to equally apply to any of the wireless devices in the one or more groups of wireless devices 130.

[0118] At each state, the first wireless device 131 may consume some surplus energy compared to the deep_sleep mode. Each OFF state may also have two additional parameters: transition energy and transition time, which may be understood to characterize costs of going to and coming out from the respective sleeping mode. It may be noted that while Table 1 is based on Tables 18 and 19 of 3GPP TR 38.840 version 16.0.0, different vendors may have their own values. Knowing these values may be understood to help the first node 111, and as will be described later in Figure 5, the second node 112, to better orchestrate the parameters of the first wireless device 131. However, in the worst case scenario, when the vendor of the first wireless device 131 may not share such information, a RAN operator may use standard models such as the one depicted in Table 1.

[0119] Sleeping Relative power Transition Transition

[0120] Sleeping mode UE state mode code consumption energy Time

[0121] UE-S1 Deep sleep 1 450 20

[0122] UE-S2 Light sleep 20 100 6 OFF state UE-S3 Micro sleep 45 0 0

[0123] PDCCH only, or

[0124] UE-S4 SSB / CSI-RS 100 N / A

[0125] proc, only

[0126] Active reception,

[0127] PDCCH +

[0128] UE-S5 300 N / A

[0129] PDSCH. ON state ACK / NACK

[0130] Active

[0131] transmission,

[0132] UE-S6 700 N / A

[0133] Long PUCCH or

[0134] PUSCH

[0135]

[0136] Table 1.

[0137] The obtaining 401 in this Action 401 may be understood to be part of an ML learning process by the first node 111, which may be understood to be iterative. An iteration or cycle may comprise, among others, collection of data, as described in this Action 401, and a determination, as described in the next Action 402, based on the collected data. Accordingly, the obtaining 401 in this Action 401 may be performed for every iteration, referred to herein as, “first iteration”, and prior to every determining as described in the next Action 402.

[0138] In one example, the wireless devices of the one or more groups of wireless devices 130 may be required to report their energy-saving values or indicators at every first iteration or a collective number of first iterations. This may enable the radio network node 110 to optimize DRX configurations based on the reported data. The process may involve a request for such a report from the first node 111 and subsequent adjustments to the DRX configuration.

[0139] In a particular non-limiting example of Action 401, for every first iteration, the first node 111 may obtain: a) total energy saving of wireless devices due to DRX for each of the one or more groups of wireless devices 130 g (i.e. , fg e [<7]), b) total energy saved at the radio network node 110 due to the sleeping modes, c) extra energy consumption at the radio network node 110 due to surplus power allocation to some of the one or more groups of wireless devices 130, d) statistics of the KPIs, latency, average bit rate, etc., of the wireless devices including but not limited to the average and standard deviation, for each of the one or more groups of wireless devices 130 g, i.e., Vg e [C], and e) average traffic in the cell 121.

[0140] Action 402

[0141] In this Action 402, the first node 111 determines, jointly, one or more settings for a sleep mode of the radio network node 110 serving the one or more groups of wireless devices 130 and a DL transmission power of the radio network node 110.

[0142] The sleep mode may be understood to be one of the possible states for the radio network node 110, which may be defined based on 3GPP TR 38.864, version 18.1.0, similarly to the states of the first wireless device 131 shown in T able 1. This technical report may be understood to also characterize models and numbers for the transition time and energy for each OFF state of a BS such as the radio network node 110, e.g., deep_sleep, light_sleep, or micro_sleep or similar names that various vendors may use.

[0143] The determining in this Action 402 is performed iteratively using machine learning, so that based on a respective target, the one or more key performance indicators and energy saving, are improved over a first period of time.

[0144] The settings, which may be considered to be decision variables in this Action 402, may be, e.g., hyperparameters of various thresholds characterizing an energy saving algorithm of the radio network node 110, including but not limited to a) a threshold on the maximum time a packet may stay in the downlink buffer without being transmitted, b) a threshold on the maximum permissible downlink buffer size of the radio network node 110 while being in a sleeping state, c) a threshold on the minimum resource utilization, e.g., measured by physical resource block (PRB) in LTE and symbol-level resource utilization in 5G NR, requirement to activate the transmitter, etc.

[0145] The one or more of the settings for the DL transmit power of the radio network node 110, which may be considered to be a first decision variable (LL1) may optimize the maximum downlink transmit power of all wireless devices in a group g, denoted by pmgk. When a service may require higher energy saving via reduced: i) ON-duration within one long cycle of UE group g, where g e [ff] (ON_PERIOD), referred to herein as vg, and ii) duration within which wireless devices associated with the g th group may remain ON after receiving a PDCCH (INACTIVITY_TIMER), referred to herein as ug, the first node 111 may increase their allocated transmit power to maintain a certain average transmission rate toward them. Increased transmission power may often translate into an increased transmission rate. There may often be a model to calculate estimated rate or equivalently estimated PRB utilization for each wireless device as a function of the channel condition of the channel condition, number of packets, transmit power and the selected adaptive modulation and coding scheme.

[0146] In another example, this power may be fed into a power allocation algorithm that may return an allocated power to each wireless device in a subframe, e.g., based on a power level set by the first node 111, as well as instantaneous channel conditions.

[0147] The DL transmit power may be understood to be one of the setting that may be optimized. Alternatively, or even additionally, the determining in this Action 402 of the one or more settings for the DL transmit power may involve modulation and coding scheme (MCS) selection, which may impact the allocation of resource blocks for any given wireless device. This way, the first node 111 may not need to change the transmit power per wireless device.

[0148] The one or more of the settings for the sleep mode of the radio network node 110 may also be referred to herein as BS sleeping model decision maker, or energy saver scheduler, hyper-parameter, which may be considered to be a second decision variable (LL2). Every sleeping algorithm, or low-energy scheduler, of a base station such as the radio network node 110, may be characterized by at least one hyper-parameter determining a tradeoff between energy saving and service KPI degradation. One example may be a queue threshold to activate scheduling. This may be understood to refer to a threshold on the downlink buffer size to activate the scheduler or keep staying in some sleeping mode. The radio network node 110 may stay in sleeping and accumulate incoming traffic / packets in its downlink buffer and activate the transmitter to empty the buffer only when the total accumulated bits may exceed a certain threshold. The longer the threshold, the more delay the wireless devices may experience, but the more energy saving may be experienced by the RAN, e.g., by the radio network node 110. Similar parameters may be defined for other sleeping mode algorithms.

[0149] The ML used in embodiments herein may be an RL algorithm. However, other ML procedures may be possible.

[0150] The determining in this Action 402 for a respective first iteration may be based on the respective one or more first indicators, one or more second indicators and one or more third indicators for the respective first iteration, obtained in Action 401. Respective may be understood to refer to the indicators collected for any given iteration. However, as stated earlier, in one example, the wireless devices of the one or more groups of wireless devices 130 may be required to report their energy-saving values or indicators at a collective number of first iterations.

[0151] The respective target one or more key performance indicators and energy saving may have been obtained from the second node 112 operating in the communications system 100, as will be described in relation to Figure 5.

[0152] In some embodiments, the determining in this Action 402 may be based on the radio network node 110 using a low energy scheduler. Action 403

[0153] In this Action 403, the first node 111 outputs one or more first indications indicating a result of the determination.

[0154] Outputting may be e.g., sending, e.g., to the second node 112 or to another node, displaying on an interface of the first node 111, etc.

[0155] Action 404

[0156] In some embodiments, at least one first first indication of the one or more first indications for the respective first iteration may initiate performing an action in the environment comprising the radio network node 110 and the one or more groups of wireless devices 130, implementing the determined one or more settings for the sleep mode and the DL transmission power of the radio network node 110.

[0157] Initiating may be understood as starting itself, or enabling, triggering or facilitating that another node may start the action. The action space may comprise triggering an implementation of the determined one or more settings for the sleep mode of the radio network node 110 serving the one or more groups of wireless devices 130, LL2, and the DL transmission power of the radio network node 110, LL1.

[0158] It may be noted that the first node 111 may not directly optimize the sleeping decisions of the radio network node 110, and it may be assumed that the algorithm describing the sleeping selection of the radio network node 1110 may be given to the environment. Instead, first node 111 may only optimize its hyper-parameters, e.g., what may be the queue size in an algorithm controlling a scheduler targeting low energy consumption to start scheduling wireless devices.

[0159] In another example, the framework of embodiments herein may be extended to include other decision variables including BS sleeping optimization or change the DL transmission power of the radio network node 110, LL, 1 from power allocation to allocation of a number of resource blocks to each wireless device or the one or more groups of wireless devices 130.

[0160] In some of such embodiments, in this Action 404, the first node 111 may obtain, for every first iteration, a first reward obtained from the environment in response to the implemented action. The first reward may be based on one or more first utility functions. The determining, as described in Action 401, for a subsequent first iteration immediately following the respective first iteration may be based on the obtained first reward.

[0161] Any of the one or more first utility functions may be understood to be a measurable function that may be optimized by embodiments herein. Example utility functions may be average network energy saving during a period of time, service KPIs, average cell throughput, etc.. A first utility function may be defined for each of the one or more groups of wireless devices 130, g, as the energy saving, as well as service KPIs intended for that group.

[0162] For the wireless device energy saving component of this utility function, according to a first group of examples (a), the first node 111 may rely on one or more of the one or more third indicators of the energy usage by the at least one of the one or more groups of wireless devices 130, and the first wireless device 131 of the one or more groups of wireless devices 130, being obtained from the wireless devices 130. As mentioned earlier, multiple approaches for such feedback may be defined, as described e.g., in WO2022229233A1 , based on various levels of privacy at the wireless devices 130.

[0163] For the wireless device energy saving component of this utility function, according to a second group of examples (b), the first node 111 may use the energy model of 3GPP TR 38.840, version 16.0.0, to compute an approximate for the energy saving potential of the wireless devices 130, and use that in the calculations by the first node 111, or by another node, e.g., the fourth node 114.

[0164] It may be noted that the actual energy-saving potential of a wireless device may depend on both, the duration of the dormancy period within the DRX cycle and the specific implementation of the wireless device. Embodiments herein may be understood to offer a capability to optimize this dormancy period across a plurality of the wireless devices 130. And that may be enough to allow wireless devices to reduce their radio energy consumption.

[0165] Further gain for wireless devices may be achieved if they share their implementation details, which may be unlikely due to privacy and intellectual property reasons. It may be understood to be important to note that even in the absence of detailed UE implementations and models, embodiments herein may still leverage simple energy indication feedback based on energy consumption data from the wireless devices, as mentioned in the first group of examples (a) above. In the worst-case scenario, where neither the energy consumption model of the wireless device nor energy consumption indications may be available, embodiments herein may use either the approach of the second group of examples (b) above or just utilize the dormancy periods of each wireless device, which may be available in the radio network node 110, as a proxy for energy-saving potential. In this case, the higher the dormancy period, the greater the energy saving of the wireless device.

[0166] If some of the wireless devices 130 have very different energy saving performance with the same DRX configurations, e.g., due to different wireless device implementation, in an alternative example, the one or more groups of wireless devices 130 may be based on similarities in wireless device implementations, among other parameters. This may be understood to ensure that the energy-saving capabilities within each group of wireless devices 130 may be more uniform. A similar logic may also be applied to various radio network node implementations. A second utility function may be defined as a RAN utility function for the radio network node 110 and as the average saved energy during one cycle, corresponding to a second iteration by the second node 112, which will be described in reference to Figure 5, and which may be referred to herein as an HL cycle. This may be a function of the sleeping algorithm used by the radio network node 110, the scheduler, being a low energy scheduler or other scheduler, as well as DRX configurations of the one or more groups of wireless devices 130. The third node 113, which may be intent management function (IMF) and operations support system (OSS), may provide the following information before running the training algorithm inside the first node 111, and, as will be described later in reference to Figure 5, inside the second node 112.

[0167] Otherwise, the information described as provided by the third node 113 may be otherwise obtained by any of the first node 111 and the second node 112 from other, e.g., internal sources. According to a first option, the third node 113 may provide a) parameters describing service constraints: such as minimum KPI requirements in terms of for example latency or throughput. According to a second option, the third node 113 may provide b) parameters describing constraints on the actions, including feasible configurations of DRX as well as maximum feasible length for DRX cycle. For example, since embodiments herein may be performed while the wireless devices 130 may be in RRC_CONNECTED mode, DRX cycles may not be able to extend beyond the PRACH configuration period, which may be as specified in Release 17 as 160 ms, e.g., based on Table 8.1-1 of 3GPP TS 38.213, version 18.4.0,. If there are some constraints on the hyper-parameters of the sleeping modes for action LL2, the sleep mode of the radio network node 110 serving the one or more groups of wireless devices 130, they may need to also be registered to the orchestrator as well. According to a third option, the third node 113 may provide c) parameters describing sleeping modes and sleeping algorithms of the wireless devices 130 and the radio base station 110. For example, if the radio base station 110 may need to be ON at some certain subframes for some reasons, this may be incorporated in the sleeping algorithm. According to a fourth option, the third node 113 may provide d) parameters describing protocol limitations. For example, signals such as Random Access CHannel (RACH) Period, Channel Quality Indicator (CQI) periodicity, CSI-RS periodicity, and Synchronization Signal Block (SSB) periodicity may need to be aligned to occur within the coordinated DRX ON duration, with a period equal to the DRX cycle duration. By informing the first node 111 and the second node 112 about these limitations, embodiments herein may ensure such signals, e.g., RACH Period, CQI periodicity, CSI-RS periodicity, and nr SS periodicity, may be adjusted to occur in the (coordinated) DRX ON duration, with a period that may be equal to DRX cycle duration.

[0168] The first reward may be a linear combination of the first utility functions (ZUE) of the one or more groups of wireless devices 130 and the second utility functions for the radio network node 110, that is, the utility RAN utility functions (fes), e.g., a fuE + (1- cr) fes for a given constant a e [0,1], A multi-objective optimization framework may be also used here. The details about the aforementioned utility functions and their relation to wireless device or cell KPIs and energy saving may be provided by the third node 113 or other network functions.

[0169] Action 405

[0170] As stated earlier, the respective target one or more key performance indicators and energy saving may have been obtained from the second node 112 operating in the communications system 100.

[0171] In this Action 405, the first node 111 may obtain, following a first number of first iterations, updated respective target one or more key performance indicators and energy saving from the second node 112, to be used in a subsequent second number of first iterations.

[0172] The first node 111 may iterate the obtaining of Action 401 , the determining of Action 402, the outputting of Action 403, the obtaining of Action 404 and the obtaining of Action 405 until an obtained first reward may exceed a threshold for a number of iterations. That is, until the first node 111 may have learned to determine the best one or more actions consistently over the number of iterations.

[0173] Embodiments of a computer-implemented method performed by the communications system 100, will now be described with reference to the flowchart depicted in Figure 5. The method may be understood to be for handling one or more settings. The communications system 100 comprises the first node 111 and the second node 112.

[0174] The method may comprise the following actions. Several embodiments are comprised herein. In some embodiments, the method may comprise all the actions. In other embodiments, the method may comprise one or more actions. One or more embodiments may be combined, where applicable. All possible combinations are not described to simplify the description. It should be noted that the examples herein are not mutually exclusive.

[0175] Components from one example may be tacitly assumed to be present in another example and it will be obvious to a person skilled in the art how those components may be used in the other examples. In Figure 5, optional actions are depicted with dashed lines. In particular embodiments, Action 502, 503, 506, 508 and 509 may be performed.

[0176] The detailed description of some of the following corresponds to the same references provided above, in relation to the actions described for the first node 111 and will thus not be repeated here to simplify the description. For example, in some examples, the settings may hyperparameters. Action 501

[0177] In this Action 501, the communications system 100 may obtain, by the first node 111, for every first iteration, and prior to every determining as described in the next Action 502 performed by the first node 111 , the following.

[0178] The first node 111 may obtain the one or more first indicators of the one or more KPIs. The one or more key performance indicators may be of the one or more of: at least one of the one or more groups of wireless devices 130, the first wireless device 131 of the one or more groups of wireless devices 130, and the first service.

[0179] The first node 111 may also obtain the one or more second indicators of the traffic in the cell 121 served by the radio network node 110 where the one or more groups of wireless devices 130 may be operating.

[0180] The first node 111 may further the one or more third indicators of energy usage by the one or more of: the radio network node 110, the at least one of the one or more groups of wireless devices 130, and the first wireless device 131 of the one or more groups of wireless devices 130.

[0181] This Action 501 may be understood to correspond to Action 401 described in relation to Figure 4.

[0182] Action 502

[0183] In this Action 502, the communications system 100 determines, by the first node 111, jointly, the one or more settings for the sleep mode of the radio network node 110 serving the one or more groups of wireless devices 130 and the DL transmission power of the radio network node 110.

[0184] The determining in this Action 502 is performed iteratively with a first periodicity using machine learning, so that based on the respective target, the one or more key performance indicators and energy saving, are improved over the first period of time.

[0185] The determining in this Action 502 for the respective first iteration may be based on the respective one or more first indicators, one or more second indicators and one or more third indicators for the respective first iteration, obtained in Action 501.

[0186] The respective target one or more key performance indicators and energy saving may have been obtained from the second node 112 operating in the communications system 100.

[0187] The first periodicity may comprise a plurality of first cycles, also referred to herein as first iterations. One first iteration may span up to one PRACH and may comprise one or multiple connected mode DRX (C-DRX) cycles.

[0188] In some embodiments, the determining in this Action 502 by the first node 111 may be based on the radio network node 110 using a low energy scheduler. This Action 502 may be understood to correspond to Action 402 described in relation to Figure 4.

[0189] Action 503

[0190] In this Action 503, the communications system 100 outputs, by the first node 111 , the one or more first indications indicating the result of the determination.

[0191] This Action 503 may be understood to correspond to Action 403 described in relation to Figure 3.

[0192] Action 504

[0193] In some embodiments, at least the one first first indication of the one or more first indications for the respective first iteration may initiate performing the action in the environment comprising the radio network node 110 and the one or more groups of wireless devices 130, implementing the determined the one or more settings for the sleep mode and the DL transmission power of the radio network node 110.

[0194] In this Action 504, the communications system 100 may obtain, by the first node 111, for every first iteration, the first reward obtained from the environment in response to the implemented action. The first reward may be based on the one or more first utility functions. The determining, as described in Action 501 , for the subsequent first iteration immediately following the respective first iteration may be based on the obtained first reward.

[0195] This Action 504 may be understood to correspond to Action 404 described in relation to Figure 4.

[0196] Action 505

[0197] In some embodiments, a second periodicity (K) may comprise second iterations by the second node 112, and for every second iteration, the method may further comprise, in this Action 505, that the communications system 100 may obtain, by the second node 112, one or more second indications indicating statistics within a respective first number of iterations. The one or more second indications may indicate the one or more first indications for the respective first number of first iterations.

[0198] The one or more first indications for the respective first number of first iterations may comprise statistics over the first number of first iterations of one or more of: the one or more first indicators, the one or more second indicators and the one or more third indicators. The one or more second indications may therefore reflect how the one or more first indicators, the one or more second indicators and the one or more third indicators may evolve based on the actions implemented by the first node 111 on the environment over the first number of first iterations. The second periodicity may comprise a plurality of second cycles, also referred to herein as second iterations. One second iteration may comprise K first iterations, where K may be a hyper-parameter according to examples of embodiments herein. That is, one of the settings.

[0199] Action 506

[0200] In this Action 506, the communications system 100 determines, by the second node 112 and based on the one or more first indications for the first number of first iterations: i) respective DRX configurations for the one or more groups of wireless devices 130, and ii) updated respective target one or more key performance indicators and energy saving. That is, updated respective target one or more KPIs and updated respective target energy saving.

[0201] The determining in this Action 506 by the second node 112 is performed iteratively with the second periodicity and using machine learning, so that based on the respective target, the one or more key performance indicators and energy saving, are improved over a second period of time.

[0202] In some embodiments, the determining in this Action 506 by the second node 112 may be based on the obtained one more second indications.

[0203] The respective DRX configurations may be kept the same for all first cycles within the same second cycle

[0204] The second node 112 may perform the determining in this Action 506 using the following decision variables.

[0205] A first decision variable of the second node 112, HL1, may be the respective DRX configurations for the one or more groups of wireless devices 130 and may involve one or more of the following parameters.

[0206] One parameter may be Tg, which may be understood to be DRX cycle length of wireless device group g, where g e [G] (DRX_CYCLE and subframe time). This parameter may be set for individual wireless devices as well if each wireless device group may comprise only one wireless device. It may also include service sets by grouping wireless devices based on the target service.

[0207] Another parameter may be vg, which may be understood to be ON-duration within one long cycle of wireless device group g, where g e [G] (ON_PERIOD).

[0208] A further parameter may be xg, which may be understood to be Offset in long cycle to start the ON duration for wireless device group g, where g e [G] (OFFSET).

[0209] Yet another parameter may be ugwhich may be understood to be a duration within which wireless devices associated with the gth group may remain ON after receiving a PDCCH (INACTIVITY_TIMER).

[0210] These configurations may be applied via Radio Resource Control (RRC). A second decision variable of the second node 112, HL2, may be understood to be a respective target energy saving, that is, the target energy saving goals. The respective target energy saving may comprise energy saving targets per group of wireless devices 130. Through this parameter, the radio network node 110 may be able to prioritize energy saving of some groups of wireless devices 130 over others and set higher targets for them, based on some predefined agreements or the service and / or type of wireless device, e.g., UE type. The respective target energy saving may additionally or alternatively comprise an energy saving target due to RAN sleeping.

[0211] A decision variable of the second node 112, HL3, may be understood to be the respective target one or more key performance indicators, that is, the target long-term service KPIs for each service, one or more groups of wireless devices 130, and / or individual wireless devices, such as the first wireless device 131.

[0212] The actions of the second node 112, as determined in this Action 506, may need to satisfy a few constraints. One set of constraints may be service requirements of all wireless device and / or DRX groups that may need to be met. The requirements may be as given in (a) in Action 404, via the third node 113. Another set of constraints may be that the DRX decision variables (Tg, Vg, Ug,00Vg) may have to be in the set of specified values by 3GPP or other predefined set of values, which may be informed via (b) as described above in Action 404. Another constraint may be that vg< Tg00Vg, that is, for all g. Yet another constrained may be that xg< vg- rg,00Vg.

[0213] In some embodiments, the determining in this Action 506 by second node 112 may be based on the radio network node 110 using a low energy scheduler.

[0214] In some embodiments, one or more of the following may apply: a) the determining in Action 502 by the first node 111 may be based on the radio network node 110 using a low energy scheduler, and b) the determining in this Action 506 by second node 112 may be based on the radio network node 110 using the low energy scheduler.

[0215] Action 507

[0216] In some embodiments, the second periodicity may comprise the second iterations by the second node 112, and for every second iteration, the method may further comprise, in this Action 507, that the communications system 100 may provide, by the second node 112, the updated respective target one or more key performance indicators and energy saving to the first node 111.

[0217] It may be noted that the action space of embodiments herein may comprise LL1, LL2, LL3, HL1, and HL2 as actions on various levels. The communications system 100 may not directly optimize the sleeping decisions of the radio network node 110 and may assume that the algorithm describing sleeping selection by the network node 110 may be given to the environment. Instead, the communications system 100 may only optimize its hyper- parameters, such as what the queue size may be in a scheduler, such as a low energy scheduler, to start scheduling wireless devices 130.

[0218] In other examples, the framework of embodiments herein may be extended to include other decision variables including sleeping optimization of the radio network node 110, or change LL1 from power allocation to allocation of the number of resource blocks to each wireless device or one or more groups of wireless devices 130.

[0219] Action 508

[0220] In this Action 508, the communications system 100 obtains, by the first node 111, following the first number of first iterations, the updated respective target one or more key performance indicators and energy saving from the second node 112 to be used in a subsequent second number of first iterations.

[0221] This Action 508 may be understood to correspond to Action 404 described in relation to Figure 4.

[0222] Action 509

[0223] In this Action 509, the communications system 100 provides, by the second node 112, the determined respective DRX configurations for the one or more groups of wireless devices 130 to the radio network node 110.

[0224] The radio network node 110 may establish communication with the wireless devices in the one or more groups of wireless devices 130 through radio resource control (RRC) messages to implement the determined respective DRX configurations, that is, to manage DRX parameters. The PRACH configuration period, CSI-RS periodicity, and SSB periodicity may be aligned with the selected DRX cycle length.

[0225] Action 510

[0226] In some embodiments, the second periodicity may comprise the second iterations by the second node 112, and for every second iteration, the method may further comprise, in some embodiments, in this Action 510, that the communications system 100 may obtain, by the second node 112, for every second iteration, a second reward. The second reward may be understood to be a reward for an action taken by the second node 112.

[0227] The second reward may be obtained from the environment in response to the provided updated respective target one or more key performance indicators and energy saving and determined respective DRX configurations. The second reward may be based on one or more second utility functions. The determining in Action 506 by the second node 112 for a subsequent second iteration immediately following the respective second iteration may be based on the obtained second reward. The second reward may be a linear combination of the utility functions of the radio network node 110, the RAN utility functions (fes), and the utility functions of the one or more groups of wireless devices 130, that is the UE groups utility functions (ZUE), e.g., a fuE + (1- a) fBSfor a given constant a e [0,1], A multi-objective optimization framework may be also used here. The details about the aforementioned utility functions and their relation to wireless device or cell KPIs and energy saving may be provided by the third node 113, e.g., OSS or IMF or other network functions.

[0228] According to examples of embodiments herein, the communications system 110 may apply a new action to the environment and feed the collected observations to the third node 113, e.g., the DPU. The third node 113 may process the first reward and a next state as a result of a first action by the first node 111, and pass them to the first node 111 for the next action. At the end of every second cycle, that is, of every cycle of the second node 112, the third node 113 may provide a second reward and the next state of the higher timescale to the second node 112 for a new policy. Some of the models required for predictions, e.g., traffic prediction or energy saving models of the wireless devices 130 may be retrieved from another node, e.g., a Network Data Analytics Function (NWDAF).

[0229] In other examples, the third node 113 may provide estimations of the channel quality for the next first cycle and feed them to the first node 111. The first node 111 may use such information to optimize the downlink transmit power of each of the one or more groups of wireless devices 130, not only based on their target KPIs and remained power budget, but also the channel quality.

[0230] In some embodiments, the communications system 100 may iterate the actions just described until the obtained second reward may exceed a certain threshold for a given number of iterations.

[0231] Some embodiments herein will now be further described with some non-limiting examples, which may be combined with the embodiments just described.

[0232] In the following description, any reference to a / the LLO, simply a / the “LL” may be understood to equally refer the first node 111; any reference to a / the HLO, or simply “HL”, may be understood to equally refer to the second node 112; any reference to a / the DPU may be understood to equally refer the third node 113; any reference to a / the OSS and / or a / the IMF may be understood to equally refer to the fourth node 114; any reference to a / the “network” or simply a / the “NW” may be understood to equally refer to the communications system 100; any reference to a / the “UE” and / or “UEs” may be understood to equally refer to any of the wireless devices 130. Figure 6 is a schematic diagram illustrating the first periodicity and the second periodicity, that is the control cycles, according to embodiments herein. Some embodiments herein may be understood to use a two-level control periodicity or loop or cycle: the first periodicity, comprising the first iterations, also referred to herein as LL cycles, and the second periodicity, comprising the second iterations, also referred to herein as HL cycles. Figure 6 illustrates both cycles. One LL cycle may span up to one PRACH and may contain one or multiple connected mode DRX (C-DRX) cycles, that is N DRX cycles. One HL cycle may comprise K LL cycles, where K may be understood to be a hyper-parameter of the framework of embodiments herein. One HL cycle may be understood to correspond to NK DRX cycles. Two DRX cycles are schematically represented at the bottom of the figure. The description of the DRX cycles as depicted may be understood to correspond to that provided in Figure 1.

[0233] Figure 7 is a schematic diagram illustrating a non-limiting example of the architecture of the multi-level optimization framework of embodiments herein and its relationship with Reinforcement Learning at a high level. Each of the first node 111 and the second node 112 may be understood to run RL agents, LLO and HLO, respectively, which may be trained, according to embodiments herein, using the architecture illustrated in Figure 7. Figure 7 illustrates how in some examples, observations from the environment comprising the wireless devices 130 and the radio network node 110 may be obtained by the third node 113, a DPU in this example. The first node 111, according to Action 401 and 501 may obtain a LL state of the environment: the one or more first indicators of the one or more KPIs of the one or more groups of wireless devices 130, the one or more second indicators of the traffic in the cell 121 and a remained energy budget for this second cycle as the one or more third indicators of energy usage via the third node 113. The first node 111 may then, according to Action 402 and Action 502, determine, jointly, the one or more settings for the sleep mode and the DL transmission power of the radio network node 110, that is, the LL actions. The first node 111 may then, according to Action 403 and Action 503, initiate implementing the determined settings in the environment. As a result, according to Action 403 and Action 503, the first node 111 may obtain the first reward, depicted as LL reward. The second node 112, according to Action 505 may obtain the one or more second indications indicating statistics of service KPIs and energy saving as well as a long-term cell load prediction, that is a state of the environment for the second node 112, depicted as HL state. Then, according to Action 506, the second node 112 may determine the actions to take on the environment as the respective DRX configurations for the one or more groups of wireless devices 130, and on the first node 111, as the updated respective target one or more KPIs and energy saving goals for the one or more groups of wireless devices 130 and the radio network node 110. The second node 112 then, according to action 507, may provide the updated respective target one or more KPIs and energy saving to the first node 111. In accordance with Action 405 and Action 508, the first node 111 may then obtain, the updated respective target one or more KPIs and energy saving from the second node 112 to be used in a subsequent second number of first iterations. The second node 112 may also, according to action 508, provide the determined respective DRX configurations for the one or more groups of wireless devices 130 to the radio network node 110. The second node 110 may finally obtain a second reward, depicted as HL reward, in accordance with Action 510.

[0234] Figure 8 is a schematic diagram illustrating a non-limiting example of a use case for the methods disclosed in this document, according to embodiments herein. The non-limiting example depicted in Figure 8 illustrates a use case wherein a coverage cell 801 operating on a certain frequency, and served by a first radio network node 802 is overlaid by two capacity booster cells 803, 804 with non-overlapping coverage areas operating on a different frequency, and served by a second radio network node 805 and a third radio network node 806, respectively. Without loss of generality, it may be assumed that the capacity booster cells 802, 803 may have much lower downlink transmit power than the coverage cell 801. In this case, the UEs served by one of the capacity booster cells may be power-limited, that is, an increase in downlink transmit power for those UEs may cause an increase in the channel capacity for them and, thus, an increase in their KPIs, and a reduction in their energy consumption. In this scenario, proper configuration of their DRX parameters may reduce their energy consumption without impacting their KPIs. The scope and the application of the embodiments herein may be understood to be very general and may be applied to any other scenario where there may exist users with diverse services, and at least one BS which may have at least one energy saving mode in short timescale, in the order of hundreds of milliseconds in accordance to 3GPP specifications.

[0235] Figure 9 is a schematic diagram illustrating experimental results obtained using methods according to embodiments herein, simulated in a 3GPP-compliant simulator. Figure 9 shows the achieved RAN energy saving in various simulated scenarios. Specifically, the following features have been implemented in the 3GPP-compliant simulator: a low energy scheduling approach (1), and UEs with DRX capability. Figure 9 depicts the RAN energy saving compared to a baseline with the low energy scheduling (1) approach and users without DRX capability. The low energy scheduling (1) approach-only (1-only) shows the performance with no UE DRX, but with RAN energy saving based on [1], In another low energy scheduling (1) approach, the users may be understood to have DRX capability, where the parameters may be configured using legacy behavior. This is depicted in Figure 9 as “(l)-DRX-uncoor”. In yet another low energy scheduling (1) approach, the DRX parameters may be adjusted based on embodiments herein. This is depicted in Figure 9 as “(l)-DRX-coor”. The simulations have been performed using energy models from 3GPP for both UEs, according to 3GPP TR 38.840, version 16.0.0, and BS 3GPP TR 38.864, version 18.1.0. Other simulation parameters are reported in Table 2. As depicted in Figure 9, the low energy scheduling (1) approach has been observed to be able to increase RAN energy savings by almost 17%. Surprisingly, the addition of DRX capability to the UEs in the group denoted as “1-DRX-uncoor”, without jointly optimizing its parameters, as depicted in the outer loop of Figure 7, results in a substantial drop in energy savings. The energy saving drops nearly to zero. This may be due to inefficient handling of DRX traffic by the low energy scheduling (1) approach, which may keep the scheduler active almost all the time, leading to no sleeping opportunity for RAN. Although the final number may be specific for the low energy scheduling (1) approach, other BS sleeping algorithms, other than the one enabled by the low energy scheduling (1) approach, may face a similar challenge in the face of uncoordinated DRX UEs. As may be observed in from Figure 9, by jointly optimizing the DRX configuration of the UEs, in the group denoted as “1-DRX-coor”, the same level of RAN energy savings may be regained, while also enabling UEs to enjoy some energy savings through DRX. It may be noted that in this experiment, the transmit power for UEs was not optimized. By incorporating such feature, further improvements in KPIs may be anticipated.

[0236]

[0237] Parameter Value

[0238] Deployment I gNB

[0239] Bandwidth 20MHz

[0240] Duplex TDD

[0241] Carrier

[0242] Frequency 3500 MHz

[0243] Pattern DDDUU

[0244] UL / DL transmit 0.2 / 40 Watt

[0245] power

[0246] # gNB / UE

[0247] 2 / 2

[0248] antennas

[0249] Multi path 30 km / h

[0250] speed

[0251] Numerology 1 (30 KHz ss)

[0252]

[0253] Table 2 Figure 10 illustrates a sequence diagram showing a non-limiting example of an implementation of the method performed by the communications system 100, according to embodiments herein. Particularly, the sequence diagram illustrates the multi-level optimization of the DRX configuration for one group of wireless devices 130, an arbitrary UE group, indicated as UE_g with the required signaling between different components: the first node 111, depicted as LLO, the second node 112, depicted as HLO, the radio network node 110, depicted as gNB and the fourth network node 114 depicted as a DPU. Starting in panel a) of Figure 10, each of the wireless devices in the group may wait for the RACH, send a PRACH Preamble according to Msg1 to the radio network node 110, receive a RACH response according to Msg 2 from the radio network node 110, send a Msg 3 including an RRC setup request to the radio network node 110, receive an RRC setup in response from the network node 110, including a preliminary C-DRX configuration, and in response, send an RRC Setup Complete message to the network node 110. It may be noted that in the operational phase, the preliminary RRC setup, may be set via the trained RL model. Once in RRC Connected mode, the method performed by the second node 112, indicated in Figure 9 as “HL optimization”, may loop with the second periodicity. Inside the second periodicity, the method performed by the first node 111, indicated as “LL optimization” may loop K times. Within every first cycle, the wireless devices 130 may send their state “state_g” indicating the statistics on the KPIs of the group of wireless devices 130 to the radio network node 110. The radio network node 110 may then report the state of the environment, comprising the group of wireless devices 130 and the radio network node 110, to the third node 113. The state may comprise the state_g, the cell traffic statistics (stats) and the energy metric statistics. In one example, the wireless devices 130 may be required to report their energy-saving values or indicators at every LL cycle or a collective number of LL cycles. This may enable the radio network node 110 to optimize DRX configurations based on the reported data. The process may involve a request for such a report and subsequent adjustments to the DRX configuration. The third node 113 may then send the state of the environment indicating the statistics of the first cycle to the first node 111, in accordance with Action 401 and Action 501. The first node 111 may then, in accordance with Action 402 and Action 502, determine, jointly, the one or more settings for the sleep mode and the DL transmission power of the radio network node 110. The first node 111 may then, according to Action 403 and Action 503, output the one or more first indications indicating the maximum (Max) DL transmit power and sleep mode settings of the radio network node 110 “BS SM settings” to be implemented by the radio network node 110. The wireless devices 130 may then send a reward (reward_g) back to the radio network node 110 according to a UE groups utility function (f_UE). The radio network node 110 may in turn report a reward to the third node 113 based on the reward of the group and Ran utility functions. The third node 113 may report the first reward with the first cycle to the first node 111 in accordance with Action 404 and Action 504. Continuing on panel b) of Figure 10, within the second cycle of the second node 112, the second node 112 may receive, according to Action 510, a second reward for a previous action (-1) implemented on the environment by the second node 112, e.g., computed over the last K first cycles. The second node 112 may also obtain the state of the environment from the third node 113 in accordance with Action 505, indicating the statistics within the second cycle. The second node 112 may then determine, in accordance with Action 506, the respective DRX configurations for the one or more groups of wireless devices 130, and the updated respective target one or more KPIs and energy saving. The second node 112 may then in accordance with Action 507 and Action 508, send the target energy saving goals and the target long-term service KPIs to the first node 111. The second node 112 may also provide, in agreement with Action 509, the determined respective C-DRX configurations for the one or more groups of wireless devices 130 to the radio network node 110. In response, the radio network node 110 may send an RRC Reconfiguration message to the wireless devices 130, including the C-DRX configuration. The wireless devices 130 may send back an RRC Reconfiguration Complete message. The second node 112 may know a result of implementing the determined respective C-DRX configurations for the one or more groups of wireless devices 130, that is, the second node 112 may obtain the second reward, after the next K of cycles performed by the first node 111, in agreement with Action 510.

[0254] As a summarized overview of the foregoing, embodiments herein in a first group of embodiments may be understood to relate to a new multi-level orchestration mechanism at the radio network node 110 that may be understood to optimize jointly DRX configurations as well as long-term and short-term power allocations to wireless devices to ensure target levels of service KPIs and energy saving. Such problem formulation, joint DRX and RAN parameters optimization in multiple levels and control cycles cannot be addressed with the existing methods.

[0255] Certain embodiments disclosed herein may provide one or more of the following technical advantage(s), which may be summarized as follows.

[0256] In a first aspect, embodiments herein may be understood to enable reducing energy consumption. Embodiments herein may be understood to introduce a new feature which may enable the provision of extra-low energy consumption for a selected group of wireless devices, achieved, for example, by allocating additional downlink power to them, with marginal or no extra overall energy consumption for the RAN. Services relying on wireless devices that may require low energy consumption may be understood to stand to gain significantly from embodiments herein, such as in the case of industrial loT devices. Embodiments herein may be understood to introduce a novel application of Al in RAN to effectively reduce overall RAN energy consumption. Embodiments herein may use a new joint DRX coordination mechanism designed to minimize the energy consumption of wireless devices. Embodiments herein may offer energy saving targets at the second node 112 as well as dynamic tuning of energy-saving parameters at first node 111 cycles, guided by the target being set at the second node 112.

[0257] In a second aspect, embodiments herein may be understood to enable maintenance of service KPIs. Embodiments herein may ensure, in conjunction with energy savings, the preservation of service KPIs for wireless devices at both higher (longer-term statistics) and lower levels (short-term statistics). This flexibility may be particularly beneficial in scenarios where wireless devices may experience temporary spikes in downlink traffic load, necessitating adjustments to parameters such as ON_DURATION or INACTIVITY_TIMER. These adaptations may lead to lower energy savings for the UE and fewer sleeping opportunities for the radio network node 110.

[0258] In a third aspect, embodiments herein may be understood to enable multi-layer decision making. Embodiments herein may implement a multi-layer architecture that may significantly reduce the signalling required to calculate the state and reward. The second node 112 may change its action infrequently, once after many first cycles of the first node 111, eliminating the need for constant feedback or calculating reward / state at every first cycle. This approach may be understood to result in substantial efficiency gains over traditional one-level optimization methods, enhancing speed and scalability.

[0259] In a fourth aspect, embodiments herein may be understood to enable support multiple service types, DRX Groups, and UE Groups. Embodiments herein may be understood to facilitate the management and optimization of diverse service types, DRX groups, and UE groups, providing a versatile and adaptable approach for various network scenarios and configurations.

[0260] Figure 11 depicts an example of the arrangement that the first node 111 may comprise to perform embodiments herein. The first node 111 may be understood to be for handling the one or more settings. The first node 111 is configured to operate in the communications system 100.

[0261] Several embodiments are comprised herein. It should be noted that the examples herein are not mutually exclusive. One or more embodiments may be combined, where applicable. All possible combinations are not described to simplify the description.

[0262] Components from one embodiment may be tacitly assumed to be present in another embodiment and it will be obvious to a person skilled in the art how those components may be used in the other exemplary embodiments. The detailed description of some of the following corresponds to the same references provided above, in relation to the actions described for the first node 111 and will thus not be repeated here. For example, the settings may be configured to be hyperparameters.

[0263] In Figure 11, optional components are depicted with dashed lines.

[0264] The first node 111 is configured to determine, jointly, the one or more settings for the sleep mode of the radio network node 110 configured to serve the one or more groups of wireless devices 130 and the DL transmission power of the radio network node 110. The determining is configured to be performed iteratively using machine learning, so that based on the respective target, the one or more key performance indicators and energy saving, are improved over the first period of time.

[0265] The first node 111 is also configured to output the one or more first indications configured to indicate the result of the determination.

[0266] In some embodiments, the first node 111 may be further configured to obtain, for every first iteration and prior to every determining: i) the one or more first indicators of the one or more key performance indicators, wherein the one or more key performance indicators may be configured to be of one or more of: at least one of the one or more groups of wireless devices 130, the first wireless device 131 of the one or more groups of wireless devices 130, and the first service, ii) the one or more second indicators of the traffic in the cell 121 configured to be served by the radio network node 110 where the one or more groups of wireless devices 130 may be configured to operate, and iii) the one or more third indicators of energy usage by the one or more of: the radio network node 110, the at least one of the one or more groups of wireless devices 130, and the first wireless device 131 of the one or more groups of wireless devices 130. The determining for the respective first iteration may be configured to be based on the respective one or more first indicators, one or more second indicators and one or more third indicators for the respective first iteration.

[0267] In some embodiments, at least one first first indication of the one or more first indications for the respective first iteration may be configured to initiate performing an action in the environment configured to comprise the radio network node 110 and the one or more groups of wireless devices 130, implementing the one or more settings for the sleep mode and the DL transmission power of the radio network node 110 configured to be determined. In some of such embodiments, the first node 111 may be further configured to obtain, for every first iteration, the first reward configured to be obtained from the environment in response to the implemented action. The first reward may be configured to be based on the one or more first utility functions. The determining for the subsequent first iteration immediately following the respective first iteration may be configured to be based on the first reward configured to be obtained. In some embodiments, the respective target one or more key performance indicators and energy saving may be configured to have been obtained from the second node 112 configured to operate in the communications system 100. In some of such embodiments, the first node 111 may be further configured to obtain, following the first number of first iterations, the updated respective target one or more key performance indicators and energy saving from the second node 112, to be used in the subsequent second number of first iterations.

[0268] In some embodiments, the determining may be configured to be based on the radio network node 110 using a low energy scheduler.

[0269] The embodiments herein in the first node 111 may be implemented through one or more processors, such as a processing circuitry 1101 in the first node 111 depicted in Figure 11, together with computer program code for performing the functions and actions of the embodiments herein. A processor, as used herein, may be understood to be a hardware component. The program code mentioned above may also be provided as a computer program product, for instance in the form of a data carrier carrying computer program code for performing the embodiments herein when being loaded into the first node 111. One such carrier may be in the form of a CD ROM disc. It is however feasible with other data carriers such as a memory stick. The computer program code may furthermore be provided as pure program code on a server and downloaded to the first node 111.

[0270] The first node 111 may further comprise a memory 1102 comprising one or more memory units. The memory 1102 is arranged to be used to store obtained information, store data, configurations, schedulings, and applications etc. to perform the methods herein when being executed in the first node 111.

[0271] In some embodiments, the first node 111 may receive information from, e.g., the second node 112, the third node 113, the fourth node 114, the radio network node 110, the one or more groups of wireless devices 130, such as the first wireless device 131, another node or user equipment, and / or another structure in the communications system 100, through a receiving port 1103. In some embodiments, the receiving port 1103 may be, for example, connected to one or more antennas in first node 111. In other embodiments, the first node 111 may receive information from another structure in the communications system 100 through the receiving port 1103. Since the receiving port 1103 may be in communication with the processing circuitry 1101, the receiving port 1103 may then send the received information to the processing circuitry 1101. The receiving port 1103 may also be configured to receive other information.

[0272] The processing circuitry 1101 in the first node 111 may be further configured to transmit or send information to e.g., the second node 112, the third node 113, the fourth node 114, the radio network node 110, the one or more groups of wireless devices 130, such as the first wireless device 131, another node or user equipment, and / or another structure in the communications system 100, through a sending port 1104, which may be in communication with the processing circuitry 1101, and the memory 1102.

[0273] Those skilled in the art will also appreciate that the units comprised within the first node 111 described above as being configured to perform different actions, may refer to a combination of analog and digital circuits, and / or one or more processors configured with software and / or firmware, e.g., stored in memory, that, when executed by the one or more processors such as the processing circuitry 1101, perform as described herein. One or more of these processors, as well as the other digital hardware, may be included in a single Application-Specific Integrated Circuit (ASIC), or several processors and various digital hardware may be distributed among several separate components, whether individually packaged or assembled into a System-on-a-Chip (SoC).

[0274] Also, in some embodiments, the first node 111 may be configured to perform the actions of Figure 4, and / or any of Figures 6-10 with respective units that may be implemented as one or more applications running on one or more processors such as the processing circuitry 1101.

[0275] The first node 111 may be configured to perform any of the Actions described in relation to Figure 4, and / or any of Figures 6-10, e.g., by means of the processing circuitry 1101 within the first node 111, configured to perform any of such actions.

[0276] Thus, the methods according to the embodiments described herein for the first node 111 may be respectively implemented by means of a computer program 1105 product, comprising instructions, i.e. , software code portions, which, when executed on at least one processing circuitry 1101, cause the at least one processing circuitry 1101 to carry out the actions described herein, as performed by the first node 111. The computer program 1105 product may be stored on a computer-readable storage medium 1106. The computer-readable storage medium 1106, having stored thereon the computer program 1105, may comprise instructions which, when executed on at least one processing circuitry 1101, cause the at least one processing circuitry 1101 to carry out the actions described herein, as performed by the first node 111. In some embodiments, the computer-readable storage medium 1106 may be a non-transitory computer-readable storage medium, such as a CD ROM disc, or a memory stick. In other embodiments, the computer program 1105 product may be stored on a carrier containing the computer program 1105 just described, wherein the carrier is one of an electronic signal, optical signal, radio signal, or the computer-readable storage medium 1106, as described above.

[0277] The first node 111 may comprise a communication interface configured to facilitate, or an interface unit to facilitate, communications between the first node 111 and other nodes or devices, e.g., the second node 112, the third node 113, the fourth node 114, the radio network node 110, the one or more groups of wireless devices 130, such as the first wireless device 131, another node or user equipment, and / or another structure in the communications system 100. The interface may, for example, include a transceiver configured to transmit and receive radio signals over an air interface in accordance with a suitable standard.

[0278] In other embodiments, the first node 111 may comprise a radio circuitry 1107, which may comprise e.g., the receiving port 1103 and the sending port 1104.

[0279] The radio circuitry 1107 may be configured to set up and maintain at least a wireless connection with the second node 112, the third node 113, the fourth node 114, the radio network node 110, the one or more groups of wireless devices 130, such as the first wireless device 131, another node or user equipment, and / or another structure in the communications system 100. Circuitry may be understood herein as a hardware component.

[0280] Hence, embodiments herein also relate to the first node 111 operative to operate in the communications system 100. The first node 111 may comprise the processing circuitry 1101 and the memory 1102, said memory 1102 containing instructions executable by said processing circuitry 1101, whereby the first node 111 is further operative to perform the actions described herein in relation to the first node 111.

[0281] Figure 12 depicts an example of the arrangement that the communications system 100 may comprise to perform the embodiments herein. The communications system 100 may be understood to be for handling the one or more settings. The communications system 100 may be configured to comprise the first node 111 and the second node 112.

[0282] Several embodiments are comprised herein. It should be noted that the examples herein are not mutually exclusive. One or more embodiments may be combined, where applicable. All possible combinations are not described to simplify the description.

[0283] Components from one embodiment may be tacitly assumed to be present in another embodiment and it will be obvious to a person skilled in the art how those components may be used in the other exemplary embodiments. The detailed description of some of the following corresponds to the same references provided above, in relation to the actions described for the communications system 100 and will thus not be repeated here. For example, the settings may be configured to be hyperparameters.

[0284] In Figure 12, optional components are depicted with dashed lines.

[0285] The communications system 100 is configured to determine, by the first node 111, jointly, the one or more settings for the sleep mode of the radio network node 110 configured to serve the one or more groups of wireless devices 130 and the DL transmission power of the radio network node 110. The determining is configured to be performed iteratively with the first periodicity and using machine learning, so that based on the respective target, the one or more key performance indicators and energy saving configured to be obtained from the second node 112, are improved over the first period of time. The communications system 100 is also configured to output, by the first node 111 , the one or more first indications configured to indicate the result of the determination.

[0286] The communications system 100 is also configured to determine, by the second node 112 and based on the one or more first indications for the first number of first iterations: i) the respective DRX configurations for the one or more groups of wireless devices 130, and ii) the updated respective target one or more key performance indicators and energy saving. The determining by the second node 112 is configured to be performed iteratively with the second periodicity and using machine learning, so that based on the respective target, the one or more key performance indicators and energy saving, are improved over the second period of time.

[0287] The communications system 100 is further configured to obtain, by the first node 111, following the first number of first iterations, the respective target one or more key performance indicators and energy saving configured to be updated from the second node 112 to be used in the subsequent second number of first iterations.

[0288] The communications system 100 is additionally configured to provide, by the second node 112, the respective DRX configurations for the one or more groups of wireless devices 130 configured to be determined to the radio network node 110.

[0289] In some embodiments, the communications system 100 may be further configured to obtain, by the first node 111 , for every first iteration and prior to every determining: i) the one or more first indicators of the one or more key performance indicators, wherein the one or more key performance indicators may be configured to be of one or more of: at least one of the one or more groups of wireless devices 130, the first wireless device 131 of the one or more groups of wireless devices 130, and the first service, ii) the one or more second indicators of the traffic in the cell 121 configured to be served by the radio network node 110 where the one or more groups of wireless devices 130 may be configured to operate, and iii) the one or more third indicators of energy usage by the one or more of: the radio network node 110, the at least one of the one or more groups of wireless devices 130, and the first wireless device 131 of the one or more groups of wireless devices 130. The determining for the respective first iteration may be configured to be based on the respective one or more first indicators, one or more second indicators and one or more third indicators for the respective first iteration.

[0290] In some embodiments, at least one first first indication of the one or more first indications for the respective first iteration may be configured to initiate performing an action in the environment configured to comprise the radio network node 110 and the one or more groups of wireless devices 130, configured to implement the one or more settings configured to be determined for the sleep mode and the DL transmission power of the radio network node 110. In some of such embodiments, the communications system 100 may be further configured to obtain, by the first node 111 , for every first iteration, the first reward configured to be obtained from the environment in response to the implemented action. The first reward may be configured to be based on the one or more first utility functions. The determining of the one or more settings for the subsequent first iteration immediately following the respective first iteration may be configured to be based on the first reward configured to be obtained.

[0291] In some embodiments, the second periodicity may be configured to comprise the second iterations by the second node 112. In some of such embodiments, for every second iteration, the communications system 100 may be further configured to performed the following three configurations.

[0292] For every second iteration, the communications system 100 may be further configured to obtain, by the second node 112, the one or more second indications configured to indicate the statistics within the respective first number of iterations, the one or more second indications configured to indicate the one or more first indications for the respective first number of first iterations. In some of such embodiments, the determining by the second node 112 may be configured to be based on the one more second indications configured to be obtained.

[0293] For every second iteration, the communications system 100 may be further configured to provide, by the second node 112, the respective target one or more key performance indicators and energy saving to the first node 111 configured to be updated.

[0294] For every second iteration, the communications system 100 may be further configured to obtain, by the second node 112, for every second iteration, the second reward configured to be obtained from the environment in response to the updated respective target one or more key performance indicators and energy saving configured to be provided and respective DRX configurations configured to be determined. The second reward may be configured to be based on the one or more second utility functions. The determining by the second node 112 for the subsequent second iteration immediately following the respective second iteration may be configured to be based on the second reward configured to be obtained.

[0295] In some embodiments, one or more of the following may apply: a) the determining 502 by the first node 111 may be configured to be based on the radio network node (110) using a low energy scheduler, and b) the determining 506 by the second node 112 may be configured to be based on the radio network node 110 using the low energy scheduler.

[0296] The embodiments herein in the first node 111 may be implemented through one or more processors, such as a processing circuitry 1201 in the first node 111 depicted in Figure 12, together with computer program code for performing the functions and actions of the embodiments herein. A processor, as used herein, may be understood to be a hardware component. The program code mentioned above may also be provided as a computer program product, for instance in the form of a data carrier carrying computer program code for performing the embodiments herein when being loaded into the first node 111. One such carrier may be in the form of a CD ROM disc. It is however feasible with other data carriers such as a memory stick. The computer program code may furthermore be provided as pure program code on a server and downloaded to the first node 111.

[0297] The first node 111 may further comprise a memory 1202 comprising one or more memory units. The memory 1202 is arranged to be used to store obtained information, store data, configurations, schedulings, and applications etc. to perform the methods herein when being executed in the first node 111.

[0298] In some embodiments, the first node 111 may receive information from, e.g., the second node 112, the third node 113, the fourth node 114, the radio network node 110, the one or more groups of wireless devices 130, such as the first wireless device 131, another node or user equipment, and / or another structure in the communications system 100, through a receiving port 1203. In some embodiments, the receiving port 1203 may be, for example, connected to one or more antennas in first node 111. In other embodiments, the first node 111 may receive information from another structure in the communications system 100 through the receiving port 1203. Since the receiving port 1203 may be in communication with the processing circuitry 1201, the receiving port 1203 may then send the received information to the processing circuitry 1201. The receiving port 1203 may also be configured to receive other information.

[0299] The processing circuitry 1201 in the first node 111 may be further configured to transmit or send information to e.g., the second node 112, the third node 113, the fourth node 114, the radio network node 110, the one or more groups of wireless devices 130, such as the first wireless device 131, another node or user equipment, and / or another structure in the communications system 100, through a sending port 1204, which may be in communication with the processing circuitry 1201, and the memory 1202.

[0300] Those skilled in the art will also appreciate that the units comprised within the first node 111 described above as being configured to perform different actions, may refer to a combination of analog and digital circuits, and / or one or more processors configured with software and / or firmware, e.g., stored in memory, that, when executed by the one or more processors such as the processing circuitry 1201, perform as described herein. One or more of these processors, as well as the other digital hardware, may be included in a single Application-Specific Integrated Circuit (ASIC), or several processors and various digital hardware may be distributed among several separate components, whether individually packaged or assembled into a System-on-a-Chip (SoC).

[0301] Also, in some embodiments, the first node 111 may be configured to perform the actions of Figure 4, and / or any of Figures 6-10 with respective units that may be implemented as one or more applications running on one or more processors such as the processing circuitry 1101. The first node 111 may be configured to perform any of the Actions described in relation to Figure 4, and / or any of Figures 6-10, e.g., by means of the processing circuitry 1101 within the first node 111, configured to perform any of such actions.

[0302] Thus, the methods according to the embodiments described herein for the first node 111 may be respectively implemented by means of a computer program 1205 product, comprising instructions, i.e. , software code portions, which, when executed on at least one processing circuitry 1201, cause the at least one processing circuitry 1201 to carry out the actions described herein, as performed by the first node 111. The computer program 1205 product may be stored on a computer-readable storage medium 1206. The computer-readable storage medium 1206, having stored thereon the computer program 1205, may comprise instructions which, when executed on at least one processing circuitry 1201, cause the at least one processing circuitry 1201 to carry out the actions described herein, as performed by the first node 111. In some embodiments, the computer-readable storage medium 1206 may be a non-transitory computer-readable storage medium, such as a CD ROM disc, or a memory stick. In other embodiments, the computer program 1205 product may be stored on a carrier containing the computer program 1205 just described, wherein the carrier is one of an electronic signal, optical signal, radio signal, or the computer-readable storage medium 1206, as described above.

[0303] The first node 111 may comprise a communication interface configured to facilitate, or an interface unit to facilitate, communications between the first node 111 and other nodes or devices, e.g., the second node 112, the third node 113, the fourth node 114, the radio network node 110, the one or more groups of wireless devices 130, such as the first wireless device 131, another node or user equipment, and / or another structure in the communications system 100. The interface may, for example, include a transceiver configured to transmit and receive radio signals over an air interface in accordance with a suitable standard.

[0304] In other embodiments, the first node 111 may comprise a radio circuitry 1207, which may comprise e.g., the receiving port 1203 and the sending port 1204.

[0305] The radio circuitry 1207 may be configured to set up and maintain at least a wireless connection with the second node 112, the third node 113, the fourth node 114, the radio network node 110, the one or more groups of wireless devices 130, such as the first wireless device 131, another node or user equipment, and / or another structure in the communications system 100. Circuitry may be understood herein as a hardware component.

[0306] Hence, embodiments herein also relate to the first node 111 operative to operate in the communications system 100. The first node 111 may comprise the processing circuitry 1201 and the memory 1202, said memory 1202 containing instructions executable by said processing circuitry 1201 , whereby the first node 111 is further operative to perform the actions described herein in relation to the first node 111. The embodiments herein in the second node 112 may be implemented through one or more processors, such as a processing circuitry 1208 in the second node 112 depicted in Figure 12, together with computer program code for performing the functions and actions of the embodiments herein. A processor, as used herein, may be understood to be a hardware component. The program code mentioned above may also be provided as a computer program product, for instance in the form of a data carrier carrying computer program code for performing the embodiments herein when being loaded into the second node 112. One such carrier may be in the form of a CD ROM disc. It is however feasible with other data carriers such as a memory stick. The computer program code may furthermore be provided as pure program code on a server and downloaded to the second node 112.

[0307] The second node 112 may further comprise a memory 1209 comprising one or more memory units. The memory 1209 is arranged to be used to store obtained information, store data, configurations, schedulings, and applications etc. to perform the methods herein when being executed in the second node 112.

[0308] In some embodiments, the second node 112 may receive information from, e.g., the first node 111 , the third node 113, the fourth node 114, the radio network node 110, the one or more groups of wireless devices 130, such as the first wireless device 131, another node or user equipment, and / or another structure in the communications system 100, through a receiving port 1210. In some embodiments, the receiving port 1210 may be, for example, connected to one or more antennas in second node 112. In other embodiments, the second node 112 may receive information from another structure in the communications system 100 through the receiving port 1210. Since the receiving port 1210 may be in communication with the processing circuitry 1208, the receiving port 1210 may then send the received information to the processing circuitry 1208. The receiving port 1210 may also be configured to receive other information.

[0309] The processing circuitry 1208 in the second node 112 may be further configured to transmit or send information to e.g., the first node 111, the third node 113, the fourth node 114, the radio network node 110, the one or more groups of wireless devices 130, such as the first wireless device 131, another node or user equipment, and / or another structure in the communications system 100, through a sending port 1211, which may be in communication with the processing circuitry 1208, and the memory 1209.

[0310] Those skilled in the art will also appreciate that the units comprised within the second node 112 described above as being configured to perform different actions, may refer to a combination of analog and digital circuits, and / or one or more processors configured with software and / or firmware, e.g., stored in memory, that, when executed by the one or more processors such as the processing circuitry 1208, perform as described above. One or more of these processors, as well as the other digital hardware, may be included in a single Application-Specific Integrated Circuit (ASIC), or several processors and various digital hardware may be distributed among several separate components, whether individually packaged or assembled into a System-on-a-Chip (SoC).

[0311] Also, in some embodiments, the second node 112 may be configured to perform the actions of Figure 5, and / or any of Figures 6-10 with respective units that may be implemented as one or more applications running on one or more processors such as the processing circuitry 1208.

[0312] The second node 112 may be configured to perform any of the Actions described in relation to Figure 5, and / or any of Figures 6-10, e.g., by means of the processing circuitry 1208 within the second node 112, configured to perform any of such actions.

[0313] Thus, the methods according to the embodiments described herein for the second node 112 may be respectively implemented by means of a computer program 1212 product, comprising instructions, i.e. , software code portions, which, when executed on at least one processing circuitry 1208, cause the at least one processing circuitry 1208 to carry out the actions described herein, as performed by the second node 112. The computer program 1212 product may be stored on a computer-readable storage medium 1213. The computer-readable storage medium 1213, having stored thereon the computer program 1212, may comprise instructions which, when executed on at least one processing circuitry 1208, cause the at least one processing circuitry 1208 to carry out the actions described herein, as performed by the second node 112. In some embodiments, the computer-readable storage medium 1213 may be a non-transitory computer-readable storage medium, such as a CD ROM disc, or a memory stick. In other embodiments, the computer program 1212 product may be stored on a carrier containing the computer program 1212 just described, wherein the carrier is one of an electronic signal, optical signal, radio signal, or the computer-readable storage medium 1213, as described above.

[0314] The second node 112 may comprise a communication interface configured to facilitate, or an interface unit to facilitate, communications between the second node 112 and other nodes or devices, e.g., the first node 111, the third node 113, the fourth node 114, the radio network node 110, the one or more groups of wireless devices 130, such as the first wireless device 131, another node or user equipment, and / or another structure in the communications system 100. The interface may, for example, include a transceiver configured to transmit and receive radio signals over an air interface in accordance with a suitable standard.

[0315] In other embodiments, the second node 112 may comprise a radio circuitry 1214, which may comprise e.g., the receiving port 1210 and the sending port 1211.

[0316] The radio circuitry 1214 may be configured to set up and maintain at least a wireless connection with the first node 111 , the third node 113, the fourth node 114, the radio network node 110, the one or more groups of wireless devices 130, such as the first wireless device 131, another node or user equipment, and / or another structure in the communications system 100. Circuitry may be understood herein as a hardware component.

[0317] Hence, embodiments herein also relate to the second node 112, operative to operate in the communications system 100. The second node 112 may comprise the processing circuitry 1208 and the memory 1209, said memory 1209 containing instructions executable by said processing circuitry 1208, whereby the second node 112 is further operative to perform the actions described herein in relation to the second node 112.

[0318] When using the word "comprise" or “comprising”, it shall be interpreted as non- limiting, i.e. , meaning "consist at least of".

[0319] The embodiments herein are not limited to the above-described preferred embodiments. Various alternatives, modifications and equivalents may be used. Therefore, the above embodiments should not be taken as limiting the scope of the invention.

[0320] 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.

[0321] As used herein, the expression “at least one of:” followed by a list of alternatives separated by commas, and wherein the last alternative is preceded by the “and” term, may be understood to mean that only one of the list of alternatives may apply, more than one of the list of alternatives may apply or all of the list of alternatives may apply. This expression may be understood to be equivalent to the expression “at least one of:” followed by a list of alternatives separated by commas, and wherein the last alternative is preceded by the “or” term.

[0322] Any of the terms processor and circuitry may be understood herein as a hardware component.

[0323] As used herein, the expression “in some embodiments” has been used to indicate that the features of the embodiment described may be combined with any other embodiment or example disclosed herein. As used herein, the expression “in some examples” has been used to indicate that the features of the example described may be combined with any other embodiment or example disclosed herein.

[0324] REFERENCES

[0325] 1. F. Moradi, E. Fitzgerald, M. Pioro and B. Landfeldt, “Flexible DRX Optimization for LTE and 5G,” in IEEE Transactions on Vehicular Technology, vol. 69, no. 1, pp. 607-621, Jan. 2020.

[0326] 2. 3GPP TR 38.864 version 18.1.0 version 18.1.0, 2023. Study on network energy savings for NR”.

[0327] 3. Jang, Gunhee, et al. "Base station switching and sleep mode optimization with LSTM- based user prediction." IEEE Access 8 (2020): 222711-222723.

Claims

CLAIMS:

1. A computer-implemented method, performed by a first node (111) operating in a communications system (100), the method comprising:- determining (402), jointly, one or more settings for a sleep mode of a radio network node (110) serving one or more groups of wireless devices (130) and a Downlink, DL, transmission power of the radio network node (110), wherein the determining (402) is performed iteratively using machine learning, so that based on a respective target, one or more key performance indicators and energy saving, are improved over a first period of time, and- outputting (403) one or more first indications indicating a result of the determination.

2. The method according to claim 1, further comprising:- obtaining (401), for every first iteration and prior to every determining (402):i. one or more first indicators of the one or more key performance indicators, wherein the one or more key performance indicators are of one or more of: at least one of the one or more groups of wireless devices (130), a first wireless device (131) of the one or more groups of wireless devices (130), and a first service,ii. one or more second indicators of a traffic in a cell (121) served by the radio network node (110) where the one or more groups of wireless devices (130) are operating,iii. one or more third indicators of energy usage by one or more of: the radio network node (110), the at least one of the one or more groups of wireless devices (130), and the first wireless device (131) of the one or more groups of wireless devices (130), andwherein the determining (402) for a respective first iteration is based on the respective one or more first indicators, one or more second indicators and one or more third indicators for the respective first iteration.

3. The method according to claim 2, wherein at least one first first indication of the one or more first indications for the respective first iteration initiates performing an action in an environment comprising the radio network node (110) and the one or more groups of wireless devices (130), implementing the determined one or more settings for the sleep mode and the DL transmission power of the radio network node (110), and wherein the method further comprises:- obtaining (404), for every first iteration, a first reward obtained from the environment in response to the implemented action, the first reward being based on one or more first utility functions, and wherein the determining (402) for a subsequent first iteration immediately following the respective first iteration is based on the obtained first reward.

4. The method according to any of claims 2-3, wherein the respective target one or more key performance indicators and energy saving have been obtained from a second node (112) operating in the communications system (100), and wherein the method further comprises:- obtaining (405), following a first number of first iterations, updated respective target one or more key performance indicators and energy saving from the second node (112), to be used in a subsequent second number of first iterations.

5. The method according to any of claims 1-4, wherein the determining (402) is based on the radio network node (110) using a low energy scheduler.

6. A computer-implemented method, performed by communications system (100) comprising a first node (111) and a second node (112), the method comprising:- determining (502), by the first node (111), jointly, one or more settings for a sleep mode of a radio network node (110) serving one or more groups of wireless devices (130) and a Downlink, DL, transmission power of the radio network node (110), wherein the determining (502) is performed iteratively with a first periodicity and using machine learning, so that based on a respective target, one or more key performance indicators and energy saving obtained from the second node (112), are improved over a first period of time, and- outputting (503), by the first node (111), one or more first indications indicating a result of the determination,- determining (506), by the second node (112) and based on the one or more first indications for a first number of first iterations:i. respective discontinued reception, DRX, configurations for the one or more groups of wireless devices (130), andii. updated respective target one or more key performance indicators and energy saving,wherein the determining (506) by the second node (112) is performed iteratively with a second periodicity and using machine learning, so thatbased on the respective target, one or more key performance indicators and energy saving, are improved over a second period of time, - obtaining (508), by the first node (111), following the first number of first iterations, the updated respective target one or more key performance indicators and energy saving from the second node (112) to be used in a subsequent second number of first iterations, and- providing (509), by the second node (112), the determined respective DRX configurations for the one or more groups of wireless devices (130) to the radio network node (110).

7. The method according to claim 6, further comprising:- obtaining (501), by the first node (111), for every first iteration and prior to every determining (502) performed by the first node (111),i. one or more first indicators of the one or more key performance indicators, wherein the one or more key performance indicators are of one or more of: at least one of the one or more groups of wireless devices (130), a first wireless device (131) of the one or more groups of wireless devices (130), and a first service,ii. one or more second indicators of a traffic in a cell (121) served by the radio network node (110) where the one or more groups of wireless devices (130) are operating,iii. one or more third indicators of energy usage by one or more of: the radio network node (110), the at least one of the one or more groups of wireless devices (130), and the first wireless device (131) of the one or more groups of wireless devices (130), andwherein the determining (502) for a respective first iteration is based on the respective one or more first indicators, one or more second indicators and one or more third indicators for the respective first iteration.

8. The method according to claim 7, wherein at least one first first indication of the one or more first indications for the respective first iteration initiates performing an action in an environment comprising the radio network node (110) and the one or more groups of wireless devices (130) implementing the determined one or more settings for the sleep mode and the DL transmission power of the radio network node (110), and wherein the method further comprises:- obtaining (504), by the first node (111), for every first iteration, a first reward obtained from the environment in response to the implemented action, the firstreward being based on one or more first utility functions, and wherein the determining (502) for a subsequent first iteration immediately following the respective first iteration is based on the obtained first reward.

9. The method according to any of claims 7-8, wherein the second periodicity comprises second iterations by the second node (112), and wherein for every second iteration, the method further comprises:- obtaining (505), by the second node (112), one or more second indications indicating statistics within a respective first number of iterations, the one or more second indications indicating the one or more first indications for the respective first number of first iterations, and wherein the determining (506) by the second node (112) is based on the obtained one more second indications,- providing (507), by the second node (112), the updated respective target one or more key performance indicators and energy saving to the first node (111), and - obtaining (510), by the second node (112), for every second iteration, a second reward obtained from the environment in response to the provided updated respective target one or more key performance indicators and energy saving and determined respective DRX configurations, the second reward being based on one or more second utility functions, and wherein the determining (506) by the second node (112) for a subsequent second iteration immediately following the respective second iteration is based on the obtained second reward.

10. The method according to any of claims 6-9, wherein one or more of:a. the determining (502) by the first node (111) is based on the radio network node (110) using a low energy scheduler, andb. the determining (506) by the second node (112) is based on the radio network node (110) using the low energy scheduler.

11. A first node (111) configured to operate in a communications system (100), the first node (111) being further configured to:- determine, jointly, one or more settings for a sleep mode of a radio network node (110) configured to serve one or more groups of wireless devices (130) and a Downlink, DL, transmission power of the radio network node (110), wherein the determining is configured to be performed iteratively using machine learning, so that based on a respective target, one or more key performance indicators and energy saving, are improved over a first period of time, and- output one or more first indications configured to indicate a result of the determination.

12. The first node (111) according to claim 11, being further configured to:- obtain, for every first iteration and prior to every determining:i. one or more first indicators of the one or more key performance indicators, wherein the one or more key performance indicators are configured to be of one or more of: at least one of the one or more groups of wireless devices (130), a first wireless device (131) of the one or more groups of wireless devices (130), and a first service,ii. one or more second indicators of a traffic in a cell (121) configured to be served by the radio network node (110) where the one or more groups of wireless devices (130) are configured to operate, andiii. one or more third indicators of energy usage by one or more of: the radio network node (110), the at least one of the one or more groups of wireless devices (130), and the first wireless device (131) of the one or more groups of wireless devices (130), andwherein the determining for a respective first iteration is configured to be based on the respective one or more first indicators, one or more second indicators and one or more third indicators for the respective first iteration.

13. The first node (111) according to claim 12, wherein at least one first first indication of the one or more first indications for the respective first iteration is configured to initiate performing an action in an environment configured to comprise the radio network node (110) and the one or more groups of wireless devices (130), implementing the one or more settings for the sleep mode and the DL transmission power of the radio network node (110) configured to be determined, and wherein the first node (111) is further configured to:- obtain, for every first iteration, a first reward configured to be obtained from the environment in response to the implemented action, the first reward being configured to be based on one or more first utility functions, and wherein the determining for a subsequent first iteration immediately following the respective first iteration is configured to be based on the first reward configured to be obtained.

14. The first node (111) according to any of claims 12-13, wherein the respective target one or more key performance indicators and energy saving are configured to have beenobtained from a second node (112) configured to operate in the communications system (100), and wherein the first node (111) is further configured to:- obtain, following a first number of first iterations, updated respective target one or more key performance indicators and energy saving from the second node (112), to be used in a subsequent second number of first iterations.

15. The method according to any of claims 11-14, wherein the determining is configured to be based on the radio network node (110) using a low energy scheduler.

16. A communications system (100) configured to comprise a first node (111) and a second node (112), the communications system (100) being further configured to: - determine, by the first node (111), jointly, one or more settings for a sleep mode of a radio network node (110) configured to serve one or more groups of wireless devices (130) and a Downlink, DL, transmission power of the radio network node (110), wherein the determining is configured to be performed iteratively with a first periodicity and using machine learning, so that based on a respective target, one or more key performance indicators and energy saving configured to be obtained from the second node (112), are improved over a first period of time, and- output, by the first node (111), one or more first indications configured to indicate a result of the determination,- determining, by the second node (112) and based on the one or more first indications for a first number of first iterations:i. respective discontinued reception, DRX, configurations for the one or more groups of wireless devices (130), andii. updated respective target one or more key performance indicators and energy saving,wherein the determining by the second node (112) is configured to be performed iteratively with a second periodicity and using machine learning, so that based on the respective target, one or more key performance indicators and energy saving, are improved over a second period of time,- obtain, by the first node (111), following the first number of first iterations, the respective target one or more key performance indicators and energy saving configured to be updated from the second node (112) to be used in a subsequent second number of first iterations, and- provide, by the second node (112), the respective DRX configurations for the one or more groups of wireless devices (130) configured to be determined to the radio network node (110).

17. The communications system (100) according to claim 16, being further configured to:- obtain, by the first node (111), for every first iteration and prior to every determining performed by the first node (111),i. one or more first indicators of the one or more key performance indicators, wherein the one or more key performance indicators are configured to be of one or more of: at least one of the one or more groups of wireless devices (130), a first wireless device (131) of the one or more groups of wireless devices (130), and a first service,ii. one or more second indicators of a traffic in a cell (121) configured to be served by the radio network node (110) where the one or more groups of wireless devices (130) are configured to operate, andiii. one or more third indicators of energy usage by one or more of: the radio network node (110), the at least one of the one or more groups of wireless devices (130), and the first wireless device (131) of the one or more groups of wireless devices (130), andwherein the determining for a respective first iteration is configured to be based on the respective one or more first indicators, one or more second indicators and one or more third indicators for the respective first iteration.

18. The communications system (100) according to claim 17, wherein at least one first first indication of the one or more first indications for the respective first iteration is configured to initiate performing an action in an environment configured to comprise the radio network node (110) and the one or more groups of wireless devices (130) configured to implement the one or more settings configured to be determined for the sleep mode and the DL transmission power of the radio network node (110), and wherein the communications system (100) is further configured to:- obtain, by the first node (111), for every first iteration, a first reward obtained from the environment in response to the implemented action, the first reward being configured to be based on one or more first utility functions, and wherein the determining of the one or more settings for a subsequent first iteration immediately following the respective first iteration is configured to be based on the first reward configured to be obtained.

19. The communications system (100) according to any of claims 17-18, wherein the second periodicity is configured to comprise second iterations by the second node (112), and wherein for every second iteration, the communications system (100) is further configured to:- obtain, by the second node (112), one or more second indications configured to indicate statistics within a respective first number of iterations, the one or more second indications being configured to indicate the one or more first indications for the respective first number of first iterations, and wherein the determining by the second node (112) is configured to be based on the one more second indications configured to be obtained,- provide, by the second node (112), the respective target one or more key performance indicators and energy saving to the first node (111) configured to be updated, and- obtain, by the second node (112), for every second iteration, a second reward configured to be obtained from the environment in response to the updated respective target one or more key performance indicators and energy saving configured to be provided and respective DRX configurations configured to be determined, the second reward being configured to be based on one or more second utility functions, and wherein the determining by the second node (112) for a subsequent second iteration immediately following the respective second iteration is configured to be based on the second reward configured to be obtained.

20. The method according to any of claims 16-19, wherein one or more of:a. the determining (502) by the first node (111) is configured to be based on the radio network node (110) using a low energy scheduler, andb. the determining (506) by the second node (112) is configured to be based on the radio network node (110) using the low energy scheduler.

21. A computer program (1105), comprising instructions which, when executed on at least one processing circuitry (1101), cause the at least one processing circuitry (1101) to carry out the method according to any of claims 1-5.

22. A computer-readable storage medium (1106), having stored thereon a computer program (1105), comprising instructions which, when executed on at least one processing circuitry (1201), cause the at least one processing circuitry (1101) to carry out the method according to any of claims 1-5.

23. A computer program (1205, 1212), comprising instructions which, when executed on at least one processing circuitry (1201, 1208), cause the at least one processing circuitry (1201, 1208) to carry out the method according to any of claims 6-10.

24. A computer-readable storage medium (1206, 1213), having stored thereon a computer program (1205, 1212), comprising instructions which, when executed on at least one processing circuitry (1312, 1208), cause the at least one processing circuitry (1201, 1208) to carry out the method according to any of claims 6-10.

Images