Dynamically optimizing trajectories of mobile network nodes using tensor networks

Abstract

A method performed by a mobile network node operating in a network is disclosed. The method includes collecting user equipment, UE, parameter data from each of a plurality of UE devices operating in the network; utilizing the collected UE parameter data to cluster the plurality of UE devices into one or more cluster groups, which are subsequently converted to a respective one or more tensor networks; determining the future trajectory positions for each of the UE devices included in a cluster group assigned to the mobile network node using the one or more tensor networks; determining a most energy-efficient movement for the mobile network node to support the future trajectory positions of the plurality of UE devices; and establishing a new position for the mobile network node based on the most energy-efficient movement and an optimal data rate prediction achievable by the mobile network node.

Description

[0001] DYNAMICALLY OPTIMIZING TRAJECTORIES OF MOBILE NETWORK NODES USING TENSOR NETWORKS TECHNICAL FIELD [1] The present disclosure relates generally to communications, and more particularly to methods and related mobile devices and mobile network nodes performing wireless and / or cellular based communications and signaling. BACKGROUND [2] Unmanned aerial vehicle (UAV) technology has recently attracted a lot of attention as a candidate to meet the future 6G ubiquitous connectivity demand and boost the resiliency of terrestrial networks. To compensate for the huge availability of data towards the user equipment (e.g., mobile devices, tablets and sensors) there is a need for UAV base stations that have the ability to reposition themselves to provide improved quality of service (QoS) to handle extremely large volumes of data. With recent advancements in drone technology, researchers are now considering the possibility of deploying small cells served by base stations mounted on flying drones. A major advantage afforded by using such small drone cells is that operators can promptly provide cellular services in areas of urgent demand without having to pre-install any infrastructure. For such a configuration to be practical, the drone- based UAV should have cost-effective energy movement, thereby allowing its battery to be available over long periods of time. The issue is that there is no intelligence currently available and / or in place for UAVs to assess the movement of UEs and subsequently reposition themselves to support the UEs that come within their service range. SUMMARY [3] There currently exist certain challenge(s). Applying deep learning (DL) to various applications will greatly enhance the accuracy of machine learning (ML) models. However, deep learning deals with large volumes of data and involves vast neural networks (NNs) in the learning process. Conventional machine learning technologies have provided a “black box” type of approach, which can be challenging to apply to critical decision making. Hence, it is difficult to explain how a model reached the particular output even if the inference was correct. This issue complicates the process for artificial intelligence (AI) technology applications that demand absolute reliability. Moreover, using an Explainable AI methodology will unnecessarily overload the applications. Here Tensor networks can be replaced with deep learning neural networks (DNNs) for certain use cases (e.g., Healthcare, Finance, Telecommunication, etc.). Tensor network implementations provide relative transparency as to how a final output in the process is reached. Tensor networks may include an intuitive graphical language that can be used to reason about them. Tensor Deep Networks can solve many of the issues. [4] Certain aspects of the disclosure and their embodiments may provide solutions to these or other challenges. Notably, current solutions in marketplace do not have the capability to track the trajectory of the user devices where both UAV and UEs are mobile and frequently changing their respective positions. Further, current solutions are unable to determine the best UAV movement position, which is often related to the most energy efficient solution (i.e., a movement and / or repositioning that places the UAV in a position to support UEs with the least amount of energy expended. In addition, there are presently no optimal solutions capable of determining the best data rate based on the different UAV movements to determine a UAV optimal position. There is also no known solution currently available to subdivide the user devices based on the density of the UE devices and subsequently allocate and / or distribute the UAVs accordingly in a tensor network format. Lastly, the present solutions available in the marketplace are unable to efficiently consider signal-to-interference-plus-noise ratio (SINR) data while determining the UAV positions based on UE movement. [5] In one embodiment, the disclosed subject matter relates to a method performed by a mobile network node operating in a network, the method comprising: collecting user equipment, UE, parameter data from each of a plurality of UE devices operating in the network; utilizing the collected UE parameter data to cluster the plurality of UE devices into one or more cluster groups, which are subsequently converted to a respective one or more tensor networks; determining the future trajectory positions for each of the UE devices included in a cluster group assigned to the mobile network node using the one or more tensor networks; determining a most energy-efficient movement for the mobile network node to support the future trajectory positions of the plurality of UE devices; and establishing a new position for the mobile network node based on the most energy-efficient movement and an optimal data rate prediction achievable by the mobile network node. [6] In one embodiment, the disclosed subject matter relates to a mobile network node comprising: processing circuitry and at least one memory. The at least one memory can be configured to store instructions executable by the processing circuitry to perform operations to: collect user equipment, UE, parameter data from each of a plurality of UE devices operating in the network; utilize the collected UE parameter data to cluster the plurality of UE devices into one or more cluster groups, which are subsequently converted to a respective one or more tensor networks; determine the future trajectory positions for each of the UE devices included in a cluster group assigned to the mobile network node using the one or more tensor networks; determine a most energy efficient movement for the mobile network node to support the future trajectory positions of the plurality of UE devices; and establish a new position for the mobile network node based on the most energy efficient movement and an optimal data rate prediction achievable by the mobile network node. [7] In one embodiment, the disclosed subject matter relates to a non-transitory computer readable medium storing instructions executable by processing circuitry of a mobile network node, the instructions executed by the processing circuitry to perform operations comprising: collecting user equipment, UE, parameter data from each of a plurality of UE devices operating in the network; utilizing the collected UE parameter data to cluster the plurality of UE devices into one or more cluster groups, which are subsequently converted to a respective one or more tensor networks; determining future trajectory positions for each of the UE devices included in a cluster group assigned to the mobile network node using the one or more tensor networks; determining a most energy efficient movement for the mobile network node to support the future trajectory positions of the plurality of UE devices; and establishing a new position for the mobile network node based on the most energy-efficient movement and an optimal data rate prediction achievable by the mobile network node. BRIEF DESCRIPTION OF THE DRAWINGS [8] The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this application, illustrate certain non-limiting embodiments of inventive concepts. In the drawings: [9] Figure 1 is a schematic diagram illustrating an example of a communication network with unmanned aerial vehicle (“UAV”) base stations that include mobile edge nodes in accordance with some embodiments;

[0010] Figure 2 illustrates a diagram of an exemplary tensor decomposition according to some embodiments;

[0011] Figure 3 is a block diagram of a one-shot learning model according to some embodiments;

[0012] Figure 4 illustrates a block diagram of a system 400 that utilizes one-shot learning according to some embodiments;

[0013] Figure 5 is a flow diagram that illustrates a mobile network node repositioning procedure of according to some embodiments;

[0014] Figure 6 is a sequence diagram illustrating a mobile network node repositioning procedure according to some embodiments;

[0015] Figure 7 illustrates a block diagram of an exemplary tensor state network model according to some embodiments;

[0016] Figure 8 illustrates flow diagram of operations performed by a policy network according to some embodiments;

[0017] Figure 9 illustrates an graph illustrating data rate measurements performed at periodic epochs according to some embodiments;

[0018] Figure 10 illustrates a bar graph depicting comparing energy consumed via UAB base station movement versus legacy UAV base station movement according to some embodiments;

[0019] Figure 11 illustrates a graph plotting a data rate summation and SINR data according to some embodiments;

[0020] Figure 12 is a block diagram of a mobile network node in accordance with some embodiments;

[0021] Figure 13 is a flow chart illustrating example operations for dynamically optimizing mobile network nodes using tensor networks according to some embodiments;

[0022] Figure 14 is a block diagram of a communication system in accordance with some embodiments;

[0023] Figure 15 is a block diagram of a user equipment in accordance with some embodiments;

[0024] Figure 16 is a block diagram of a network node in accordance with some embodiments; and

[0025] Figure 17 is a block diagram of a virtualization environment in accordance with some embodiments. DETAILED DESCRIPTION

[0026] Unmanned Aerial Vehicle (UAV) on Demand is one of the key market research areas for both the private 5G environment and 6G environments. The UAV can be operated via remote control, or it can be configured to operate in an autonomous manner. The main intent of this problem is to install a UAV base station, which can run autonomously based on the criteria on trajectory movement of a user equipment (UW), line of sight (LOS), and energy consumption. Line of sight can be calculated based on multiple factors like distance(s) between the UAV and UE, the height of UAV and Non-Line-of-Sight (NLOS), Noise emission, and the like.

[0027] Moreover, tensor decompositions are effective in analyzing the structure of multidimensional data. Tensor networks, which comprise a contracted network of factor tensors, have been independently developed in several areas of science and engineering. Such networks appear in the description of physical processes and an accompanying collection of numerical techniques have elevated the use of tensor networks into a variational model of machine learning (ML). Underlying these algorithms is the compression of high-dimensional data. These compression techniques have recently proven ripe to apply to many traditional problems faced in deep learning (DL).

[0028] The interplay between tensor networks and machine learning is explained in the present disclosure. Indeed, this interplay is based not just on numerical methods but on the equivalence for optimizing and transforming tensor networks, and connections to low-rank methods for learning.

[0029] Deep reinforcement learning (DRL), which is mainly a reinforcement learning (RL) technique combined with a deep-neural-network (DNN), has the potential to handle high- dimensional inputs, learn patterns, and solve complex problems efficiently. Even though some initial works consider DRL for two dimensional UAVs placement in the presence of line of light (LoS) links only, which employs deep deterministic policy gradient (DDPG) DRL, the full potentials of DRL for three-dimensional (3D) networks still need to be assessed. Integrating dueling network architectures with DRL has shown its merits by dramatically improving and generalizing learning in various domains. In the same regard, dueling DRL, as compared to other DRL advances like double DRL shows an improvement in the learning speed in the context of UAV sensing related applications and using a UAV as a base station. Nevertheless, there is limited research and / or investigations conducted on the integration of dueling DRL and UAVs. The present disclosure provides a solution to the non-convex optimization problem of the 3D-network resources management using the disclosed dueling DRL based framework. Notably, the disclosed subject matter brings forth the most-recent advances in AI to efficiently solve the UAV placement and resources management while maximizing both users' data rate and fairness.

[0030] Figure 1 illustrates examples of a 5G or 6G network that includes a set of UAV base stations 140a-c that serve as mobile edge nodes (e.g., each UAV base station 140a-c includes a mobile network node). In this example, each of the UAV base stations 140a-c can wireless communicate with UEs 110 within their corresponding coverage areas. The UAV base stations 140a-c can also wireless communicate with a static network node 120a (e.g., a base station). Accordingly, the UAV base stations 140a-c can serve as access points for the UEs 110 to communicate with the network node 120a and a corresponding network (e.g., a private 5G network). In some examples, the UAV base stations 140a-c can directly communicate between each other. The network in Figure 2 may be considered balanced because each of the UAV base stations 140a-c include two UEs 110 in their coverage areas.

[0031] In real-world scenarios, mobile network nodes, such as UAV base stations (BSs), are expected to dynamically reposition themselves in response to the dynamic movement of UEs. While this approach considers dynamic scenarios involving UE movement, the reposition action performed by the base station is done after the movement of UEs, which may cause communication service outage. To serve the UEs seamlessly, a prediction on the future positions of UEs needs to be determined prior to the reposition action of UAV-BSs in the next time slot, which is a challenging task. Performing this prediction may be implemented by applying a Tensor Network (TN) as an enabler in the framework.

[0002]

[0032] Notably, the disclosed subject matter is designed to determine the trajectory movement of UEs with the assistance of a tensor decomposition network. Notably, the tensor decomposition network takes data relating to the existing pattern of UE movement as input and subsequently calculates the future angle of the UE(s) through the approach set forth in the present disclosure. Energy consumption is primarily attributed to the effective usage of power displacement on the drawn battery. Notably, this energy consumption cost may be calculated using the tensor based decomposition technique as detailed below. Use of this tensor based decomposition technique will support the repositioning process in an allotted time.

[0033] The disclosed subject matter pertains to the application of tensor decomposition to the various algorithms in the proposed framework and via the use of innovative AI methodology. Notably, the disclosed process is expected to be more scalable and generic to any new scenario related to the UAV and robotics world. In some embodiments, the metrics of the present disclosure are derived based on the following data through a tensor decomposition approach. First, the disclosed subject matter utilizes “one shot learning,” which uses the concept of similarities existing among the data in a tensor space. The primarily goal of using the one- shot learning is to teach the model to set its own assumptions about data similarities based on a minimal number of components to scale at the expected level of requirements.

[0034] In some embodiments, the disclosed subject matter may be used to create a density- based (DB) clustering scan for the grouping or clustering of the UEs that are ultimately assigned to a particular mobile network node (i.e., UAV base station, based on the formed cluster). In some embodiments, the DB clustering scan includes a Density-Based Spatial Clustering of Applications with Noise (DBSCAN) operation.

[0035] Moreover, the disclosed subject matter may be used to generate a tensor-based network model for deriving the energy cost of the mobile network node movement. This metric ensures that the energy displacement experienced during the UAV / mobile network node movement is kept to its minimum. In some embodiments, the disclosed subject matter creates a tensor state network to determine the trajectory position based on the existing movement of UEs.

[0036] In some embodiments, the disclosed subject matter is configured to create a Tensor network-based reinforcement learning environment that calculates the movement of mobile network node and determines how effective this movement improves communication with respect to higher data rates and bandwidths based on Line of Sight, height of the UAV, altitude positioning, and / or the distance between the mobile network node / UAV and the UE devices.

[0037] Notably, the entire process can be continuously executed based on time series allocation within periodic time frames to relocate the cluster again based on new UE positions. Hence, one advantage of this approach is that mobile network node addition and deletion can be implemented dynamically, especially since this is a continuous execution.

[0038] In some embodiments, the disclosed subject matter is primarily based on the use of a tensor-based decomposition model to relocate the mobile network node (e.g., UAV base station) position that best serves it assigned UEs, e.g., facilitating the highest data rate. The disclosed system utilizes the mobile network node as a base station (BS) that can be frequently repositioned to serve multiple users and ensuring that the energy cost of mobile network node movements should be minimized. More specifically, the repositioning of the mobile network nodes is conducted in a manner that provides an effective data rate to UEs. Since the corresponding optimization problem is non-convex, the disclosed subject matter utilizes and relies on the recent advances in artificial intelligence (AI) and proposes mobile network node tensor network decomposition techniques to reduce the memory footprint for an overall faster AI execution phase associated with minimal energy use.

[0039] As indicated above, the disclosed subject matter may be designed to utilize one-shot learning. As used herein, one-shot learning for computer vision tasks may be based on a special type of convolutional neural network (CNN) called a Siamese neural network (SNN). Notably, classic CNNs adjust their parameters throughout the training process to correctly classify each dataset. In contrast, Siamese neural networks are trained to evaluate the comparison (i.e., similarities and dissimilarities) between features present in two input datasets.

[0040] In some embodiments, the disclosed subject matter also accommodates the addition and / or deletion of mobile network nodes. Notably, the disclosed system / methodology is dynamic since it operates in a continuous execution mode where additional processing is not required when a mobile network node / UAV is to be added or deleted to the network.

[0041] In some embodiments, the disclosed subject matter takes the historical position(s) of the mobile network node so that it should have the ability to determine the future trajectory positions of UEs with the assistance of a tensor network based neural network model.

[0042] Moreover, the disclosed subject matter utilizes a reinforcement learning (RL) algorithm based on the tensor network-based algorithm. In some embodiments, this reinforcement learning based framework is employed based on the Q learning method. For example, a sequential decision-making setup may be considered, where the mobile network node interacts with the network environment, which evolves as a Markov process over discrete time steps. At each time step, the mobile network node receives (or obtains) a representation of the environment state st, and takes an action that is drawn from a set of possible actions according to a certain policy π. Notably, the mobile network node proceed to a new state st+1, and a reward rt is issued as a result.

[0043] In some embodiments, the disclosed subject matter receives and / or obtains UE related information, such as location details (e.g., latitude and longitude position data), bandwidth (e.g., data rate), carrier frequency, noise figure, thermal noise along with initial line of sight (LOS) details. Using this information, the disclosed system is configured to determine the best mobile network node / UAV position and assigns the rewards based on the data rate achieved. In some embodiments, SINR may also be added as a parameter for the overall mobile network node / UAV positioning, which corresponds to a position that the mobile network node can maintain an effective data rate.

[0044] One motivation for the disclosed solution is to have multiple mobile network node base stations to function autonomously. Further, the disclosed solution can be configured to ensure that the system can be dynamically updated to take into account the addition and / or deletion of at least one mobile network node. Notably, due to the one-shot learning approach utilized, the disclosed system can readily accommodate these associated dynamic updates.

[0045] The disclosed subject matter can be configured to achieve the ability to understand the environment on any difficult circumstances including the temporal movement of one or more UEs in addition to the energy consumption and trajectory angle of mobile network node. As referred herein, the trajectory angle may be the angle that exists between the position of the mobile network node (e.g., UAV base station) against the trajectory of a UE. One motivation for the present disclosure is based on evolving tensor decomposition techniques, which can be used to run any neural network model in a reduced memory footprint. Tensor networks provide fine-tuned tools to efficiently view the dimensions of the tensor(s) allocated within the tensor networks (as discussed below with respect to tensor decomposition functionality).

[0046] Solutions currently available are unable to understand and / or process the frequent modification of the observing space where the mobile network node and UEs are located. In contrast, the disclosed subject matter frequently runs a tensor network-based reinforcement algorithm in a time bound manner to understand the changing environment space such that the allocation of mobile network nodes efficiently accommodates the insertion / addition and deletion processes.

[0047] The disclosed system also is also capable of performing the trajectory prediction of UEs, which is executed on tensor decompositions such that the system understands the UE trajectory positions in a manner that is both memory-space efficient and energy efficient.

[0048] The disclosed subject matter affords a number of advantages over the current state of the art. For example, the disclosed solution is completely autonomous and does not require any assistance to determine the repositioning of the mobile network node (e.g., a UAV base station). Further, the disclosed solution is dynamic and can accommodate mobile network node addition and / or deletion on the fly (i.e., in real-time). The disclosed solution can determine the movement of UEs so efficiently that the mobile network node can be repositioned automatically. In some embodiments, mobile network nodes are automatically assigned to support a cluster of UEs based on the density of the UEs in a designated area. Further, the mobile network node can determine the best energy efficient movement (for the mobile network node) such that the energy requirement is efficiently satisfied. In some embodiments, mobile network node positioning can be allocated based on the height and / or position of the UEs. Moreover, the disclosed system has the ability to adapt to the continuous environment changes in an autonomous manner.

[0049] In some embodiments, the use of One-shot learning reduces the number of optimal components that are required to train the model. Notably, the use of One-shot learning has demonstrated better accuracy than any other technique experimented. Moreover, the disclosed techniques help in making the entire solution as a generalized principle and hence the solution is scalable to any number of new parameters which includes adding additional mobile network nodes, UAVs, robotic device, and UEs in any environment. Likewise, the disclosed solution efficiently determines the positioning of a newly added mobile network node on the density of UEs available in the subspace (e.g., mobile network node’s designated area of service). In some embodiments, the disclosed solution can effectively determine the amount and / or levels of energy consumed during the movement of a mobile network node.

[0050] The following description pertains to a proposed section model that starts from a tensor network, which assists in determining the trajectory position of one or more UEs. After determining trajectory position of the UEs, a Tensor Network-based reinforcement learning model is generated. Notably, this model assists in determining the optimal position of a mobile network node. Once the optimal position is determined, a cost-effective energy model is derived to reposition the mobile network node.

[0051] Notably, the disclosed subject matter is based on the use of tensor networks. Figure 2 illustrates a diagram depicting an exemplary tensor decomposition representation. Instead of having the tensor flow with regular networks, the present disclosure uses tensor decomposition networks. Notably, tensor decomposition in multilinear algebra may include any scheme for expressing a tensor as a sequence of elementary operations acting on other (often simple) tensors. In multilinear algebra, the tensor rank decomposition or the decomposition of a tensor is the decomposition of a tensor in terms of a sum of minimum tensors. The reason for using this technique is to reduce the number of trainable parameters so that it reduces the in-memory footprint. This technique takes input shape and output shape, followed by a rank. As used herein, ‘rank’ may refer to the order of tensor / dimension. Efficiently ranking the tensor helps to decompose the tensor efficiently so that it can be represented in the reduced dimension space.

[0052] The overall reduction in trainable parameters is observed in the disclosed subject matter as explained below. Tensor decomposition is considered unique if there exists only one combination of rank-1 tensors that sum to X (i.e., ‘tensor X’) up to a common scaling and / or permutation indeterminacy. Intuitively, this means that there is only one decomposition and a different arrangement of rank-1 tensors cannot be constructed to sum to X. As shown in Figure 2, tensor X 200 can be decomposed into a plurality of rank-1 sensors 2011…R. Tensor decompositions are unique under much more relaxed requirements on the tensor as compared to matrices. Hence, tensor decompositions are considered more rigid than matrices.

[0053] The use of One shot learning reduces the number of trainable components with the help of the similarity between the components. One shot learning uses a sparse data controller and similarity index to reduce the number of optimal components required to train the model. Currently, higher order tensors used in such experiments require an extremely large observation space, which makes it difficult to train in any workspace. Using one shot learning will help to solve this problem by reducing the higher order tensors to only optimal components. An example one shot learning model is depicted in Figure 3. In particular, Figure 3 illustrates a one-shot learning network model 300 that receives a plurality of input tensors 302 (e.g., input tensor 1 and input tensor 2) and produces an output 304 that includes a similarity number that indicates how the datasets of the tensors are similar.

[0054] In some embodiments, the implementation of One shot learning requires a model similar to that of the Convolution layer. For example, system 400 in Figure 4 is such an implementation includes i) generating the encodings 402 (e.g., feature vectors) for all the data inside the datasets with the help of the generated one-shot model(s) 401, ii) utilizing a customized layer 403 to compute the absolute difference (e.g., absolute value) between the encodings which provides the optimal distance between the data, and iii) computing and removing (e.g., using a sigmoid function 404) the closely correlated data from the datasets, which are ideally left with only the principal components remaining. Notably, the loss of system 400 can be represented by the following equation: Loss where Y is equal to 0 when the inputs are from the same class. Otherwise, Y is equal to 1. Further, m is the margin that defines the radius to indicate that dissimilar pairs beyond this margin will not contribute to the loss and is always greater than 0 and^^^^represents the Euclidean distance between the outputs of the model networks. The overall accuracy of the experiment has increased to the great extent with the help of one shot memory. The Difference Sigmoid Value that is received in the overall execution process is 0.5.

[0055] One intention of using one-shot learning is to attempt to reduce the amount of learning and then passing the dataset with the model learning tensors with the minimal different tensor to determine the difference in both of the results. This difference value is used as a similarity value for comparing the distance between the two datasets for generic classifications.

[0056] Figure 5 is a flow diagram that details a mobile network node repositioning procedure 500 according to some embodiments. As referred herein for Figure 5, a UAV base station (i.e., a UAV) is used as an example of a mobile network node to be used in procedure 500. Other mobile network nodes may be utilized with procedure 500 without departing from the scope of the disclosed subject matter.

[0057] In block 501, UE data is collected by a dynamic positioning manager (DPM) (e.g., a software-based positioning management module and / or algorithm locally stored in the UAV base station) and subsequently utilized for determining a new position for the UAV base station. For example, UE related parameter information such as location details (e.g., latitude and longitude data), bandwidth (e.g., data rate) data, carrier frequency data, noise figure data, thermal noise data, and / or initial line of sight (LOS) details may be obtained by the UAV base station via a wireless connection. In some embodiments, parameter data associated with the UAV base station may also be collected and utilized by the UAV base station and / or its DPM) to determine the UAV base station’s new position. Such UAV base station parameter data includes coordinates of the UAV base station, height of the UAV base station, trajectory angle of the UAV base station, and the like.

[0058] In block 502, a DB clustering scan is performed by the UAV base station and / or its DPM. For example, based on a UE’s longitude and latitude position, the UAV base station is configured to apply a DB clustering scan to determine one or more cluster groups. Notably, this clustering operation (e.g., conducted by the DPM) will group a plurality of UEs into one or more cluster groups, where each cluster group is assigned to one UAV base station.

[0059] In block 503, the UAV base station is assigned to a cluster group (by the UAV base station and / or its DPM). For example, the UAV base station and / or its DPM may be configured to convert the quantum circuits of a deep learning NN into a tensor network. In some embodiments, relevant quantum circuits are converted to a tensor network with the UE parameters collected in block 501.

[0060] In block 504, procedure 500 includes assessing the trajectory position of the UEs. In some embodiments, the UAV base station and / or its DPM is configured to provide the existing historical / past UE latitude and longitude positions as input into the tensor network (i.e., the tensor network based circuits). Such a tensor network can be used to predict the future positional parameters (e.g., UEPos data) of the UEs, which can be used to determine and / or understand the UE trajectory positions. In some embodiments, the trajectory positions of the UEs can be determined using a recurrent graph tensor network (RGTN). In some embodiments, the received future position data (e.g., UEPos data) of the UEs are converted by the UAV base station and / or its DPM into a vector space (e.g., the data representation for machine interpretable format). This data is subsequently stored by the UAV base station as input data for a reinforcement learning (RL) algorithm.

[0061] In block 506, an energy calculation is performed by UAV base station and / or its DPM to determine a most effective movement for the UAV base station. For example, using the vector space UE future position data (e.g., UEPos data), the UAV base station and / or its DPM is configured to calculate the amount of energy saved for one or more UAV base station movements. In some embodiments, the UAV base station and / or its DPM stores a list of energy efficient UAV movements and identifies the most energy efficient (i.e., least amount of energy used) UAV movement. In some embodiments, the UAV base station and / or its DPM may also be configured to determine the bandwidth and / or data rates existing between the UAV base station and each of the UEs.

[0062] In block 508, the UAV base station and / or its DPM is configured to calculate the Signal to Interference and Noise Ratio (SINR) of the signal communicated from each of the UEs to the UAV base station. In some embodiments, the SINR calculation may utilize the UEPos and UAVPos data in addition to the UAV height and bandwidth parameter data.

[0063] In block 509, the UAV base station and / or its DPM is configured to apply weights and rewards based on energy, trajectory, and data rate. Notably, weights and rewards may refer to the manner of calculating a change in position of the UAV base station based on the factors of energy, trajectory, and data rate of the UE devices configured in the environment.

[0064] In block 510, the UAV base station and / or its DPM is configured to apply tensor network-based reinforcement learning. In some embodiments, the UAV base station applies tensor network-based Q-learning by creating an observation space with the list of all of the UE cluster spaces. The UAV base station may also provide the UEPos data, UAVPos data, Bandwidth data, Noise value data, and LOS information as inputs.

[0065] In block 514, the DPM may direct the UAV base station to move to a new position. For example, the UAV base station and / or its DPM may begin applying an action by moving the UAV base station to any of the movement positions specified in the UAVPos data. IN some embodiments, the UAV base station and / or its DPM may be configured to calculate the data rate present at the new position and apply rewards. Based on the best data rate with the UEs in the cluster group (e.g., the best collective data rate determined considering all of the UEs in the cluster group), the DPM may determine the next UAV position. In some embodiments, the procedure 500 continues to block 502 to repeat the process based on time series analysis. More specifically, steps 502-516 may be repeated in preferred time intervals so that UAV base station(s) is repositioned autonomously.

[0066] In block 516, a UAV base station may be added or removed based on the presence of new requirements. From the above understanding of a new UAV position, the disclosed subject matter can infer if the convergence (e.g., how the UAV base station optimally communicates data via the data rate to UE devices) of the UAV base station is happening to an expected extent. If an anomaly (i.e., the UAV base station is not converging as expected and / or a deviation from the expected data rate) is detected, the addition or removal of UAV base station is done based on the received bandwidth (e.g., data rate). If there is no anomaly detected, the DPM may select the position of UAV base station based on which position obtains a higher reward. In some embodiments, the procedure 500 continues to block 502 to initiate a subsequent loop. More specifically, the steps 502-516 of procedure 500 may be repeated in preferred time intervals so that UAV base station(s) is repositioned autonomously.

[0067] In some embodiments, procedure 500 may be implemented by the UAV base station and / or its DPM utilizing One shot learning, which helps in reducing the optimal number of components and increases the accuracy to a great extent.

[0068] Figure 6 depicts an exemplary sequence diagram illustrating the mobile network node repositioning procedure 600. In block 601, the UAV base station and / or its DPM is configured to conduct a data collection operation. In some embodiments, UE related parameter information such as location details (e.g., latitude and longitude data), bandwidth (e.g., data rate) data, carrier frequency data, noise figure data, thermal noise data, and / or initial line of sight (LOS) details may be obtained by the UAV base station via a wireless connection. In some embodiments, parameter data associated with the UAV base station may also be collected and utilized (e.g., by UAV base station and / or its DPM) to determine the UAV base station’s new position.

[0069] In block 602, a DB clustering scanning operation is initiated by the UAV base station and / or its DPM. In some embodiments, the UAV base stations are assigned to generate cluster groups and subsequently convert the cluster groups to tensor networks.

[0070] In block 603, a UE trajectory calculation is performed by the UAV base station and / or its DPM. In some embodiments, the UAV base station and / or its DPM is configured to provide the existing historical / past UE latitude and longitude positions as input into a tensor network (e.g., the tensor network based circuits). Such a tensor network can be used to predict the future positional parameters (e.g., UEPos data) of the UEs, which can be used to determine and / or understand the UE trajectory positions. In some embodiments, the trajectory calculation includes the conversion of a predicted UE position into a vector space.

[0071] In block 604, a plurality of UAV movements and their respective energy consumption and / or efficiencies are determined by the UAV base station and / or its DPM. In some embodiments, the UAV base station and / or its DPM is configured to assess and / or understand the most energy efficient position for the UAV base station.

[0072] In block 605, reinforcement learning is applied by the UAV base station and / or its DPM. In some embodiments, the UAV base station and / or its DPM is configured to apply weights and rewards based on the energy efficiency, trajectory, and data rate associated with the optimal position for the UAV base station. This information is utilized by the DPM to determine the optimal UAV position.

[0073] In block 606, the UAV base station and / or its DPM repositions the UAV base station to its determined optimal position.

[0074] As indicated above, the disclosed subject matter may comprise a mobile network node (e.g., a UAV base station) and / or its DMP that is configured to apply a tensor network based DB clustering scan using UE positional data. The mobile network node may then utilize that information to determine one or more cluster groups for the UEs present in the network. In some embodiments, the mobile network node is configured to utilized collected UE parameter data to cluster a plurality of UE devices into one or more cluster groups, which are subsequently converted to one or more tensor networks (e.g., see block 1301 in Figure 13 described below).

[0075] Below are the randomly selected samples of UE devices and their location coordinates. Here, in Table 1, ‘LON’ stands for Longitude and ‘LAT’ stands for Latitude, respectively. TABLE 1

[0076] In some embodiments, the above input parameters are sent to clustering modelling with the aim of allocating the UAV base station for each cluster group. The following paragraph explains about the clustering analysis and how it is applied to UE positional data.

[0077] Clustering analysis (i.e., “clustering”) is basically an unsupervised learning method that divides the data points into several specific batches or groups, such that the data points existing in the same groups have similar properties while data points in different groups have different properties to some degree. In this context, the UE are clustered into multiple groups so that each mobile network node, i.e., UAV base station, can be assigned to an appropriate UE cluster group. In some embodiments, a DB Scan can be used to perform the clustering of UEs to determine the local optimal positions of each cluster to deploy UAV bases stations in the initial time slot. The UAV base stations are then permitted to these optimal positions. Fundamentally, all clustering methods use the same approach, i.e., calculation of similarities and which are then utilized to cluster the data points into groups or batches. As used herein, we will utilize the Density-based spatial clustering of applications with noise (DBSCAN) clustering method as an example although other methods may be used without departing from the scope of the disclosed subject matter. With the available Latitude (LAT) and Longitude (LON) data, a DBScan is applied and the UEs are grouped and / or converted into multiple clusters. Based on the clustering methodology, the samples fall into three different clusters as indicated in Table 2. TABLE 2

[0078] Now from the list, we have selected two sets of samples for our experiments: i) 2 UAV and 4 UE as experiment 1, and ii) 3 UAV and 4 UE as experiment 2. Samples considered for the first experiment (i.e., 2 UAV and 4 UE) are indicated in the below Table 3, which is represented for experiment 1. TABLE 3

[0079] Similarly, samples considered for the second experiment (e.g., 3 UAV and 4 UE) are indicated in Table 4 as follows. TABLE 4

[0080] Thus, two different experiment tables are presented, where Table 1 relates to two clusters and Table 2 is related to three clusters.

[0081] As indicated above, the disclosed subject matter may comprise a mobile network node (e.g., a UAV base station) and / or its DMP that is configured to assess and / or understand the trajectory position of UE and converting cluster groups into one or more Tensor state networks.

[0082] For example, the next step of the present disclosure (e.g., block 1302 in Figure 13) is to utilize the above experimental data to understand the trajectory position of the UE(s) using a tensor network approach. In order to assess or understand the pattern of each UE position, the mobile network node and / or its DPM needs to determine the future trajectory predictions based on the positional parameters of UEs. In some embodiments, the input parameters for trajectory predictions for experiment 1 and experiment 2 are shown in Table 5 and Table 6, respectively. TABLE 5 TABLE 6

[0083] In some embodiments, a tensor state network uses a three-layer architecture that helps determine the upcoming position of a UE based on the UE’s existing movement pattern. For example, the tensor network takes the existing moving altitude positions of the UEs as input. These UE positions are then passed to the tensor network space (e.g., vector space) and the weights are repeatedly reassigned based on a root mean square error (RMSE) measurement of RMSE. Once the RMSE value is reduced to the global minima, this position is determined to be the optimal next position of the UEs.

[0084] Figure 7 illustrates a block diagram of an exemplary tensor state network model 700. Notably, the structure of the Tensor state network model 700 is shown where V (e.g., input UE position 701) represents the input weight matrix, R (e.g., Hidden Layer 702) is the reservoir weight matrix, and W (e.g., output UE position 703) is the output weight matrix. In the input layer, the input vector is defined as un×1, and its dimensions is n×1. In the reservoir pool, the typical update equation for the tensor network is represented as: where xm×1is a vector of internal units in the reservoir pool. Similarly, Wm×n is the connection weight matrices that exist between input layer and reservoir pool, and ^^ ^^×^^ ^^^^is the recurrent weight matrices. In some embodiments, the reservoir pool is linearly connected to the output layer, which can be defined as: where y(n) is the output vector and ^^^^^^^^×^^represents the connection weight matrices between the reservoir pool and output layer. The measure root mean square error (RMSE) is used to evaluate the quality of this model, the expected value is and the actual result is yi. In some embodiments RMSE is defined as: where is the number of predictions, and wiare the weights. In the considered dataset, some UEs move unexpectedly (e.g., U-turn, right-angled turn, etc.), and the UE positions of these unexpected movements are assigned with the smaller weights.

[0085] Based on the past timesteps of latitude and longitude of the UE’s calculated the new positions. Below, a time step variation is disclosed.

[0086] The results of the trajectory positions corresponding to the first UE (e.g., new latitude and longitude values) for experiment 1 is indicated in Table 7. TABLE 7

[0087] For various reasons, we have taken the first UE in the above sample and taken the previous LAT and LON historical positions for the past 50 time steps. Afterwards, the trajectory model is executed and the UAV base station receives the new predicted latitude and new predicted longitude position for the UE. This process was then applied to all the UEs, and the UAV base station and / or its DPM predicted the new latitude and longitude positions. For each UE, the DPM obtains the LAT (Latitude) and LON (Longitude) data for the past 50 time steps between the time frame of 30 seconds. The trajectory position for the first experiment associated with the 2 UAV base stations and 4 UEs is represented in Table 8. TABLE 8

[0088] Similarly, trajectory position for the second experiment for 3 UAV and 4 UEs is determined and depicted in Table 9. TABLE 9

[0089] To achieve the predicted positions, the average RMSE associated with one shot learning as received in the experiment is 0.217 as shown in Table 10 Learning RMSE Method Score SVD 0.23 One Shot Learning 0.217 TABLE 10

[0090] As indicated above, the disclosed subject matter may comprise a mobile network node (e.g., a UAV base station) and / or its DMP that is configured to perform an energy calculation of UAV movement. Two experiments are considered as input for these energy calculations include: i) UAV clusters with the new predicted UE trajectories with 2 UAVs and 4 UEs and ii) UAV clusters with the new predicted UE trajectories with 3 UAVs and 4 UEs as indicated in Tables 11 and 12, respectively. TABLE 11 TABLE 12

[0091] Without considering environmental factors, the smaller the value of the summation of all the UAV base stations’ movement distances from their current positions to the positions in the next time slot is, then the less energy will be consumed. Such an Energy-Efficient Displacement Optimization (EEDO) problem can be formulated as the following definition.

[0092] Notably, at time t, the current position of a UAV BS is denoted as Li(t) ∈ R2, and the potential position of a deployed UAV BS in the next time slot is denoted as Pj (t) ∈ R2, i, j ∈ [1, n], where n is the number of UAVs, and wi,j is the energy cost for a UAV BS to fly from a position Lito a position Pj. The total energy cost of the UAV base stations’ displacements can be optimized by the following equations for calculating energy consumption movement.

[0093] The present disclosure assumes that the number of UAV-BSs is n, and the heights of the multiple UAV-BSs are the same and equal to a constant. At time t, the number of current positions and the number of predicted positions of the UAV-BSs are both equal to n. The coordinate of the ith current position is defined as Li(t) = [xi(t), yj(t)]T ∈ R2and the coordinate of the jth predicated position is defined as Pj(t) = [xj(t), yj(t)]T ∈ R2, ∀i,j ∈ [1, n], where T is a predefined time period slot between the predictions, and yi(t) are the latitude and longitude of the ith current position, respectively. Likewise, xj(t) and yj(t) are the latitude and longitude of the jth predicated position, respectively. The vertex labeling of Li(t) and Pj(t) are denoted as A[Li(t)] and B[Pj(t)], respectively. The weight between Li(t) and Pj(t) is defined as [Li(t)][Pj(t)], which can be calculated by the distance between Li(t) and Pj(t). The present disclosure aims at minimizing the energy cost, which is equivalent to determining the smallest sum of total weights. In some embodiments, a bipartite matching method is used. According to this method algorithm, A[Li(t)] and B[Pj(t)] will be explained in the below equation for vertex labeling calculations. ^^[^^^^(^^)] = 0, ∀^^,^^∈ [1, ^^]

[0094] For purposes of example and illustration, each cluster may be assigned with at least one UAV, which is designated in accordance to the naming convention of a UAV base station as indicated below in Table 13. TABLE 13

[0095] The UAV naming convention has been modified as shown below in Table 14 for representation. TABLE 14

[0096] The predicted UAV base station position at an altitude of 10 meters as per the energy driven approaches are shown below.

[0097] In some embodiments, the present disclosure utilized the Hungarian algorithm to understand and / or assess the distance between the current position of the UAVs with the predicted position of UAVs. The optimal values are presented in Table 15. TABLE 15

[0098] The distance of the UAV base station movement is based on the distance matrix from another UAV base station and on its own such that the distance of the reposition trajectory from the current position to the predicted position for this UAV base station is 267 meters (as shown in Table 15). The other distances can be deduced in the same way, and the total distance of this matching scheme will be equal to 1452 meters (e.g., 267m + 467m + 718m). The present disclosure implements a KM-based matching algorithm and use the values as the input information. The output repositions matching scheme (UAV1 → UAV1’, UAV2 → UAV2’, UAV3 → UAV3’) is shown below in Table 16. TABLE 16

[0099] Based on this experiment, it would be ideal to reduce these UAV positions and bringthem near to 1256 meters, where ^^^^^^^^ = 1256 meters.

[0100] As indicated above, the disclosed subject matter may comprise a mobile network node (e.g., a UAV base station) and / or its DMP that is configured to calculate a Physical Layer parameter like SINR towards the UAV base station and UE positions.

[0101] The signal-to-interference-and-noise ratio (SINR) is of key importance for the analysis and design of wireless networks. For addressing new requirements imposed on wireless communication, in particular high availability, a highly accurate modeling of the SINR is needed. The present disclosure proposes a stochastic model of the SINR distribution where shadow fading is characterized by random variables. Therein, the impact of shadow fading on the user association is incorporated by modification of the distributions involved. The SINR model is capable of describing all parts of the SINR distribution in detail, especially the left tail that is of interest for studies of high availability. For example, SINR may be calculated using the following equation: ^^^^^^^^(^^) = ^^ / (1 + ^^)where P is the power of the incoming signal of interest, I is the interference power of the other (interfering) signals in the network, and N is some noise term, which may be a constant or random value. Like other ratios in electronic engineering and related fields, the SINR is often expressed in decibels (dB).

[0102] The superposition coding (SC) is used at the UAV base station to transmit messages for users located in the same cluster. Notably, SC may be used to encode different messages into a single signal while assigning them different power values. The successive interference cancellation (SIC) can be used at the receiver side for signal detection. The received SINR at UEi (e.g., SINR at specific user equipment positions) is expressed as:

[0103] The simplified term SNR= Transmitted Power / (Noise Figure*Thermal Noise), such that SNR= 508370070223.5405. For both experiments, 1000 episodes, the SNR value varies close to 508370070223.5405.

[0104] At this stage, the mobile network node and / or its DPM inputs all the calculated values into tensor network-based RL framework.

[0105] As indicated above, the disclosed subject matter may comprise a mobile network node (e.g., a UAV base station) and / or its DMP that is configured to apply tensor network based reinforcement learning. For example, inputs for the aforementioned experiment 1 with 4 UEs and 2 UAV base stations in indicated below in Table 17. TABLE 17

[0106] Similarly, Inputs for the experiment 2 with 4 UEs and 3 UAV are depicted in Table 18. TABLE 18

[0107] Reinforcement learning (RL) technique combined with a tensor neural network (TNN), has the potential to handle high-dimensional inputs, learn patterns, and solve complex problems efficiently.

[0108] To use reinforcement learning successfully in situations approaching real-world complexity, however, agents are tasked to derive efficient representations of the environment from high-dimensional sensory inputs and use these to generalize past experiences to a new situation. Tasks in which an agent interacts with an environment, E, may be considered in a sequence of actions, observations, and rewards. At each time-step, the agent selects an action at from the set of legal UAV base station movement actions, A = {1, ... , K} where A is the element of energy consumed by UAV base station movements.

[0003]

[0109] A tensor network is used both as the ‘Actor network’ and ‘Critic network’, which helps in efficient weight adjustment such that the tensor network is able to determine the optimal UAV base station position for each episode. Tensor decomposition efficiently reduces the number of operational parameters that is used for the neural network in a weight matrix assignment. This approach provides the best results in terms of a reduced Mean Average Error (MAE) value with efficient memory footprints as tensor decomposition serves as an effective tool in bringing the reduced weight matrix to represent the complex neural network. In some embodiments, a policy network is generated and used as an actor network, while a target network acts as the critic network. The basic working is explained in the flow diagram 800 in Figure 8.

[0110] Even though some initial works consider RL UAV base station placement in the presence of LoS links (which employ deep deterministic policy gradient RL), the full potentials of RL for 3D networks still need to be assessed. Integrating dueling network architectures with RL has shown its merits in dramatically improving and generalizing learning in different domains (e.g., an Atari domain). In the same regard, dueling RL, compared to other RL advances like double RL, as discussed in the UAV sensing related application context, demonstrates an improvement in the learning speed. Nevertheless, there are surprisingly very few investigations conducted about integrating dueling RL and mobile network nodes. The present disclosure instead attempts to solve the non-convex optimization problem of the 3D-network resources management using the proposed dueling RL based framework. Notably, the disclosed subject matter seeks to bring forth the most-recent advances in AI to efficiently solve the UAV base station placement and resource management while maximizing both the users' data rate and to reduce fairness equally. Notable contributions are summarized as follows:

[0111] For a bandwidth ^^^^, the data rate of UEi at time step, τ, is given by an equation representative of the sum rated based on individual user equipment: ^^^^,^^=^^^^^^^^^^2(1 + ^^^^^^^^^^,^^)

[0112] The present disclosure proposes to optimize the UAV BS placement and power allocation that maximizes the total users' data rate and fairness. First, the sum users' data rate (for all user equipment) at time step τ is defined as: and, using the Jain's fairness index, the users' fairness based on data rate can be represented by the following equation:

[0113] The optimization for calculating UAV base station positions is formulated via the following equation: ^^^^^^^^ − 100 ≥ ^^^^^^^^ ≤ ^^^^^^^^ + 100 ^^^^^^^^ = ^^^^^^ ^^^^^^^^^^^^^^^^ ^^^^ ^^^^^^^^^^^^

[0114] In an experiment with 4 UE and 2 UAV base stations, the experiment can relate to each UAV base station as follows in Table 19. TABLE 19

[0115] In a second experiment with 4 UEs and 3 UAV base stations, the results for each UAV base station is respectively depicted below in Tables 20-22. Notably, Table 20 presents the UAV base station 1 result, Table 21 presents the UAV base station 2 result, and Table 22 presents the UAV base station 3 result as follows: Sum Rate Fairness Sum rate 1 Sum rate 2 Batch (^^^^^^^^^^) (^^ ^^ ^^) (^^1,^^)(^^2,^^)1 9.85 67.37 4.5 5.08 2 15.13 57.8 10.03 5.1 3 18.29 58.12 10.07 8.22 4 18.16 63.15 9.42 8.74 5 18.51 66.42 9.32 9.18 6 17.1 62.5 8.57 8.53 7 18.67 65.42 9.2 9.47 8 18.1 67.77 8.59 9.51 9 16.17 63.22 8.02 8.15 10 18.17 71.2 9.08 9.19 TABLE 20 Sum Rate Fairness Sum rate 1 Batch (^^^^^^^^^^) (^^ ^^ ^^)(^^1,^^)1 9.5 68.21 9.5 2 15.1 57.4 15.1 3 18.31 59.22 18.31 4 18.1 63.2 18.1 5 18.27 66.21 18.27 6 17.35 61.67 17.35 7 18.2 64.45 18.2 8 18.34 68.01 18.34 9 17.21 63.11 17.21 10 18.3 72 18.3 TABLE 21 Sum Rate Fairness Sum rate 1 Batch^^^ (^^^^^^^^^) (^^^^)(^^1,^^)1 9.8 67.21 9.8 2 14.9 58.4 14.9 3 18.34 57.22 18.34 4 18.24 62.2 18.24 5 18.31 65.21 18.31 6 17.21 62.67 17.21 7 18.35 63.4 18.35 8 18.56 62 18.56 9 17.4 64.21 17.4 10 18.1 69 18.1 TABLE 22

[0116] The maximum sum rate or data rate is achieved in a shorter batch itself. For example, the convergent is much faster in tensor-network-based approach as compared to deep neural network model.

[0117] In some embodiments, the disclosed subject matter provides the new UAV base station movement in meters from the current UAV base station positions and considers data rate and energy expended.

[0118] As indicated above, the disclosed subject matter may comprise a mobile network node (e.g., a UAV base station) and / or its DMP that is configured to calculate the new UAV base station position (UAVPos). For example, inputs for calculating the UAV base station position is provided in Table 23. New New Energy LON LAT Cluster Predicted predicted Saving LON LAT Movement in meters 28.17858 -25.739 2 29.18988 -24.8 1256 28.1766 -25.738 2 29.2975 27.8 1256 28.12514 26.2667 1 29.25 27.27 1256 28.10144 26.1057 3 28.20301 -25.0042 1256 TABLE 23

[0119] In some embodiments, applying the action state and rewards helps to determine the best optimal rewards based on the action. The process can be repeated for 100 episodes and can subsequently determine the best optimal UAV base station position once completed. The process can be repeated numerous times after a specific time interval so as to repeatedly move the UAV base station position. Examples for an experiment with 4 UEs and 3 UAV base stations is presented int Table 24, while the experiment with 4 UEs and 2 UAV base stations is presented in Table 25. Energy Predicted predicted predicted New New Saving UAV 1 UAV 2 UAV 3 LON LAT Cluster Predicted predicted SINR Movement movement Movement in movement LON LAT in meters in meters meters in meters8.17858 -25.7388 2 29.18988 -24.8 1256 5.0837E+11 1270 1244 13088.1766 -25.738 2 29.2975 27.8 1256 5.0837E+11 1270 1244 13088.12514 26.2667 1 29.25 27.27 1256 5.0837E+11 1270 1244 13088.10144 26.1057 3 28.20301 -25.0042 1256 5.0837E+11 1270 1244 1308 TABLE 24 Ne Energy Predicted predicted AT Ne w New LON L w Saving SINR UAV 1 UAV 2 Cluster Predicted Predicted Mo movement Movement LON LON vement in meters in meters in meters 28.17858 -25.7388 2 27.17006 -26.8888 12565.08E+11 1268 129628.1766 -25.738 2 29.1896 -26.8854 12565.08E+11 1268 129627.83239 -26.5372 1 28.8119 -27.5388 12565.08E+11 1268 129628.12514 -26.2667 1 28.1315 -27.3776 12565.08E+11 1268 1296TABLE 25

[0120] In both the experiments, the final predicted UAV base station positions in meters are almost near to the energy saving value in meters. It clearly highlights the repositioning of UAV base stations to approximately near to energy saving values by considering the change in trajectory of the UE.

[0121] As indicated above, the disclosed subject matter may comprise a mobile network node (e.g., a UAV base station) and / or its DMP that is configured to conduct the addition or deletion of one or more UAV base stations.

[0122] The convergence data from the above tensor based reinforcement algorithm is taken as an input for deciding the need of addition or deletion of a UAV base station.

[0123] To further illustrate the recommendation process for a new UAV base station addition or deletion, the present disclosure details some experiments where 2 UAV base stations tried with 3 UEs, which were taken initially and incremented the number of UEs to see the consistent convergence in achieving the data rate.

[0124] After the number of UEs was increased from 4 UEs to 5 UEs, a sudden drop in the convergence is exhibited.

[0125] To explain about the convergence, the sum data rate achieved during 400 EPOCHS always surface between 17.5 Gbps (gigabits per second) to 19.0 Gbps (which is ideal as per the 5G specification) for each UE until the number of UEs is 4 with 2 UAV base stations. As per the learning, if the threshold of Gbps is within the range of 17.5 to 19.0 Gbps in 400 EPOCHS, after the training of 1000 EPOCHS, the Gbps always settle at approximately at 18.5 Gbps.

[0126] However, the sum data rate has dropped further below 17.5 Gbps when the UE are incremented because the distance in which the UAV base station needs to provide the data is getting wider as the number of UEs is increasing, and hence it reduces the data rate. The same scenario occurred when the number of UEs decreased and the resulting data rate convergence happened.

[0127] When the UEs are reduced, the data rate for each UE is increased and hence the UAV base station has more than enough signal strength and data rate and hence one of the UAV base stations in the service pool can be recommended for removal.

[0128] Example pseudocode for the same is presented below: #Calculate the average number of EPOCHS where the data rate reaches 17.5 gbps in the #past 10 experiments, for example 400 EPOCHS. average_Convergence_Epochs=400 Min_Threshold_ingbps=17.5 max_threshold_ingbps=19.0 # Check in new experiment if the experiment convergence is equal to average convergence if(new_Experiment_Convergence=average_Convergence_Epochs): #Check if the data rate is greater than Min_Threshold_inmbps and less than max_threshold_ingbps if(experiment_data_rate>=Min_Threshold_ingbps): #Recommend new UAV base station addition recommendUAVaddition() elif(experiment_data_rate<=max_threshold_ingbps): #Remove one of the UAV base station recommendUAVRemoval() else: # No action required and continue with the existing number of UAV base stations and UEs Continue. Average Data Rate in New Experiment New Experiment Number in every 100 EPOCHs Data Rate in every Data Rate in every 100 EPOCHS with 2 UAVs and 4 100 EPOCHs with 2 100 EPOCHs with 2 UEs UAVs and 5 UEs UAVs and 2 UEs 1 8.5 5.95 9.25 2 12.98 8.92 12.45 3 16.75 12.95 16.8 4 17.61 15.25 19.01 5 18.08 17.08 18.54 6 18 17.59 18.48 TABLE 25

[0129] From the above Table 25, which presents Convergence Inference data, below are the inferences. First, Column 3 in Table 25 infers that this experiment recommends for addition of 1 UAV. Moreover, Column 4 in Table 25 infer that this experiment recommends for removal of 1 UAV. Additional Data rate information is depicted in plots 901-903 in graph 900 of Figure 9.

[0130] Experimental Results comparing the existing UAV base station Repositioning vs. UAV base station Trajectory Optimization through Tensor Networks

[0131] As indicated above, the present disclosure may achieve greater accuracy with One Shot Accuracy.

[0132] Table 26 indicates the accuracy with respect to the One Shot Learning against the Self supervised algorithm which is highly utilized for most of the reinforcement learning. Learning Techniques Accuracy Self-Supervised Algorithm 70.80% One shot solution (un- supervised) 96% TABLE 26

[0133] Table 27 indicates the total execution time for Legacy RL and One-Shot Learning. Learning Techniques Accuracy Legacy Reinforcement learning 259 Seconds One shot learning 125 seconds TABLE 27

[0134] The experiments above show that the one-shot solution provides better accuracy than self-supervised algorithm with a minimal number of components in a less time.

[0135] Advantages of using energy consumption UAV movement vs. Legacy UAV movement

[0136] Legacy Approach of UAV movement – Without providing a reward for Energy efficient movements

[0137] Energy-Minimization – Applying a reward for Energy Efficient movements Average Average Energy Average acceleration power efficiency speed (m / s) (m2 / s) (Watts) (kbits / Joule) Legacy Approach of UAV movement 8.4 3.74 585.66 16.32 Energy- minimization 29.61 1.1 102.74 19.6 TABLE 28

[0138] The above experiment results shown in Table 28 and graph 1000 in Figure 10 indicate that the energy minimization approach (i.e., approach “A”) provides reduced power usage as compared to the legacy-based UAV movement (i.e., approach “B”). Notably, graph 1000 indicates that the energy efficiency 1001A of approach A is greater than energy efficiency 1001Bof approach B. Moreover, graph 1000 indicates a bar representing average power 1002Aof approach A that is less than a bar representing the average power 1002Bof approach B. This is notable because graph 1000 indicates that the average UAV speed 1004A of approach A is still significantly greater than approach B’s average speed 1004B.

[0139] SINR ratio for better efficient data rate

[0140] A healthy SINR is critical to maximizing data capacity and throughput. SINR is one of the most fundamental and useful parameters. Similar to how a person’s body weight is indicative of susceptibility to a wide range of health issues, SINR is an excellent indicator of overall network health. Even though two subscribers may use the same amount of spectrum (i.e., the number of Orthogonal Frequency-Division Multiple Access (OFDMA) sub-carriers), their respective signal quality (i.e., SINR) determines their throughput. This is because the higher the SINR, the higher the quadrature amplitude modulation (QAM) that can be achieved, and hence the higher the data rate that the subscriber will experience. In the lower frequency bands and for busy cell sites, a low SINR has an outsized negative impact so SINR optimization is particularly important in these cases.

[0141] The experiment shown in graph 1100 of Figure 11 indicates that a consistent maintenance of data rate helps in maintaining the better data rate (e.g., Sum rate 1101 as compared to SINR 1102).

[0142] Optimized UAV base station position is achieved by providing rewards based on the action that is performed on each of the movement performed by UAV base station.

[0143] Experiment for addition and removal of UAV base station(s)

[0144] As previously indicated by the present disclosure, one experiment already inferred the below statements: 1) When the number of UAV BSs is 2 and when the number of UEs reaches 5, then the data rate of each UE decreases in the convergence of RL. 2) When the number of UAV BSs is 2 and when the number of UEs reduces to 2, then data rate of each UE increases in the convergence of RL. 3) For point 1, the number of UAV BSs has increased to 3. 4) For point 2, the number of UAV BSs has increased to 2.

[0145] Below in Table 29 is the experiment results as per the point 3 (Addition) and point 4 (Removal) and inferred the data rates.

[0146] Number in 100 UAV Addition UAV Removal EPOCHS 1 8.52 8.43 2 12.56 12.98 3 16.72 16.08 4 17.58 17.52 5 18.34 18.21 6 18.01 17.98 TABLE 29

[0147] The above UAV addition and deletion ensures that the data rate is reaching the consistency after the recommended fulfillment of UEs.

[0148] Efficient data rate based on better energy consumption UAV movement and trajectory position of UEs.

[0149] Communications in 5G can be significantly faster than 4G, delivering up to 20 Gigabits- per-second (Gbps) peak data rates and 100+ Megabits-per-second (Mbps) average data rates. 5G has more capacity than 4G.5G is designed to support a 100x increase in traffic capacity and network efficiency.

[0150] The current solution available has the ability to provide a best data rate to more than 18GB consistently in recent experiments as shown in Table 30. Batches Sum of Rate training 1 9.35 2 14.59 3 18.22 4 18.16 5 18.81 6 18.27 7 18.66 8 18.61 9 16.9 10 18.56 TABLE 30

[0151] Figure 12 depicts a mobile network node 1200 in accordance with some embodiments. As used herein, network node refers to equipment capable, configured, arranged and / or operable to communicate directly or indirectly with a UE and / or with other network nodes or equipment, in a telecommunication network. Examples of network nodes include, but are not limited to, access points (APs) (e.g., radio access points), base stations (BSs) (e.g., radio base stations, Node Bs, evolved Node Bs (eNBs) and NR NodeBs (gNBs)), O-RAN nodes or components of an O-RAN node (e.g., O-RU, O-DU, O-CU). Notably, the mobile network nodes described herein are mobile devices, such as a UAV base station. In some embodiments, mobile network node 1200 has the same functionality as the network node and / or base station described in Figure 15.

[0152] Mobile network node 1200 includes a processing circuitry 1202, a memory 1204, a communication interface 1206, and a power source (not shown). The processing circuitry 1202 may comprise a combination of one or more of a microprocessor, controller, microcontroller, central processing unit, digital signal processor, application-specific integrated circuit, field programmable gate array, or any other suitable computing device, resource, or combination of hardware, software and / or encoded logic operable to provide, either alone or in conjunction with other network node components, such as the memory 1204, to provide mobile network node 1200 functionality.

[0153] In some embodiments, the processing circuitry 1202 includes a system on a chip (SOC). In some embodiments, the processing circuitry 1202 includes one or more of radio frequency (RF) transceiver circuitry and baseband processing circuitry. In some embodiments, the radio frequency (RF) transceiver circuitry (not shown) and the baseband processing circuitry (not shown) may be on separate chips (or sets of chips), boards, or units, such as radio units and digital units. In alternative embodiments, part or all of RF transceiver circuitry and baseband processing circuitry may be on the same chip or set of chips, boards, or units.

[0154] The memory 1204 may comprise any form of volatile or non-volatile computer-readable memory including, without limitation, persistent storage, solid-state memory, remotely mounted memory, magnetic media, optical media, random access memory (RAM), read-only memory (ROM), mass storage media (for example, a hard disk), removable storage media (for example, a flash drive, a Compact Disk (CD) or a Digital Video Disk (DVD)), and / or any other volatile or non-volatile, non-transitory device-readable and / or computer-executable memory devices that store information, data, and / or instructions that may be used by the processing circuitry 1202. The memory 1204 may store any suitable instructions, data, or information, including a computer program, software, an application including one or more of logic, rules, code, tables, and / or other instructions capable of being executed by the processing circuitry 1202 and utilized by the network node 1200. The memory 1204 may be used to store any calculations made by the processing circuitry 1202 and / or any data received via the communication interface 1206. In some embodiments, the processing circuitry 1202 and memory 1204 is integrated. In some embodiments, memory 1204 is configured to store DPM 1208. Notably, DPM 1208 may include operations (as set forth herein and / or in the steps set forth in Figure 13) for dynamically optimizing UAVs using tensor networks that can be executed by processing circuitry 1202.

[0155] The communication interface 1206 is used in wired or wireless communication of signaling and / or data between a network node, access network, and / or UE. As illustrated, the communication interface 1206 comprises port(s) / terminal(s) (not shown) to send and receive data, for example to and from a network over a wired connection. The communication interface 1206 also includes radio front-end circuitry that may be coupled to, or in certain embodiments a part of, the antenna.. The radio front-end circuitry may receive digital data that is to be sent out to other network nodes or UEs via a wireless connection. The radio front- end circuitry may convert the digital data into a radio signal having the appropriate channel and bandwidth parameters using a combination of filters and / or amplifiers. The radio signal may then be transmitted via the antenna to another network node and / or UE(s). Similarly, when receiving data, the antenna may collect radio signals which are then converted into digital data by the radio front-end circuitry. The digital data may be passed to the processing circuitry. In other embodiments, the communication interface may comprise different components and / or different combinations of components. Mobile network node 1200 may also include a mobility module 1210 that is configured to facilitate movement of mobile network node 1200. Mobile module 1210 may include hardware components, software components, and / or a combination of hardware and software components that are configured to receive instruction signals and / or direction from DPM 1208 and / or processing circuitry 1202.

[0156] Figure 13 is a flow chart illustrating method 1300 depicting exemplary operations for dynamically optimizing mobile network nodes using tensor networks according to one embodiment. Operations of the mobile network node (e.g., UAV base station) which may be implemented using the structure of the block diagram of Figure 12, will now be discussed with reference to the flow chart of Figure 13 according to some embodiments. For example, one or more modules may be stored in memory 1204 of Figure 12, and these modules may provide instructions so that when the instructions of a module are executed by respective processing circuitry 1202, the mobile network node 1200 performs respective operations 1301-1305 of the flow chart 1300. In some embodiments, memory 1204 may store a DPM application 1208 that performs the operations described herein. In some embodiments, the mobile network node is configured to operate autonomously. In some embodiments, the mobile network node includes a UAV base station, a drone, or a mobile base station. In some embodiments, each of the plurality of UEs is mobile.

[0157] In block 1301, method 1300 includes collecting user equipment, UE, parameter data from each of a plurality of UE devices operating in the network. In some embodiments, the UE parameter data includes one or more of: longitude and latitude location information, bandwidth data, height data, carrier frequency data, noise figure data, thermal noise data, and initial line of sight (LoS).

[0158] In block 1302, method 1300 includes utilizing the collected UE parameter data to cluster the plurality of UE devices into one or more cluster groups, which are subsequently converted to a respective one or more tensor networks.

[0159] In block 1303, method 1300 includes determining the future trajectory positions for each of the UE devices included in a cluster group assigned to the mobile network node using the one or more tensor networks.

[0160] In block 1304, method 1300 includes determining a most energy-efficient movement for the mobile network node to support the future trajectory positions of the plurality of UE devices. In some embodiments, determining the most energy efficient movement further comprises calculating an amount of the energy saved by the most energy efficient movement for the mobile network node. In some embodiments, determining a most energy-efficient movement for the mobile network node includes determining an angle existing from the mobile network node to a future trajectory position for at least one of the UE devices. In some embodiments, tensor decomposition is used to determine the future trajectory positions.

[0161] In block 1305, method 1300 includes establishing a new position for the mobile network node based on the most energy-efficient movement and an optimal data rate prediction achievable by the mobile network node. In some embodiments, establishing a new position for the mobile network node includes i) utilizing a reinforcement learning trained tensor network to determine the optimal data rate achievable by the mobile network node for the UE devices in the cluster group in accordance with the most energy efficient movement, and ii) establishing a new position for the mobile network node based on the determined data rate. In some embodiments, establishing a new position for the mobile network node is conducted in a single orthogonal frequency division multiplexing (OFDM) time slot.

[0162] In some embodiments, method 1300 further comprises using the tensor network to assign the mobile network node to a plurality of UEs based on a density exhibited by the UEs.

[0163] In some embodiments, method 1300 further comprises adding or deleting at least one mobile network node to or from the network based on a data rate received from each of the UE devices included in the cluster group.

[0164] In some embodiments, the mobile network node is configured to repeat the steps of method 1300 in preferred time intervals.

[0165] To summarize the present disclosure’s aim to achieve the best data rate, the disclosed subject matter took the positional parameters of the UE latitude and longitude positions under the tensor network framework. The disclosed subject matter then applies a DB scan for efficient clustering of UE devices such that each UAV base station can be assigned to one cluster.

[0166] The present disclosure as a whole is focused on One shot learning which reduces the number of observation components. Based on UE movements, the disclosed subject matter is configured to predict the future movements (e.g., trajectories) of UE positions and calculate the most energy efficient movement so as to provide longer energy sustainability. The system further applies a Tensor network-based reinforcement algorithm based on trajectory UE movements and energy Efficient movements as input.

[0167] In some embodiment, the disclosed subject matter may then calculate the SINR relevant to UE and UAV BS movements. Moreover, the disclosed subject matter may pass the SINR and bandwidth as input and perform an action by moving UAV BS positions related to new UE positions. The disclosed subject matter can also be configured to arrive to the best UAV positions where it achieves the best data rate based on energy efficient measurement(s).

[0168] The disclosed subject matter can also repeat the experiment in time bound positions.

[0169] The disclosed subject matter may apply one-Shot learning, which helps in reducing the overall learning time which implicitly reduces the model running time. There is a significant increase in terms of accuracy when using one shot learning.

[0170] This solution can be scaled up by adding additional UAV BSs, UEs, and robotic tools as it is more generalization technique. Further, applying this one-shot learning has significantly reduced the RMSE value. Finally, the disclosed subject matter has demonstrated the scalability and sustainability (energy efficiency) of the proposed algorithm via rigorous experimental evaluations on a dataset. Since the solution is based on one-shot learning, which requires a low footprint, it can be run either in a centralized cloud or on any resource- constrained edge node with sufficient compute and storage. In some embodiments, the disclosed subject matter may include any UAV BS that serves as an edge node.

[0171] Some of the embodiments contemplated herein will now be described more fully with reference to the accompanying drawings. Embodiments are provided by way of example to convey the scope of the subject matter to those skilled in the art.

[0172] Figure 14 shows an example of a communication system 1400 in accordance with some embodiments.

[0173] In the example, the communication system 1400 includes a telecommunication network 1402 that includes an access network 1404, such as a radio access network (RAN), and a core network 1406, which includes one or more core network nodes 1408. The access network 1404 includes one or more access network nodes, such as network nodes 1410a and 1410b (one or more of which may be generally referred to as network nodes 1410), or any other similar 3rdGeneration Partnership Project (3GPP) access nodes or non-3GPP access points. Moreover, as will be appreciated by those of skill in the art, a network node is not necessarily limited to an implementation in which a radio portion and a baseband portion are supplied and integrated by a single vendor. Thus, it will be understood that network nodes include disaggregated implementations or portions thereof. For example, in some embodiments, the telecommunication network 1402 includes one or more Open-RAN (ORAN) network nodes. An ORAN network node is a node in the telecommunication network 1402 that supports an ORAN specification (e.g., a specification published by the O-RAN Alliance, or any similar organization) and may operate alone or together with other nodes to implement one or more functionalities of any node in the telecommunication network 1402, including one or more network nodes 1410 and / or core network nodes 1408.

[0174] Examples of an ORAN network node include an open radio unit (O-RU), an open distributed unit (O-DU), an open central unit (O-CU), including an O-CU control plane (O- CU-CP) or an O-CU user plane (O-CU-UP), a RAN intelligent controller (near-real time or non-real time) hosting software or software plug-ins, such as a near-real time control application (e.g., xApp) or a non-real time control application (e.g., rApp), or any combination thereof (the adjective “open” designating support of an ORAN specification). The network node may support a specification by, for example, supporting an interface defined by the ORAN specification, such as an A1, F1, W1, E1, E2, X2, Xn interface, an open fronthaul user plane interface, or an open fronthaul management plane interface. Moreover, an ORAN access node may be a logical node in a physical node. Furthermore, an ORAN network node may be implemented in a virtualization environment (described further below) in which one or more network functions are virtualized. For example, the virtualization environment may include an O-Cloud computing platform orchestrated by a Service Management and Orchestration Framework via an O-2 interface defined by the O-RAN Alliance or comparable technologies. The network nodes 1410 facilitate direct or indirect connection of user equipment (UE), such as by connecting UEs 1412a, 1412b, 1412c, and 1412d (one or more of which may be generally referred to as UEs 1412) to the core network 1406 over one or more wireless connections.

[0175] Example wireless communications over a wireless connection include transmitting and / or receiving wireless signals using electromagnetic waves, radio waves, infrared waves, and / or other types of signals suitable for conveying information without the use of wires, cables, or other material conductors. Moreover, in different embodiments, the communication system 1400 may include any number of wired or wireless networks, network nodes, UEs, and / or any other components or systems that may facilitate or participate in the communication of data and / or signals whether via wired or wireless connections. The communication system 1400 may include and / or interface with any type of communication, telecommunication, data, cellular, radio network, and / or other similar type of system.

[0176] The UEs 1412 may be any of a wide variety of communication devices, including wireless devices arranged, configured, and / or operable to communicate wirelessly with the network nodes 1410 and other communication devices. Similarly, the network nodes 1410 are arranged, capable, configured, and / or operable to communicate directly or indirectly with the UEs 1412 and / or with other network nodes or equipment in the telecommunication network 1402 to enable and / or provide network access, such as wireless network access, and / or to perform other functions, such as administration in the telecommunication network 1402.

[0177] In the depicted example, the core network 1406 connects the network nodes 1410 to one or more hosts, such as host 1416. These connections may be direct or indirect via one or more intermediary networks or devices. In other examples, network nodes may be directly coupled to hosts. The core network 1406 includes one more core network nodes (e.g., core network node 1408) that are structured with hardware and software components. Features of these components may be substantially similar to those described with respect to the UEs, network nodes, and / or hosts, such that the descriptions thereof are generally applicable to the corresponding components of the core network node 1408. Example core network nodes include functions of one or more of a Mobile Switching Center (MSC), Mobility Management Entity (MME), Home Subscriber Server (HSS), Access and Mobility Management Function (AMF), Session Management Function (SMF), Authentication Server Function (AUSF), Subscription Identifier De-concealing function (SIDF), Unified Data Management (UDM), Security Edge Protection Proxy (SEPP), Network Exposure Function (NEF), and / or a User Plane Function (UPF).

[0178] The host 1416 may be under the ownership or control of a service provider other than an operator or provider of the access network 1404 and / or the telecommunication network 1402, and may be operated by the service provider or on behalf of the service provider. The host 1416 may host a variety of applications to provide one or more service. Examples of such applications include live and pre-recorded audio / video content, data collection services such as retrieving and compiling data on various ambient conditions detected by a plurality of UEs, analytics functionality, social media, functions for controlling or otherwise interacting with remote devices, functions for an alarm and surveillance center, or any other such function performed by a server.

[0179] As a whole, the communication system 1400 of Figure 14 enables connectivity between the UEs, network nodes, and hosts. In that sense, the communication system may be configured to operate according to predefined rules or procedures, such as specific standards that include, but are not limited to: Global System for Mobile Communications (GSM); Universal Mobile Telecommunications System (UMTS); Long Term Evolution (LTE), and / or other suitable 2G, 3G, 4G, 5G standards, or any applicable future generation standard (e.g., 6G); wireless local area network (WLAN) standards, such as the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standards (WiFi); and / or any other appropriate wireless communication standard, such as the Worldwide Interoperability for Microwave Access (WiMax), Bluetooth, Z-Wave, Near Field Communication (NFC) ZigBee, LiFi, and / or any low-power wide-area network (LPWAN) standards such as LoRa and Sigfox.

[0180] In some examples, the telecommunication network 1402 is a cellular network that implements 3GPP standardized features. Accordingly, the telecommunications network 1402 may support network slicing to provide different logical networks to different devices that are connected to the telecommunication network 1402. For example, the telecommunications network 1402 may provide Ultra Reliable Low Latency Communication (URLLC) services to some UEs, while providing Enhanced Mobile Broadband (eMBB) services to other UEs, and / or Massive Machine Type Communication (mMTC) / Massive IoT services to yet further UEs.

[0181] In some examples, the UEs 1412 are configured to transmit and / or receive information without direct human interaction. For instance, a UE may be designed to transmit information to the access network 1404 on a predetermined schedule, when triggered by an internal or external event, or in response to requests from the access network 1404. Additionally, a UE may be configured for operating in single- or multi-RAT or multi-standard mode. For example, a UE may operate with any one or combination of Wi-Fi, NR (New Radio) and LTE, i.e. being configured for multi-radio dual connectivity (MR-DC), such as E-UTRAN (Evolved-UMTS Terrestrial Radio Access Network) New Radio – Dual Connectivity (EN- DC).

[0182] In the example, the hub 1414 communicates with the access network 1404 to facilitate indirect communication between one or more UEs (e.g., UE 1412c and / or 1412d) and network nodes (e.g., network node 1410b). In some examples, the hub 1414 may be a controller, router, content source and analytics, or any of the other communication devices described herein regarding UEs. For example, the hub 1414 may be a broadband router enabling access to the core network 1406 for the UEs. As another example, the hub 1414 may be a controller that sends commands or instructions to one or more actuators in the UEs. Commands or instructions may be received from the UEs, network nodes 1410, or by executable code, script, process, or other instructions in the hub 1414. As another example, the hub 1414 may be a data collector that acts as temporary storage for UE data and, in some embodiments, may perform analysis or other processing of the data. As another example, the hub 1414 may be a content source. For example, for a UE that is a VR headset, display, loudspeaker or other media delivery device, the hub 1414 may retrieve VR assets, video, audio, or other media or data related to sensory information via a network node, which the hub 1414 then provides to the UE either directly, after performing local processing, and / or after adding additional local content. In still another example, the hub 1414 acts as a proxy server or orchestrator for the UEs, in particular if one or more of the UEs are low energy IoT devices.

[0183] The hub 1414 may have a constant / persistent or intermittent connection to the network node 1410b. The hub 1414 may also allow for a different communication scheme and / or schedule between the hub 1414 and UEs (e.g., UE 1412c and / or 1412d), and between the hub 1414 and the core network 1406. In other examples, the hub 1414 is connected to the core network 1406 and / or one or more UEs via a wired connection. Moreover, the hub 1414 may be configured to connect to an M2M service provider over the access network 1404 and / or to another UE over a direct connection. In some scenarios, UEs may establish a wireless connection with the network nodes 1410 while still connected via the hub 1414 via a wired or wireless connection. In some embodiments, the hub 1414 may be a dedicated hub – that is, a hub whose primary function is to route communications to / from the UEs from / to the network node 1410b. In other embodiments, the hub 1414 may be a non-dedicated hub – that is, a device which is capable of operating to route communications between the UEs and network node 1410b, but which is additionally capable of operating as a communication start and / or end point for certain data channels.

[0184] Figure 15 shows a UE 1500 in accordance with some embodiments. As used herein, a UE refers to a device capable, configured, arranged and / or operable to communicate wirelessly with network nodes and / or other UEs. Examples of a UE include, but are not limited to, a smart phone, mobile phone, cell phone, voice over IP (VoIP) phone, wireless local loop phone, desktop computer, personal digital assistant (PDA), wireless cameras, gaming console or device, music storage device, playback appliance, wearable terminal device, wireless endpoint, mobile station, tablet, laptop, laptop-embedded equipment (LEE), laptop-mounted equipment (LME), smart device, wireless customer-premise equipment (CPE), vehicle, vehicle-mounted or vehicle embedded / integrated wireless device, etc. Other examples include any UE identified by the 3rd Generation Partnership Project (3GPP), including a narrow band internet of things (NB-IoT) UE, a machine type communication (MTC) UE, and / or an enhanced MTC (eMTC) UE.

[0185] A UE may support device-to-device (D2D) communication, for example by implementing a 3GPP standard for sidelink communication, Dedicated Short-Range Communication (DSRC), vehicle-to-vehicle (V2V), vehicle-to-infrastructure (V2I), or vehicle-to-everything (V2X). In other examples, a UE may not necessarily have a user in the sense of a human user who owns and / or operates the relevant device. Instead, a UE may represent a device that is intended for sale to, or operation by, a human user but which may not, or which may not initially, be associated with a specific human user (e.g., a smart sprinkler controller). Alternatively, a UE may represent a device that is not intended for sale to, or operation by, an end user but which may be associated with or operated for the benefit of a user (e.g., a smart power meter).

[0186] The UE 1500 includes processing circuitry 1502 that is operatively coupled via a bus 1504 to an input / output interface 1506, a power source 1508, a memory 1510, a communication interface 1512, and / or any other component, or any combination thereof. Certain UEs may utilize all or a subset of the components shown in Figure 15. The level of integration between the components may vary from one UE to another UE. Further, certain UEs may contain multiple instances of a component, such as multiple processors, memories, transceivers, transmitters, receivers, etc.

[0004]

[0187] The processing circuitry 1502 is configured to process instructions and data and may be configured to implement any sequential state machine operative to execute instructions stored as machine-readable computer programs in the memory 1510. The processing circuitry 1502 may be implemented as one or more hardware-implemented state machines (e.g., in discrete logic, field-programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), etc.); programmable logic together with appropriate firmware; one or more stored computer programs, general-purpose processors, such as a microprocessor or digital signal processor (DSP), together with appropriate software; or any combination of the above. For example, the processing circuitry 1502 may include multiple central processing units (CPUs).

[0188] In the example, the input / output interface 1506 may be configured to provide an interface or interfaces to an input device, output device, or one or more input and / or output devices. Examples of an output device include a speaker, a sound card, a video card, a display, a monitor, a printer, an actuator, an emitter, a smartcard, another output device, or any combination thereof. An input device may allow a user to capture information into the UE 1500. Examples of an input device include a touch-sensitive or presence-sensitive display, a camera (e.g., a digital camera, a digital video camera, a web camera, etc.), a microphone, a sensor, a mouse, a trackball, a directional pad, a trackpad, a scroll wheel, a smartcard, and the like. The presence-sensitive display may include a capacitive or resistive touch sensor to sense input from a user. A sensor may be, for instance, an accelerometer, a gyroscope, a tilt sensor, a force sensor, a magnetometer, an optical sensor, a proximity sensor, a biometric sensor, etc., or any combination thereof. An output device may use the same type of interface port as an input device. For example, a Universal Serial Bus (USB) port may be used to provide an input device and an output device.

[0189] In some embodiments, the power source 1508 is structured as a battery or battery pack. Other types of power sources, such as an external power source (e.g., an electricity outlet), photovoltaic device, or power cell, may be used. The power source 1508 may further include power circuitry for delivering power from the power source 1508 itself, and / or an external power source, to the various parts of the UE 1500 via input circuitry or an interface such as an electrical power cable. Delivering power may be, for example, for charging of the power source 1508. Power circuitry may perform any formatting, converting, or other modification to the power from the power source 1508 to make the power suitable for the respective components of the UE 1500 to which power is supplied.

[0190] The memory 1510 may be or be configured to include memory such as random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic disks, optical disks, hard disks, removable cartridges, flash drives, and so forth. In one example, the memory 1510 includes one or more application programs 1514, such as an operating system, web browser application, a widget, gadget engine, or other application, and corresponding data 1516. The memory 1510 may store, for use by the UE 1500, any of a variety of various operating systems or combinations of operating systems.

[0191] The memory 1510 may be configured to include a number of physical drive units, such as redundant array of independent disks (RAID), flash memory, USB flash drive, external hard disk drive, thumb drive, pen drive, key drive, high-density digital versatile disc (HD- DVD) optical disc drive, internal hard disk drive, Blu-Ray optical disc drive, holographic digital data storage (HDDS) optical disc drive, external mini-dual in-line memory module (DIMM), synchronous dynamic random access memory (SDRAM), external micro-DIMM SDRAM, smartcard memory such as tamper resistant module in the form of a universal integrated circuit card (UICC) including one or more subscriber identity modules (SIMs), such as a USIM and / or ISIM, other memory, or any combination thereof. The UICC may for example be an embedded UICC (eUICC), integrated UICC (iUICC) or a removable UICC commonly known as ‘SIM card.’ The memory 1510 may allow the UE 1500 to access instructions, application programs and the like, stored on transitory or non-transitory memory media, to off-load data, or to upload data. An article of manufacture, such as one utilizing a communication system may be tangibly embodied as or in the memory 1510, which may be or comprise a device-readable storage medium.

[0192] The processing circuitry 1502 may be configured to communicate with an access network or other network using the communication interface 1512. The communication interface 1512 may comprise one or more communication subsystems and may include or be communicatively coupled to an antenna 1522. The communication interface 1512 may include one or more transceivers used to communicate, such as by communicating with one or more remote transceivers of another device capable of wireless communication (e.g., another UE or a network node in an access network). Each transceiver may include a transmitter 1518 and / or a receiver 1520 appropriate to provide network communications (e.g., optical, electrical, frequency allocations, and so forth). Moreover, the transmitter 1518 and receiver 1520 may be coupled to one or more antennas (e.g., antenna 1522) and may share circuit components, software or firmware, or alternatively be implemented separately.

[0193] In the illustrated embodiment, communication functions of the communication interface 1512 may include cellular communication, Wi-Fi communication, LPWAN communication, data communication, voice communication, multimedia communication, short-range communications such as Bluetooth, near-field communication, location-based communication such as the use of the global positioning system (GPS) to determine a location, another like communication function, or any combination thereof. Communications may be implemented in according to one or more communication protocols and / or standards, such as IEEE 802.11, Code Division Multiplexing Access (CDMA), Wideband Code Division Multiple Access (WCDMA), GSM, LTE, New Radio (NR), UMTS, WiMax, Ethernet, transmission control protocol / internet protocol (TCP / IP), synchronous optical networking (SONET), Asynchronous Transfer Mode (ATM), QUIC, Hypertext Transfer Protocol (HTTP), and so forth.

[0194] Regardless of the type of sensor, a UE may provide an output of data captured by its sensors, through its communication interface 1512, via a wireless connection to a network node. Data captured by sensors of a UE can be communicated through a wireless connection to a network node via another UE. The output may be periodic (e.g., once every 15 minutes if it reports the sensed temperature), random (e.g., to even out the load from reporting from several sensors), in response to a triggering event (e.g., when moisture is detected an alert is sent), in response to a request (e.g., a user initiated request), or a continuous stream (e.g., a live video feed of a patient).

[0195] As another example, a UE comprises an actuator, a motor, or a switch, related to a communication interface configured to receive wireless input from a network node via a wireless connection. In response to the received wireless input the states of the actuator, the motor, or the switch may change. For example, the UE may comprise a motor that adjusts the control surfaces or rotors of a drone in flight according to the received input or to a robotic arm performing a medical procedure according to the received input.

[0196] A UE, when in the form of an Internet of Things (IoT) device, may be a device for use in one or more application domains, these domains comprising, but not limited to, city wearable technology, extended industrial application and healthcare. Non-limiting examples of such an IoT device are a device which is or which is embedded in: a connected refrigerator or freezer, a TV, a connected lighting device, an electricity meter, a robot vacuum cleaner, a voice controlled smart speaker, a home security camera, a motion detector, a thermostat, a smoke detector, a door / window sensor, a flood / moisture sensor, an electrical door lock, a connected doorbell, an air conditioning system like a heat pump, an autonomous vehicle, a surveillance system, a weather monitoring device, a vehicle parking monitoring device, an electric vehicle charging station, a smart watch, a fitness tracker, a head-mounted display for Augmented Reality (AR) or Virtual Reality (VR), a wearable for tactile augmentation or sensory enhancement, a water sprinkler, an animal- or item-tracking device, a sensor for monitoring a plant or animal, an industrial robot, an Unmanned Aerial Vehicle (UAV), and any kind of medical device, like a heart rate monitor or a remote controlled surgical robot. A UE in the form of an IoT device comprises circuitry and / or software in dependence of the intended application of the IoT device in addition to other components as described in relation to the UE 1500 shown in Figure 15.

[0197] As yet another specific example, in an IoT scenario, a UE may represent a machine or other device that performs monitoring and / or measurements, and transmits the results of such monitoring and / or measurements to another UE and / or a network node. The UE may in this case be an M2M device, which may in a 3GPP context be referred to as an MTC device. As one particular example, the UE may implement the 3GPP NB-IoT standard. In other scenarios, a UE may represent a vehicle, such as a car, a bus, a truck, a ship and an airplane, or other equipment that is capable of monitoring and / or reporting on its operational status or other functions associated with its operation.

[0198] In practice, any number of UEs may be used together with respect to a single use case. For example, a first UE might be or be integrated in a drone and provide the drone’s speed information (obtained through a speed sensor) to a second UE that is a remote controller operating the drone. When the user makes changes from the remote controller, the first UE may adjust the throttle on the drone (e.g. by controlling an actuator) to increase or decrease the drone’s speed. The first and / or the second UE can also include more than one of the functionalities described above. For example, a UE might comprise the sensor and the actuator, and handle communication of data for both the speed sensor and the actuators.

[0199] Figure 16 shows a network node 1600 in accordance with some embodiments. As used herein, network node refers to equipment capable, configured, arranged and / or operable to communicate directly or indirectly with a UE and / or with other network nodes or equipment, in a telecommunication network. Examples of network nodes include, but are not limited to, access points (APs) (e.g., radio access points), base stations (BSs) (e.g., radio base stations, Node Bs, evolved Node Bs (eNBs) and NR NodeBs (gNBs)), O-RAN nodes or components of an O-RAN node (e.g., O-RU, O-DU, O-CU). In some embodiments, network node 1600 may be a mobile network node such that network node 1600 is equipped with some means for movement (e.g., integrated propellers, wheels, and / or the like) or may be incorporated into another mobile device, such as a drone or other UAV device.

[0200] Base stations may be categorized based on the amount of coverage they provide (or, stated differently, their transmit power level) and so, depending on the provided amount of coverage, may be referred to as femto base stations, pico base stations, micro base stations, or macro base stations. A base station may be a relay node or a relay donor node controlling a relay. A network node may also include one or more (or all) parts of a distributed radio base station such as centralized digital units, distributed units (e.g., in an O-RAN access node) and / or remote radio units (RRUs), sometimes referred to as Remote Radio Heads (RRHs). Such remote radio units may or may not be integrated with an antenna as an antenna integrated radio. Parts of a distributed radio base station may also be referred to as nodes in a distributed antenna system (DAS).

[0201] Other examples of network nodes include multiple transmission point (multi-TRP) 5G access nodes, multi-standard radio (MSR) equipment such as MSR BSs, network controllers such as radio network controllers (RNCs) or base station controllers (BSCs), base transceiver stations (BTSs), transmission points, transmission nodes, multi-cell / multicast coordination entities (MCEs), Operation and Maintenance (O&M) nodes, Operations Support System (OSS) nodes, Self-Organizing Network (SON) nodes, positioning nodes (e.g., Evolved Serving Mobile Location Centers (E-SMLCs)), and / or Minimization of Drive Tests (MDTs).

[0202] The network node 1600 includes a processing circuitry 1602, a memory 1604, a communication interface 1606, and a power source 1608. The network node 1600 may be composed of multiple physically separate components (e.g., a NodeB component and a RNC component, or a BTS component and a BSC component, etc.), which may each have their own respective components. In certain scenarios in which the network node 1600 comprises multiple separate components (e.g., BTS and BSC components), one or more of the separate components may be shared among several network nodes. For example, a single RNC may control multiple NodeBs. In such a scenario, each unique NodeB and RNC pair, may in some instances be considered a single separate network node. In some embodiments, the network node 1600 may be configured to support multiple radio access technologies (RATs). In such embodiments, some components may be duplicated (e.g., separate memory 1604 for different RATs) and some components may be reused (e.g., a same antenna 1610 may be shared by different RATs). The network node 1600 may also include multiple sets of the various illustrated components for different wireless technologies integrated into network node 1600, for example GSM, WCDMA, LTE, NR, WiFi, Zigbee, Z-wave, LoRaWAN, Radio Frequency Identification (RFID) or Bluetooth wireless technologies. These wireless technologies may be integrated into the same or different chip or set of chips and other components within network node 1600.

[0203] The processing circuitry 1602 may comprise a combination of one or more of a microprocessor, controller, microcontroller, central processing unit, digital signal processor, application-specific integrated circuit, field programmable gate array, or any other suitable computing device, resource, or combination of hardware, software and / or encoded logic operable to provide, either alone or in conjunction with other network node 1600 components, such as the memory 1604, to provide network node 1600 functionality.

[0204] In some embodiments, the processing circuitry 1602 includes a system on a chip (SOC). In some embodiments, the processing circuitry 1602 includes one or more of radio frequency (RF) transceiver circuitry 1612 and baseband processing circuitry 1614. In some embodiments, the radio frequency (RF) transceiver circuitry 1612 and the baseband processing circuitry 1614 may be on separate chips (or sets of chips), boards, or units, such as radio units and digital units. In alternative embodiments, part or all of RF transceiver circuitry 1612 and baseband processing circuitry 1614 may be on the same chip or set of chips, boards, or units.

[0005]

[0205] The memory 1604 may comprise any form of volatile or non-volatile computer-readable memory including, without limitation, persistent storage, solid-state memory, remotely mounted memory, magnetic media, optical media, random access memory (RAM), read-only memory (ROM), mass storage media (for example, a hard disk), removable storage media (for example, a flash drive, a Compact Disk (CD) or a Digital Video Disk (DVD)), and / or any other volatile or non-volatile, non-transitory device-readable and / or computer-executable memory devices that store information, data, and / or instructions that may be used by the processing circuitry 1602. The memory 1604 may store any suitable instructions, data, or information, including a computer program, software, an application including one or more of logic, rules, code, tables, and / or other instructions capable of being executed by the processing circuitry 1602 and utilized by the network node 1600. The memory 1604 may be used to store any calculations made by the processing circuitry 1602 and / or any data received via the communication interface 1606. In some embodiments, the processing circuitry 1602 and memory 1604 is integrated.

[0206] The communication interface 1606 is used in wired or wireless communication of signaling and / or data between a network node, access network, and / or UE. As illustrated, the communication interface 1606 comprises port(s) / terminal(s) 1616 to send and receive data, for example to and from a network over a wired connection. The communication interface 1606 also includes radio front-end circuitry 1618 that may be coupled to, or in certain embodiments a part of, the antenna 1610. Radio front-end circuitry 1618 comprises filters 1620 and amplifiers 1622. The radio front-end circuitry 1618 may be connected to an antenna 1610 and processing circuitry 1602. The radio front-end circuitry may be configured to condition signals communicated between antenna 1610 and processing circuitry 1602. The radio front-end circuitry 1618 may receive digital data that is to be sent out to other network nodes or UEs via a wireless connection. The radio front-end circuitry 1618 may convert the digital data into a radio signal having the appropriate channel and bandwidth parameters using a combination of filters 1620 and / or amplifiers 1622. The radio signal may then be transmitted via the antenna 1610. Similarly, when receiving data, the antenna 1610 may collect radio signals which are then converted into digital data by the radio front-end circuitry 1618. The digital data may be passed to the processing circuitry 1602. In other embodiments, the communication interface may comprise different components and / or different combinations of components.

[0207] In certain alternative embodiments, the network node 1600 does not include separate radio front-end circuitry 1618, instead, the processing circuitry 1602 includes radio front-end circuitry and is connected to the antenna 1610. Similarly, in some embodiments, all or some of the RF transceiver circuitry 1612 is part of the communication interface 1606. In still other embodiments, the communication interface 1606 includes one or more ports or terminals 1616, the radio front-end circuitry 1618, and the RF transceiver circuitry 1612, as part of a radio unit (not shown), and the communication interface 1606 communicates with the baseband processing circuitry 1614, which is part of a digital unit (not shown).

[0208] The antenna 1610 may include one or more antennas, or antenna arrays, configured to send and / or receive wireless signals. The antenna 1610 may be coupled to the radio front-end circuitry 1618 and may be any type of antenna capable of transmitting and receiving data and / or signals wirelessly. In certain embodiments, the antenna 1610 is separate from the network node 1600 and connectable to the network node 1600 through an interface or port.

[0209] The antenna 1610, communication interface 1606, and / or the processing circuitry 1602 may be configured to perform any receiving operations and / or certain obtaining operations described herein as being performed by the network node. Any information, data and / or signals may be received from a UE, another network node and / or any other network equipment. Similarly, the antenna 1610, the communication interface 1606, and / or the processing circuitry 1602 may be configured to perform any transmitting operations described herein as being performed by the network node. Any information, data and / or signals may be transmitted to a UE, another network node and / or any other network equipment.

[0210] The power source 1608 provides power to the various components of network node 1600 in a form suitable for the respective components (e.g., at a voltage and current level needed for each respective component). The power source 1608 may further comprise, or be coupled to, power management circuitry to supply the components of the network node 1600 with power for performing the functionality described herein. For example, the network node 1600 may be connectable to an external power source (e.g., the power grid, an electricity outlet) via an input circuitry or interface such as an electrical cable, whereby the external power source supplies power to power circuitry of the power source 1608. As a further example, the power source 1608 may comprise a source of power in the form of a battery or battery pack which is connected to, or integrated in, power circuitry. The battery may provide backup power should the external power source fail.

[0211] Embodiments of the network node 1600 may include additional components beyond those shown in Figure 16 for providing certain aspects of the network node’s functionality, including any of the functionality described herein and / or any functionality necessary to support the subject matter described herein. For example, the network node 1600 may include user interface equipment to allow input of information into the network node 1600 and to allow output of information from the network node 1600. This may allow a user to perform diagnostic, maintenance, repair, and other administrative functions for the network node 1600.

[0212] Figure 17 is a block diagram illustrating a virtualization environment 1700 in which functions implemented by some embodiments may be virtualized. In the present context, virtualizing means creating virtual versions of apparatuses or devices which may include virtualizing hardware platforms, storage devices and networking resources. As used herein, virtualization can be applied to any device described herein, or components thereof, and relates to an implementation in which at least a portion of the functionality is implemented as one or more virtual components. Some or all of the functions described herein may be implemented as virtual components executed by one or more virtual machines (VMs) implemented in one or more virtual environments 1700 hosted by one or more of hardware nodes, such as a hardware computing device that operates as a network node, UE, core network node, or host. Further, in embodiments in which the virtual node does not require radio connectivity (e.g., a core network node or host), then the node may be entirely virtualized. In some embodiments, the virtualization environment 1700 includes components defined by the O-RAN Alliance, such as an O-Cloud environment orchestrated by a Service Management and Orchestration Framework via an O-2 interface.

[0213] Applications 1702 (which may alternatively be called software instances, virtual appliances, network functions, virtual nodes, virtual network functions, etc.) are run in the virtualization environment Q400 to implement some of the features, functions, and / or benefits of some of the embodiments disclosed herein.

[0214] Hardware 1704 includes processing circuitry, memory that stores software and / or instructions executable by hardware processing circuitry, and / or other hardware devices as described herein, such as a network interface, input / output interface, and so forth. Software may be executed by the processing circuitry to instantiate one or more virtualization layers 1706 (also referred to as hypervisors or virtual machine monitors (VMMs)), provide VMs 1708a and 1708b (one or more of which may be generally referred to as VMs 1708), and / or perform any of the functions, features and / or benefits described in relation with some embodiments described herein. The virtualization layer 1706 may present a virtual operating platform that appears like networking hardware to the VMs 1708.

[0215] The VMs 1708 comprise virtual processing, virtual memory, virtual networking or interface and virtual storage, and may be run by a corresponding virtualization layer 1706. Different embodiments of the instance of a virtual appliance 1702 may be implemented on one or more of VMs 1708, and the implementations may be made in different ways. Virtualization of the hardware is in some contexts referred to as network function virtualization (NFV). NFV may be used to consolidate many network equipment types onto industry standard high volume server hardware, physical switches, and physical storage, which can be located in data centers, and customer premise equipment.

[0216] In the context of NFV, a VM 1708 may be a software implementation of a physical machine that runs programs as if they were executing on a physical, non-virtualized machine. Each of the VMs 1708, and that part of hardware 1704 that executes that VM, be it hardware dedicated to that VM and / or hardware shared by that VM with others of the VMs, forms separate virtual network elements. Still in the context of NFV, a virtual network function is responsible for handling specific network functions that run in one or more VMs 1708 on top of the hardware 1704 and corresponds to the application 1702.

[0217] Hardware 1704 may be implemented in a standalone network node with generic or specific components. Hardware 1704 may implement some functions via virtualization. Alternatively, hardware 1704 may be part of a larger cluster of hardware (e.g. such as in a data center or CPE) where many hardware nodes work together and are managed via management and orchestration 1710, which, among others, oversees lifecycle management of applications 1702. In some embodiments, hardware 1704 is coupled to one or more radio units that each include one or more transmitters and one or more receivers that may be coupled to one or more antennas. Radio units may communicate directly with other hardware nodes via one or more appropriate network interfaces and may be used in combination with the virtual components to provide a virtual node with radio capabilities, such as a radio access node or a base station. In some embodiments, some signaling can be provided with the use of a control system 1712 which may alternatively be used for communication between hardware nodes and radio units.

[0218] Although the computing devices described herein (e.g., UEs, network nodes, hosts) may include the illustrated combination of hardware components, other embodiments may comprise computing devices with different combinations of components. It is to be understood that these computing devices may comprise any suitable combination of hardware and / or software needed to perform the tasks, features, functions and methods disclosed herein. Determining, calculating, obtaining or similar operations described herein may be performed by processing circuitry, which may process information by, for example, converting the obtained information into other information, comparing the obtained information or converted information to information stored in the network node, and / or performing one or more operations based on the obtained information or converted information, and as a result of said processing making a determination. Moreover, while components are depicted as single boxes located within a larger box, or nested within multiple boxes, in practice, computing devices may comprise multiple different physical components that make up a single illustrated component, and functionality may be partitioned between separate components. For example, a communication interface may be configured to include any of the components described herein, and / or the functionality of the components may be partitioned between the processing circuitry and the communication interface. In another example, non-computationally intensive functions of any of such components may be implemented in software or firmware and computationally intensive functions may be implemented in hardware.

[0219] In certain embodiments, some or all of the functionality described herein may be provided by processing circuitry executing instructions stored on in memory, which in certain embodiments may be a computer program product in the form of a non-transitory computer- readable storage medium. In alternative embodiments, some or all of the functionality may be provided by the processing circuitry without executing instructions stored on a separate or discrete device-readable storage medium, such as in a hard-wired manner. In any of those particular embodiments, whether executing instructions stored on a non-transitory computer- readable storage medium or not, the processing circuitry can be configured to perform the described functionality. The benefits provided by such functionality are not limited to the processing circuitry alone or to other components of the computing device, but are enjoyed by the computing device as a whole, and / or by end users and a wireless network generally.

Claims

CLAIMS:

1. A method performed by a mobile network node (140, 1200, 1600) operating in a network, the method comprising: collecting (1301) user equipment, UE, parameter data from each of a plurality of UE devices (110, 1500) operating in the network; utilizing (1302) the collected UE parameter data to cluster the plurality of UE devices into one or more cluster groups, which are subsequently converted to a respective one or more tensor networks; determining (1303) future trajectory positions for each of the UE devices included in a cluster group assigned to the mobile network node using the one or more tensor networks; determining (1304) a most energy-efficient movement for the mobile network node to support the future trajectory positions of the plurality of UE devices; and establishing (1305) a new position for the mobile network node based on the most energy-efficient movement and an optimal data rate prediction achievable by the mobile network node.

2. The method of claim 1 wherein the UE parameter data includes one or more of: longitude and latitude location information, bandwidth data, height data, carrier frequency data, noise figure data, thermal noise data, and initial line of sight, LOS.

3. The method of any of claims 1 to 2 wherein determining the most energy efficient movement further comprises calculating an amount of the energy saved by the most energy efficient movement for the mobile network node.

4. The method of any of claims 1 to 3 further comprising adding or deleting at least one mobile network node to or from the network based on a data rate received from each of the UE devices included in the cluster group .

5. The method of any of claims 1 to 4 wherein the establishing a new position for the mobile network node is conducted in a single orthogonal frequency division multiplexing, OFDM, time slot.

6. The method of any of claims 1 to 5 wherein the mobile network node is configured to operate autonomously.

7. The method of any of claims 1 to 6 wherein each of the plurality of UEs is mobile.

8. The method of any of claims 1 to 7 wherein the mobile network node is configured to repeat the steps of the method in preferred time intervals.

9. The method of any of claims 1 to 8 further comprising using the tensor network to assign the mobile network node to a plurality of UEs based on a density exhibited by the UEs.

10. The method of any of claims 1 to 9 wherein tensor decomposition is used to determining the future trajectory positions. 11 The method of any of claims 1 to 10 wherein the mobile network node includes an unmanned aerial vehicle, UAV, base station or a mobile base station.

12. The method of any of claims 1 to 11 wherein establishing a new position for the mobile network node includes utilizing a reinforcement learning trained tensor network to determine the optimal data rate achievable by the mobile network node for the UE devices in the cluster group in accordance with the most energy efficient movement, and establishing a new position for the mobile network node based on the determined data rate.

13. The method of any of claims 1 to 12 wherein determining a most energy-efficient movement for the mobile network node includes determining an angle existing from the mobile network node to a future trajectory position for at least one of the UE devices.

14. A mobile network node (140, 1200, 1600) comprising: processing circuitry (1202, 1602); and at least one memory (1204, 1604) storing instructions executable by the processing circuitry to perform operations to: collect (1301) user equipment, UE, parameter data from each of a plurality of UE devices (110, 1500) operating in the network; utilize (1302) the collected UE parameter data to cluster the plurality of UE devices into one or more cluster groups, which are subsequently converted to a respective one or more tensor networks; determine (1303) future trajectory positions for each of the UE devices included in a cluster group assigned to the mobile network node using the one or more tensor networks;determine (1304) a most energy efficient movement for the mobile network node to support the future trajectory positions of the plurality of UE devices; and establish (1305) a new position for the mobile network node based on the most energy efficient movement and an optimal data rate prediction achievable by the mobile network node.

15. The mobile network node of claim 14 wherein the UE parameter data includes one or more of: longitude and latitude location information, bandwidth data, height data, carrier frequency data, noise figure data, thermal noise data, and initial line of sight, LOS.

16. The mobile network node of any of claims 14 to 15 wherein the mobile network node is further configured to calculate an amount of the energy saved by the most energy efficient movement for the mobile network node.

17. The mobile network node of any of claims 14 to 16 further configured to add or delete at least one mobile network node to or from the network based on a data rate received from each of the UE devices included in the cluster group.

18. The mobile network node of any of claims 14 to 17 wherein the establishing a new position for the mobile network node is conducted in a single orthogonal frequency division multiplexing, OFDM, time slot.

19. The mobile network node of any of claims 14 to 18 wherein the mobile network node is configured to operate autonomously.

20. The mobile network node of any of claims 14 to 19 wherein each of the plurality of UEs is mobile.

21. The mobile network node of any of claims 14 to 20 wherein the mobile network node is configured to repeat the operations in preferred time intervals.

22. The mobile network node of any of claims 14 to 21 wherein the mobile network node is configured to use the tensor network to assign the mobile network node to a plurality of UEs based on a density exhibited by the UEs.

23. The mobile network node of any of claims 14 to 22 wherein tensor decomposition is used to determine the future trajectory positions.

24. The mobile network node of any of claims 14 to 23 wherein the mobile network node includes an unmanned aerial vehicle, UAV, base station or a mobile base station.

25. The mobile network node of any of claims 14 to 24 wherein the mobile network node is further configured to utilize a reinforcement learning trained tensor network to determine the optimal data rate achievable by the mobile network node for the UE devices in the cluster group in accordance with the most energy efficient movement, and establish a new position for the mobile network node based on the determined data rate.

26. The mobile network node of any of claims 14 to 25 wherein the mobile network node is further configured to determine an angle existing from the mobile network node to a future trajectory position for at least one of the UE devices.

27. A non-transitory computer readable medium storing instructions executable by processing circuitry (1202, 1602) of a mobile network node (140, 1200, 1600), the instructions executed by the processing circuitry to perform operations comprising: collecting (1301) user equipment, UE, parameter data from each of a plurality of UE devices (110, 1500) operating in the network; utilizing (1302) the collected UE parameter data to cluster the plurality of UE devices into one or more cluster groups, which are subsequently converted to a respective one or more tensor networks; determining (1303) future trajectory positions for each of the UE devices included in a cluster group assigned to the mobile network node using the one or more tensor networks; determining (1304) a most energy-efficient movement for the mobile network node to support the future trajectory positions of the plurality of UE devices; and establishing (1305) a new position for the mobile network node based on the most energy-efficient movement and an optimal data rate prediction achievable by the mobile network node.

28. The non-transitory computer readable medium of claim 27, wherein further instructions are executed by the processing circuitry to perform further operations comprising operations of any one of claims 2 to 13.

Images