Planning over-the-air network synchronization topology with knowledge transfer

Abstract

A computer-implemented method implemented by a computing device is provided for selecting an over-the-air network synchronization topology in a planned target network. The method comprises: identifying (206) a first antenna reference point (ARP) pair in the source network; generating (208) a first representation of the first ARP pair; and generating (218) a second representation of a second ARP pair in the planned target network. The method further comprises using (220) a trained machine learning, ML, model to classify the respective pairs in the planned target network as at least one of similar and dissimilar, the trained ML model comprising transfer knowledge of the learned similarity; and selecting (222) a subset of the second ARP pair based on an identification of a minimum number of ARPs that are classified as similar that meet a criterion. The selected subset defines the over-the-air network synchronization topology.

Description

Technical Field

[0001] The present disclosure generally relates to a computer-implemented method, performed by a computing device, for selecting an over-the-air network synchronization topology in a planned target network, and related methods and apparatus. Background Art

[0002] Fifth-generation (5G) and sixth-generation (6G) wireless communication networks are envisioned to deliver enhanced end-user Quality of Experience (QoE) and unlock more time-critical and industrial applications. The evolving requirements for these use cases make time synchronization crucial. 5G and 6G wireless communication technologies rely on network synchronization and time alignment between antenna reference points (ARPs) in the communication network.

[0003] Traditionally, network synchronization implementations seek to minimize the relative time error (rTE) at the ARPs of all radio access network (RAN) nodes in a communication network with respect to a common reference time (CRT) by continuously adjusting the clock of each radio access network node to a local reference traceable to the CRT. The CRT may be, for example, the Global Positioning System (GPS) system time. At each radio access network node, the local reference may be provided by a global navigation satellite system (GNSS) receiver that receives the CRT. Alternatively, the CRT may be carried to the radio access network nodes over a backhaul network, for example, via a timing protocol such as the Precision Time Protocol (PTP). Thus, by adjusting the clock of each radio access network node to within a specific time error (TE) of the CRT, direct time alignment requirements (e.g., rTE) between ARPs may be achieved. 5G has attracted attention as a possible synchronization distribution approach, but challenges remain, such as the lack of effective synchronization for various scenarios. See, e.g., 5G synchronization requirements and solutions, https: / / www.ericsson.com / en / reports-and-papers / ericsson-technology-review / articles / 5g-synch ronization-requirements-and-solutions (accessed November 3, 2022). Summary of the Invention

[0004] Several challenges currently exist. As mentioned above, existing approaches may seek to achieve network synchronization at the ARPs of RAN nodes by minimizing rTE at the ARPs without using over-the-air / 5G channels. This approach may present challenges. For example, this trial-and-error approach may not be scalable. In dense networks, the number of possible cell pair candidates is large. Therefore, evaluating faulty relationships can be very difficult. Due to hardware limitations and network configuration, using a large number of cell connections may be impractical. Determining which relationships between ARP pairs of RAN nodes are reliable can be a labor-intensive process. Furthermore, in greenfield deployments, the network can be planned, and then the optimization of the network synchronization configuration can occur after the network is deployed. Therefore, pre-deployment optimization of network synchronization may be lacking. Furthermore, some existing approaches may not model the relationships between ARPs as multi-dimensional (e.g., technological, geospatial, etc.) entities and may also lack consideration of geospatial and demographic changes. As business demands increase, there may be a trend towards network densification. Therefore, the increasing complexity and diversity of 5G synchronization requirements and constraints may pose significant challenges to effectively finding topological configurations and robust network synchronization solutions (e.g., optimal solutions).

[0005] Certain aspects of the present disclosure and embodiments thereof may provide solutions to these and other challenges.

[0006] In some embodiments, a computer-implemented method implemented by a computing device is provided for selecting an air network synchronization topology in a planned target network. The method includes: identifying a plurality of first ARP pairs in a source network; generating a first representation of the plurality of first ARP pairs in the source network; and generating a second representation of a plurality of second ARP pairs in the planned target network. The method also includes: using a trained ML model to classify corresponding pairs from the second representation of the plurality of second ARP pairs in the planned target network as at least one of similar and dissimilar, the trained ML model including transfer knowledge of learned similarities between the plurality of first ARP pairs in the first representation. The method also includes: selecting a subset of the second ARP pairs based on identifying a minimum number of ARPs from the corresponding pairs that are classified as similar that meet a criterion. The selected subset defines the air network synchronization topology in the planned target network.

[0007] In some embodiments, a computing device is provided that is configured to select an air network synchronization topology in a planned target network. The computing device includes processing circuitry and a memory coupled to the processing circuitry. The memory includes instructions that, when executed by the processing circuitry, cause the computing device to perform operations. The operations include identifying a plurality of first ARP pairs in a source network; generating a first representation of the plurality of first ARP pairs in the source network; and generating a second representation of a plurality of second ARP pairs in the planned target network. The operations also include using a trained ML model to classify corresponding pairs from the second representation of the plurality of second ARP pairs in the planned target network as at least one of similar and dissimilar, the trained ML model including transferred knowledge of learned similarities between the plurality of first ARP pairs in the first representation. The operations also include selecting a subset of the second ARP pairs based on identifying a minimum number of ARPs from the corresponding pairs classified as similar that meet a criterion. The selected subset defines the air network synchronization topology in the planned target network.

[0008] In some embodiments, a computing device is provided that is configured to select an air network synchronization topology in a planned target network. The computing device is suitable for performing operations. The operations include: identifying a plurality of first ARP pairs in a source network; generating a first representation of the plurality of first ARP pairs in the source network; and generating a second representation of a plurality of second ARP pairs in the planned target network. The operations also include: using a trained ML model to classify corresponding pairs from the second representation of the plurality of second ARP pairs in the planned target network as at least one of similar and dissimilar, the trained ML model including the transferred knowledge of learned similarities between the plurality of first ARP pairs in the first representation. The operations also include: selecting a subset of the second ARP pairs based on identifying a minimum number of ARPs from the corresponding pairs classified as similar that meet a criterion. The selected subset defines the air network synchronization topology in the planned target network.

[0009] In some embodiments, a computer program is provided that includes program code to be executed by processing circuitry of a computing device configured to select an air network synchronization topology in a planned target network. Execution of the program code causes the computing device to perform operations. The operations include identifying a plurality of first ARP pairs in a source network; generating a first representation of the plurality of first ARP pairs in the source network; and generating a second representation of a plurality of second ARP pairs in the planned target network. The operations also include using a trained ML model to classify corresponding pairs from the second representation of the plurality of second ARP pairs in the planned target network as at least one of similar and dissimilar, the trained ML model including learned transfer knowledge of similarities between the plurality of first ARP pairs in the first representation. The operations also include selecting a subset of the second ARP pairs based on identifying a minimum number of ARPs from the corresponding pairs classified as similar that meet a criterion. The selected subset defines the air network synchronization topology in the planned target network.

[0010] In some embodiments, a computer program product is provided, comprising a non-transitory storage medium comprising program code to be executed by processing circuitry of a computing device. Execution of the program code causes the computing device to perform operations. The operations include identifying a plurality of first ARP pairs in a source network; generating a first representation of the plurality of first ARP pairs in the source network; and generating a second representation of a plurality of second ARP pairs in a planned target network. The operations also include using a trained ML model to classify corresponding pairs from the second representation of the plurality of second ARP pairs in the planned target network as at least one of similar and dissimilar, the trained ML model including transferred knowledge of learned similarities between the plurality of first ARP pairs in the first representation. The operations also include selecting a subset of the second ARP pairs based on identifying a minimum number of ARPs from the corresponding pairs classified as similar that meet a criterion. The selected subset defines the air network synchronization topology in the planned target network. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The accompanying drawings illustrate certain non-limiting embodiments of the inventive concept and are included to provide a further understanding of the present disclosure, and are incorporated into and constitute a part of this application. In the drawings:

[0012] Figure 1 is a flowchart illustrating the operation of a computing device according to some embodiments;

[0013] Figure 2is a flowchart illustrating the operation of a computing device according to some embodiments;

[0014] Figure 3 is a schematic diagram illustrating the mapping / representation of data of an example embodiment;

[0015] Figure 4 is a diagram illustrating an example of extracting data of a node ARP pair according to an example embodiment;

[0016] Figure 5 is a schematic diagram showing an example of a cycle;

[0017] Figure 6 is a flow chart illustrating the operation of an example embodiment for identifying unreliable measurements in a source network using looping;

[0018] Figure 7 is a diagram illustrating the operation of an example embodiment for generating embeddings for cell relations;

[0019] Figure 8 is a schematic diagram illustrating an example embodiment of training an ML model;

[0020] Figure 9 is a flowchart of operations for training an ML model according to some embodiments of the present disclosure;

[0021] Figure 10 is a schematic diagram illustrating an example embodiment of a selection result of an anchor node in the figure;

[0022] Figure 11 is a block diagram of a cloud environment in which some embodiments of the present disclosure may be implemented;

[0023] Figure 12 is a block diagram of a communication system according to some embodiments;

[0024] Figure 13 is a block diagram of a user equipment according to some embodiments

[0025] Figure 14 is a block diagram of a network node according to some embodiments;

[0026] Figure 15 is a block diagram of a host computer in communication with a user device according to some embodiments;

[0027] Figure 16 is a block diagram of a virtualization environment according to some embodiments;

[0028] Figure 17 An implementation example of a specific embodiment of the present disclosure is shown;

[0029] Figure 18illustrates an optional implementation example of a specific embodiment of the present disclosure; and

[0030] Figure 19 Three examples of computing devices that can be used to implement certain embodiments of the present disclosure are shown. DETAILED DESCRIPTION

[0031] The inventive concept will be described more fully below with reference to the accompanying drawings, which illustrate examples of embodiments of the inventive concept. However, the inventive concept can be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be comprehensive and complete and fully convey the scope of the inventive concept to those skilled in the art. It should also be noted that these embodiments are not mutually exclusive. Components in one embodiment may be assumed to be present / used in another embodiment.

[0032] The following description presents some embodiments of the disclosed subject matter. These embodiments are presented as teaching examples and should not be interpreted as limiting the scope of the disclosed subject matter. For example, certain details of the described embodiments may be modified, omitted, or expanded without departing from the scope of the described subject matter.

[0033] Mobile operators may often rely on a trial-and-error process for newly designed technologies (e.g., this may be common for new 5G capabilities). Mobile operators may prefer to conduct field trials in which the performance of new services, such as 5G synchronization, is monitored to mitigate network failures or avoid performance degradation. However, problems often arise due to a lack of knowledge about configuring new technologies in deployed networks. Furthermore, network planning may rely not only on network data and RAN configuration, but also consider established infrastructure and terrain factors.

[0034] Some approaches may have drawbacks. Some existing approaches may seek to achieve network synchronization at the ARPs of RAN nodes by minimizing rTE at the ARPs without using air / 5G channels. This approach can use trial and error, but may not be scalable. In dense networks, the number of possible cell pair candidates is large. Therefore, evaluating defect relationships may be very difficult. Due to hardware limitations and network configuration, it may not be practical to use a large number of cell connections. Determining which relationships between ARP pairs of RAN nodes are reliable can be a labor-intensive process. Furthermore, in greenfield deployments, the network may be planned and then seeking to optimize the network synchronization configuration may occur after the network is deployed. Therefore, there may be a lack of optimization of network synchronization before deployment.

[0035] Furthermore, some existing approaches may not model the relationships between ARPs as multi-dimensional (e.g., technological, geospatial, etc.) entities and may also lack consideration of geospatial and demographic changes. As business demands increase, there may be a trend toward network densification. Therefore, the increasing complexity and diversity of 5G synchronization requirements and constraints may pose significant challenges to effectively finding topology configurations and robust network synchronization solutions (e.g., optimal solutions). One of the challenges mobile operators may face is accommodating a large number of cells without significantly increasing operational and infrastructure costs.

[0036] Therefore, the increasing complexity and diversity of 5G synchronization requirements and constraints may pose significant challenges to effectively finding optimal topology configurations and robust network synchronization solutions.

[0037] As referred to herein, the term "relationship" refers to a relationship between RAN nodes that have antennas that can "hear" each other and therefore can perform over-the-air time alignment measurements and provide over-the-air synchronization to each other.

[0038] Certain aspects of the present disclosure and embodiments thereof may provide solutions to these or other challenges. A communication network comprises a set of interconnected nodes (including ARPs) located on a physical site. Each cell in a site provides coverage in a geographic area and is typically linked to adjacent cells to facilitate communication functions (e.g., over-the-air synchronization, handover, etc.). Each link between a cell and its adjacent cells may be referred to as a "relationship." A cell relationship comprises paired cells, each of which is characterized by a set of configuration parameters (azimuth, etc.) and a set of corresponding performance indicators (throughput, etc.). An observability system can be used to passively monitor system performance and signal failures.

[0039] When the primary reference for the RAN node's clock is disabled, a software (SW) feature may allow the use of the 5G channel between ARPs to maintain the RAN node's clock. While this SW feature may also enable features for synchronizing the ARPs over the 5G channel without the need for an additional local reference, this may require the 5G channel between ARPs to meet the necessary characteristics.

[0040] In order to effectively utilize 5G's capabilities to distribute synchronization, it may be necessary to know during the planning stages of a new 5G rollout which ARP pairs in the network will have a good enough 5G link between them for synchronization. Knowing this knowledge in advance can allow for smarter and more cost-effective synchronization solutions, where only a few nodes (referred to herein as "anchor nodes") require a local time reference (e.g., a GNSS receiver), with the remaining nodes synchronized via the 5G channel.

[0041] Therefore, for an operational network, the 5G link quality between ARP pairs in the RAN can be obtained by using a feature that measures the time alignment between ARPs. Transferring this knowledge from an operational network to a newly planned network can improve the shortcomings of existing methods.

[0042] In an example, operations are provided for planning a network synchronization topology for a planned target network by transferring knowledge about ARP-ARP channel quality from an existing source network.

[0043] Neural networks can be used to model and generate deep representations of each ARP pair in the source network of the operation, and then similarity learning is used to transfer knowledge from the source network of the operation to the planned target network.

[0044] In certain operations, a process is provided to find reliable ARP pairs created in a source network of an operation and use such pairs to transfer knowledge from the source network to a planned destination network.

[0045] These operations may provide a process for finding a reliable ARP pair in a planned target network.

[0046] These operations may also provide techniques for selecting an optimal set of anchor nodes in the planned target network.

[0047] Technical advantages provided by certain embodiments of the present disclosure may include: based on the method including modeling and transferring synchronization settings from a deployed source network to selected ARP pairs in a planned target network, network synchronization planning can be optimized or improved. For example, the method may allow the use of 5G channels between ARPs to maintain the clocks of RAN nodes when the primary reference of the RAN node's clock is disabled based on the use of machine learning (ML) techniques, which may result in the efficient management and automation of the process of planning new synchronization relationships in the planned target network deployment. In addition, based on the inclusion of knowledge transfer, the method may be scalable and can be used for any target mobile network.

[0048] Furthermore, the method may reduce or minimize the labor used to evaluate synchronization relationships by alleviating the tasks and expense of troubleshooting activities typically handled by testers and experienced network engineers.

[0049] For ease of discussion, example embodiments are described herein in the non-limiting context of a 5G source network and a 5G planned target network, each including a gNodeB (gNB) ARP pair. However, the present disclosure is not limited thereto, and some embodiments include other radio access technologies (e.g., 6G) and nodes other than gNBs (e.g., evolved NodeBs (eNBs)).

[0050] As used herein, the term "network node" refers to, but is not limited to, a base station, a gNodeB (gNB), and an evolved NodeB (eNB), etc. The terms "network node" and "node" in this document are interchangeable.

[0051] Provides operations for planning node (e.g., gNB) ARP pairs for the target network. These operations may include the following:

[0052] 1. Data Collection (referred to as “Operation 1” in this article): Operation 1 can consider two data sources:

[0053] - Live network data, such as site configuration data, and

[0054] - Urban infrastructure data, such as spatial data (eg terrain type) of the area where the network is deployed.

[0055] 2. Extract the synchronization node ARP from the deployed source network (referred to as "Operation 2" herein).

[0056] 3. Identify robust and reliable node ARP pairs (referred to herein as "Operation 3").

[0057] 4. Generate a representation for a reliable node ARP pair (referred to herein as "operation 4").

[0058] 5. Train the ML model to learn the similarity between node ARP pairs. For example, a Siamese neural network can be used to learn similarity and overcome the problem of limited data (referred to as "Operation 5" in this article).

[0059] 6. Generate ARP pairs of nodes in the target network of interest. For example, a trained Siamese neural network can be used to estimate the reliability of setting new relationships (referred to herein as "operation 6").

[0060] 7. Identify the anchor node in the graph of node ARP pairs (referred to herein as "operation 7").

[0061] Figure 1 1 is a flowchart 100 illustrating operations 1-7 performed by a computing device. In the following discussion, the above seven operations are referred to as operations 1-7. For some embodiments of computing devices and related methods, Figure 1 The operations in the flowchart of may be optional. For example, the operations of blocks 102-108, blocks 114-122, and block 128 may be optional.

[0062] The operation of the computing device can be performed by Figure 5 、 Figure 17 、 Figure 18 or Figure 19According to some embodiments of the present disclosure, the operation of the computing device (using Figure 14 、 Figure 17 、 Figure 18 or Figure 19 For example, the modules can be stored in Figure 14 、 Figure 17 、 Figure 18 or Figure 19 At least one memory 14304, 17102, 18102, 19518, 19548 of the present invention, and these modules can provide instructions so that when the instructions of the module are Figure 14 、 Figure 17 、 Figure 18 or Figure 19 When executed by the corresponding computing device processing circuits 14302, 17101, 18101, 19512, 19542, the computing device 502, 506, 510, 17100, or 18000 implements the corresponding operations of the flowchart. In some embodiments, the computing device includes one of the following: a centralized computing device communicatively connected to the source network and the planned target network, and a cloud-based distributed computing device, the cloud-based distributed computing device including one or more of the following modules (discussed further herein): (i) a data collection module, (ii) an over-the-air measurement evaluation module, and (iii) a cross-network knowledge transfer module.

[0063] For some embodiments of computing devices and related methods, Figure 2 Various operations in the flowchart may be optional. For example, Figure 2 The operations of blocks 200-204, blocks 210-216, and block 224 may be optional.

[0064] refer to Figure 1 , perform the operations of blocks 102 - 114 on the deployed source network, and perform the operations of blocks 116 - 130 on the planned target network.

[0065] The data collection of operation 1 may include the operations of blocks 102 - 106 for the source network and the operations of blocks 116 - 120 for the planned target network.

[0066] For the source network, SW characteristics ( Figure 1An over-the-air time alignment measurement software (SW) feature (shown as an over-the-air time alignment measurement software) is deployed 102 in an operational source network. The SW feature can, for example, enable the use of telecommunication channels (e.g., 5G channels) between deployed ARPs to maintain a node's clock when its primary reference is disabled. The SW feature is used to obtain 104 node ARP pair data, including configuration management (CM) and performance management (PM) data.

[0067] For the planned target network, data is obtained from the network plan of the target network 116, 118. The data may include candidate node ARP pair data (eg, configuration management (CM) data, such as cell configuration).

[0068] For the source network, in block 106, node ARP pair data is accessed from the data obtained in block 104. For the planned target network, candidate node pair data is accessed 118 from the data obtained in block 116.

[0069] In block 106 (for the source network) and block 120 (for the target network), ARP pair data is obtained. The ARP pair data includes two data streams: network data and spatial infrastructure data.

[0070] exist Figure 1 In the example of FIG, the network data for the source network includes information about cell configuration attributes and performance management (PM) data, such as, for each node ARP pair, antenna position, direction, transmit power, and over-the-air measurements (e.g., signal quality and signal path loss; time alignment error measurements from the deployed source network). For the planned target network, the network data may include at least one or more of the following data: cell configuration; antenna position; expected signal quality and expected signal path loss.

[0071] Spatial infrastructure data includes data extracted from the spatial multi-source data geographic attributes of the corresponding tiles of the source network and the planned target network 108 (for the source network), 122 (for the target network). Spatial data can include but is not limited to building layers, points of interest (POI), terrain types, etc. Currently, some methods rely solely on telecommunication data. However, environmental conditions can play a role in how the signal propagates and how the synchronization function works. In contrast, in Figure 1 In the example, the following data is collected: building footprints to capture the shape and size of buildings; and POIs (e.g., bus stops, airports, etc.) and land use information (e.g., commercial, educational, etc.) to understand traffic patterns and locations (e.g., cities) that change over time.

[0072] Figure 3is a schematic diagram illustrating a mapping / representation of data 300 collected from node ARP pairs 306, 308 in first and second cells 310, 312, respectively, and / or infrastructure 302, 304, for which the data is collected from a source network or from a planned target network.

[0073] The extraction of operation 2 may include the operation of block 108 for the source network. The extraction may be performed, for example, to understand and mark the behavior of the time alignment of the SW features of block 102 in an attempt to deploy reliable relationships in the planned target network. To achieve this, Figure 1 In the example above, data is processed from a controlled environment or source network where SW features have been deployed.

[0074] Figure 4 is a schematic diagram showing an example of extracted data for a node ARP pair 306, 308. In this example, node / ARP 400 is not included in the node ARP pair. For the node ARP pair 306, 308 of the first cell 310. Attribute 402 is extracted for node / ARP 306, and attribute 406 is extracted for node / ARP pair 308. In addition, time series measurements 404 are extracted for node / ARP pairs 306, 308. The extracted node ARP pair data can be stored in a managed object (MO) category. When the SW feature is activated, the SW feature can capture time series data, referred to herein as time alignment error measurements. PM data can be used to measure SW feature performance. PM data can be captured periodically. The time alignment error PM counter can be a scalar. Table 1 below shows an example of node ARP pair attribute data, and Table 2 below shows an example of node ARP PM data. Table 3 below shows another example of a node APP PM data list.

[0075] Table 1:

[0076] Node ARP pair configuration data Data collection time Global cell ID for each pair Radio technology generation (e.g. 4G, 5G, 6G, etc.) The downlink channel bandwidth in the cells used for each pair. The uplink channel bandwidth in the cells used for each pair. The azimuth used for each pair. The distance between pairs etc.

[0077] Table 2:

[0078]

[0079]

[0080] Table 3:

[0081] Node ARP to PM data Data collection time gNB ARP global cell ID Time alignment error counter

[0082] Operation 3 of identifying an ARP pair that produces a reliable time alignment measurement may include Figure 1 Operation of block 110. Once the time alignment error data is collected, the historical time series measurements can be labeled using the following process.

[0083] A check may first be performed to see if the feature has been activated and if the cell is operational. Attributes used for this may include availability status, operational status, and measurement quality. For example, these attributes may indicate whether a measurement was captured because it was unreliable or due to other operational settings. At this stage, some measurements may be marked as unreliable.

[0084] The time alignment error data may be in the form of a time series of data. Each data point may represent a time alignment error measurement of the node's ARP.

[0085] While this example includes existing measurement quality attributes that proactively assess the reliability of a measurement, some measurements assessed as reliable may still be unreliable due to various reasons, such as quality or external factors (e.g., antenna orientation, alignment issues, environmental factors, etc.). As discussed below, such situations can be identified by exploiting "cycles" in the network.

[0086] The topology of the source network can be used to obtain cycles in the source network. Figure 5 is a diagram illustrating an example of a loop of size 3 (e.g., a triangle illustrated by the three edges between a first node / ARP 502 in cell A 500 and a second node / ARP 506 in cell B 504, between a third node / ARP 510 in cell C 508 and the second node / ARP 506, and between the third node / ARP 510 and the first node / ARP 502).

[0087] In an ideal scenario, the sum of the time differences can be equal to or close to 0 in each cycle. For example, Figure 5 In the example, given three cells A 500, B 504, and C 508 forming a loop (here a triangle), the sum of the time differences in an ideal scenario can be close to or equal to 0, i.e., Δ (A,B) +Δ (B,C) +Δ (c,A) ≈0. Then, a threshold of the total time difference of a loop can be close enough to 0 and set as the loop quality threshold (or, for example, the triangle quality threshold in the case of a triangle). If a loop does not meet this condition, it can be marked as defective and the measurement of its edges is considered unreliable. Otherwise, if a loop meets this condition, it can be marked as good and the measurement of its edges is considered reliable. If an edge exists in a defective loop as well as a good loop, its measurement can be considered reliable. Although reference (discussed below) Figure 5 and Figure 6The example in describes this channel for identifying additional unreliable measurements, but the disclosure is not limited to cycles of size 3 (eg, triangles) and may include cycles of other sizes with more than two sides.

[0088] Figure 6 is a flow chart illustrating operations for identifying unreliable measurements in a source network using a loop, shown in this example as corresponding to Figure 5 Three cycles.

[0089] In block 600 , an air measurement topology is obtained for a source network; in block 602 , an air measurement topology map is generated according to the obtained air measurement topology.

[0090] In block 604, triangles are identified in the air measurement topology map. For each triangle T(A, B, C), in block 606, the time difference is calculated to obtain the sum of the time differences: Δ τ =Δ (A,B) +Δ (B,C) +Δ (C,A) .

[0091] In block 608, the cycle is marked as defective or good. For example, if Δ τ ≥ triangle quality threshold: mark T(A,B,C) as defective; or if Δ τ <Triangle quality threshold: mark T(A,B,C) as good.

[0092] In block 610 , common edges between the defective triangle and the good triangles are checked; in block 612 , the edges of the defective triangle are marked as unreliable, and the remaining edges (eg, in the good triangles) are marked as reliable.

[0093] Operation 4 of generating a reliable node ARP pair representation may include Figure 1 Operation 112. Figure 7 is a diagram showing the process for generating embeddings for each cell relationship. Once reliable node ARP pairs are identified from operation 3, a set of different types of features can be used to generate latent representations. Using such high-dimensional feature representations can mitigate domain differences. In addition, as Figure 7 As shown in the example of

[15] , leveraging multi-source data and utilizing transfer learning techniques can overcome the label scarcity problem.

[0094] To generate embeddings for node ARP pairs, Figure 7 In the example (which corresponds to Figure 5 and Figure 6), using the two sources of data discussed in Operation 1. That is: (1) node ARP pair configurations, (2) node ARP pair key performance indicators (KPIs), and (3) spatial context data. Figure 7 The operations for generating embeddings using data are shown. Figure 7 As shown in FIG, the node ARP pair configuration data 202a, 202b collected for the node ARP pairs 502, 506 and the spatial context data 7000 including the cell location related attributes 204 are provided to the multi-layer perceptrons (MLPs) 700, 704, 706 and the gated recurrent units (GRUs) 702, 708. These data can be collected and prepared in the form of tabular data. The MLPs 700, 704, 706 can be used to generate deep potential representations of tabular features. On the other hand, the node ARP pair KPIs are time data points. Therefore, the GRUs 702, 708 can create a time representation of the internal time counter. The embeddings generated in the previous steps are connected using the potential representation vector to represent 710 the node ARP relationship (e.g., ARP pair).

[0095] Training an ML model for operation 5 to learn similarities between pairs of node APRs may include Figure 1 Operation 114. Figure 8 is a schematic diagram showing an example of training an ML model. In the example shown, in order to compare between two node ARP pairs 502, 506 and 506, 510 using potential representation vectors, a Siamese neural network (NN) architecture is used that can learn to distinguish between two inputs 202b, 204, 202b and 214a, 216, 214b. The Siamese NN of this example includes two parallel identical NN components 802, 804 that share weights 806 and a function for calculating the similarity of the outputs of the two NNs 802, 804. The embedding generated in the example of the previous step contains differentiable properties. The embedding is then used as a proxy to calculate the similarity (e.g., similarity score 808) between the node ARP pairs 502, 506 and 506, 510. If the two node ARP pair embeddings share the same label, they are similar, and if they do not share the same label, the pair is considered to be dissimilar.

[0096] The weights 806 of the Siamese NN are shared between the two parts of the network 802, 804, which can enable the network to learn similarities and dissimilarities. The ML model can be trained using contrastive loss.

[0097] Figure 8The Siamese NN uses training data. In this example, the Siamese NN is trained by teaching the following: two reliable pairs are similar; two unreliable pairs are dissimilar; and a reliable pair and an unreliable pair are dissimilar. Therefore, the training data for the ARP node pair can be generated as follows:

[0098] For each node ARP pair (X i ), extract k node ARP pairs (X n ). That is, for example, if (X i ) is reliable, then (X n ) is reliable, and if (X i ) is unreliable, then (X n ) is unreliable.

[0099] For each node ARP pair (X i ), do not extract k node ARP pairs from the same category (e.g., dissimilar) (X r ). That is, for example, if (X i ) is reliable, then (X r ) is unreliable, and if (X i ) is unreliable,

[0100] Then (X r )reliable.

[0101] The advantage of using Siamese NN is that it can overcome the data shortage problem. Fine-grained node ARP-level knowledge transfer can be applied to any location (e.g., a city) regardless of its size. Location-independent contextual features and KPIs existing in areas where SW features have not yet been deployed can be used.

[0102] Figure 9 It is used for training and Figure 8 Flowchart of the operation of the corresponding ML model. In the loop of operation 900 and operation 902, for each ARP relationship (e.g., ARP pair) in the source network, the air synchronization relationship is obtained (operation 900). In block 902, the air synchronization relationship latent representation vector X is extracted. S The operation then continues, in block 904, according to X S Candidate similar and dissimilar pairs are generated. The candidate pairs generated from operation 904 are then used to train a Siamese NN in operation 906.

[0103] For the target network, in the loop of operations 908 and 910, for each air synchronization relationship in the target network, an air synchronization relationship is obtained (operation 908). In block 910, the air synchronization relationship potential representation vector X is extracted. tThe operation then continues, in block 912, according to X t Generate pairs for each relation, according to X s Generate K random candidates from each category (e.g., reliable and unreliable categories). Then use the trained Siamese NN in operation 914 to calculate the similarity of each pair from operation 912. The calculation can be implemented as discussed further herein. In operation 916, a majority vote is applied and the category with the highest similarity score is selected. Thus, in the example, from source X s Select K random candidates from each category (e.g., reliable and unreliable) in X t The right and from X s and apply majority voting to determine the pair from X t Is the pair more similar to a reliable pair or an unreliable pair.

[0104] Generating an ARP pair of nodes in the target network for operation 6 may include Figure 1 Operation 124 in . Continuing with the example discussed above, once the Siamese NN model is trained, the trained model is used to classify whether two node ARP pairs are similar or dissimilar.

[0105] In this example, for this purpose, all possible node ARP pairs in the target network are generated. Based on the generated ARP pairs, a representation embedding is generated for each node ARP pair (as discussed previously herein).

[0106] For each node ARP pair, k random node ARP pairs are used from each category (i.e., similar to reliable pairs, not similar to reliable pairs). Then, the trained Siamese NN model is used to evaluate the similarity. If most of the similarity scores indicate that the node ARP pair is a reliable pair (e.g., if the pair is mostly similar to a reliable pair), it is marked (e.g., as a "possible reliable pair" and vice versa). If most of the similarity scores indicate that the node ARP pair is not a reliable pair (e.g., if the pair is mostly similar to an unreliable pair), it is marked accordingly (e.g., as a "possible unreliable pair").

[0107] Identifying the anchor node in the graph of node ARP pairs for operation 7 may include Figure 1 Continuing with the above example, the node ARP pairs that have been classified as reliable pairs are used to construct a graph of node ARP pairs (eg, where pairs are edges in the graph).

[0108] In some cases, a cell may be restricted to a maximum number of connections. In this case, for each cell, pairs related to that cell, labeled as, for example, "likely reliable pairs," are ranked from high to low based on the average of their similarity scores with the reliable pairs obtained during previous operations using the Siamese NN. The top N pairs are then retained, where N is the maximum number of connections allowed. These selected 128-node ARP pairs are then used to construct 130 a synchronization topology map for the target network.

[0109] Continuing with the example, once the graph is constructed, the minimum number of nodes to be selected is identified in order to reach all other nodes in the graph, where the selected nodes reach nodes up to k hops away (e.g., k=1 means that the node only reaches its direct neighbors). In this example, the minimum number of nodes is selected as follows.

[0110] Given a graph constructed using the selected node-ARP pairs, construct an adjacency matrix from the graph, where a vertex is connected to other vertices that are k hops away (including itself), and calculate the degree of each vertex. Create a state dictionary containing a binary state for each vertex, reflecting whether the vertex has been reached. Calculate the conditional degree of each vertex, which captures the number of vertices it is connected to that have not yet been reached. Select the vertex with the highest conditional degree and add it to the minimum node set. Then change the state of this vertex and all the vertices it is connected to to reflect that these vertices have been reached. Then calculate the conditional degree again, and continue this process until all vertices have been reached.

[0111] Note that a node can be reached multiple times, and such nodes are referred to herein as "overlapping nodes." Overlapping nodes are encouraged to minimize the number of nodes required to reach all nodes. Therefore, when selecting the node with the highest conditional degree, in the event of a draw, the vertex with the highest degree is chosen because it allows for the largest number of "overlapping nodes." If a draw still exists, several alternatives can be proposed.

[0112] In this example, once a minimum number of nodes are identified, these nodes are marked as anchor nodes.As used herein, an "anchor node" is a node that should have traceability capabilities to a Primary Reference Time Clock (PRTC).

[0113] Figure 10 is a schematic diagram showing an example of the selection result of the anchor node in the figure, where the anchor node can reach the node at a distance of 1 hop. Figure 10 As shown in , 4 anchor nodes (node numbers 2, 5, 6, and 10, in the 11-node graph) are selected. Figure 10 The selected anchor nodes 2, 5, 6, and 10 can reach nodes that are at most 1 hop away.

[0114] Some embodiments relate to methods implemented by a computing device (e.g., Figure 5 、 Figure 17 、 Figure 18 or Figure 19 502, 506, 510, 17100, 18000 in FIG. ). Reference will now be made to FIG. 5 according to some embodiments. Figure 2 The flowchart discusses the operation of the computing device.

[0115] like Figure 2 As shown in , a computer-implemented method implemented by a computing device is provided for selecting an air network synchronization topology in a planned target network. The method includes: identifying (206) a plurality of first ARP pairs in a source network; generating (208) a first representation of the plurality of first ARP pairs in the source network; and generating (218) a second representation of a plurality of second ARP pairs in the planned target network. The method also includes: using (220) a trained ML model to classify corresponding pairs from the second representation of the plurality of second ARP pairs in the planned target network as at least one of similar and dissimilar, the trained ML model including transferred knowledge of learned similarities between the plurality of first ARP pairs in the first representation. The method also includes: selecting (222) a subset of the second ARP pairs based on identifying a minimum number of ARPs from the corresponding pairs classified as similar that meet a criterion. The selected subset defines the air network synchronization topology in the planned target network.

[0116] The knowledge may include antenna data and radio environment data from the source network.

[0117] In some embodiments, identifying (206) the plurality of first ARP pairs comprises: (i) marking a plurality of historical time series measurements as at least one of reliable and unreliable, the plurality of historical time series measurements comprising time alignment error data for pairs of the plurality of first ARP pairs, (ii) identifying and marking unreliable and reliable ARP pairs, and (iii) using the marked ARP pairs to select pairs of the second ARP pairs.

[0118] In some embodiments, identifying (206) further comprises: (i) identifying a plurality of cycles in the first representation of the plurality of first ARP pairs in the source network, the cycles comprising time differences between corresponding ARPs in a group of three or more ARPs, (ii) calculating a sum of the time differences for each cycle in the source network, (iii) comparing the sum of the time differences for each cycle to a specified cycle threshold, (iv) marking the cycle as acceptable and marking a plurality of edges in the cycle as reliable when the cycle satisfies the cycle threshold, (v) marking the cycle as unacceptable and marking a plurality of edges in the cycle as unreliable when the cycle fails to satisfy the cycle threshold, and (vi) considering the edge as reliable if the edge exists in both the unacceptable cycle and the acceptable cycle.

[0119] In some embodiments, generating (208) a first representation of a plurality of first ARP pairs in a source network comprises: (i) accessing data for an ARP pair in the first ARP pairs, the data comprising cell configuration, synchronization key performance indicators (KPIs) comprising air time alignment measurements, and spatial context data, (ii) generating embeddings using the accessed data, and (iii) connecting the embeddings for the representations using a latent representation vector.

[0120] In another embodiment, the method further comprises: training (210) an ML model to learn to distinguish corresponding pairs of the plurality of first ARP pairs in the first representation, the corresponding pairs to be compared to verify similarity of corresponding second pairs in the planned target network.

[0121] In some embodiments, selecting (222) a subset of the second ARP pairs includes: (i) generating a plurality of first ARP pairs in the source network that are identified as similar or dissimilar, (ii) training an ML model on the identified first ARP pairs based on the generated plurality of first ARP pairs identified as similar or dissimilar, (iii) using the trained ML model to generate a plurality of second ARP pairs in the planned target network and comparing the generated plurality of second ARP pairs with the first plurality of first ARP pairs in the source network that are identified as similar or dissimilar, and (iv) marking corresponding pairs of the plurality of second ARP pairs in the planned target network as at least one of candidate reliable pairs and candidate unreliable pairs.

[0122] ARPs may be included in the nodes, and selecting (222) a subset of a second ARP pair that defines an over-the-air synchronization topology in the planned target network may include: (i) constructing a graph of candidate reliable pairs, (ii) identifying a minimum number of nodes that reach all other nodes in the graph, and (iii) marking the minimum number of nodes as the selected subset of the second ARP pair.

[0123] In some embodiments, identifying (206) a plurality of first ARP pairs in the source network is based on the method further comprising: accessing (200) data from the source network, including configuration management and performance data; and extracting (202) data from the accessed data, including at least one of the following for each of the plurality of first ARP pairs: cell configuration, location of the antenna in the ARP, signal quality, signal path loss, and time alignment error measurement.

[0124] In another embodiment, the plurality of second ARP pairs in the planned target network are identified based on the method further comprising: accessing (212) data for the planned target network, including configuration management and performance data; and extracting data (214) from the accessed data, including at least one of the following for each of the plurality of second ARP pairs: cell configuration, location of the antenna in the ARP, expected signal quality, and expected signal path loss.

[0125] In some embodiments, generating (220) a second representation of a plurality of second ARP pairs in the planned target network that are classified as similar is based on the method further comprising: extracting (204, 216) spatial infrastructure data for a geographic area of the planned target network, including at least one of: building footprints, points of interest, land use information, and terrain type.

[0126] In another embodiment, identifying (206) the plurality of first ARP pairs in the source network is identified from a plurality of historical time series measurements including time alignment error data for pairs of the plurality of first ARP pairs in the source network.

[0127] In some embodiments, the trained ML model includes at least one neural network that learns to distinguish corresponding pairs of a plurality of first ARP pairs in the first representation and then compares them to verify similarity of corresponding second pairs in the planned target network.

[0128] In some embodiments, the at least one neural network comprises a Siamese neural network comprising two identical deep neural networks, and training the ML model to learn similarities between corresponding pairs of the plurality of ARP pairs in the first representation comprises: (i) comparing corresponding latent representation vectors of two ARP pairs from the plurality of first ARP pairs from the first representation to obtain a similarity score, (ii) based on the comparison, the similarity score comprises a score identifying the two ARP pairs as similar when the two ARP pairs share the same label, and (iii) based on the comparison, the similarity score comprises a score identifying the two ARP pairs as dissimilar when the two ARP pairs have different labels.

[0129] In some embodiments, the criteria include: a minimum number of ARPs from the corresponding pair that are classified as similar reach the remaining number of ARPs from the corresponding pair within a specified hop distance.

[0130] In some embodiments, the method further comprises identifying (224) an anchor node in the planned target network from the subset of the second ARP pairs.

[0131] Figure 11 11 is a block diagram of a cloud environment in which some embodiments of the present disclosure may be implemented. For example, ML models and other components for implementing some embodiments may be deployed in a cloud environment, including, but not limited to, decoupling of components for three groups of operations: (1) data collection (e.g., at site data collection 1106); (2) network synchronization feature evaluation (e.g., at over-the-air measurement evaluation component 1120); and (3) learning similarity and cross-network knowledge transfer (e.g., in target network 1136).

[0132] like Figure 11 As shown in the example of , a live customer network 1102 (e.g., a mobile network) and external data 1104 (e.g., spatial data such as city infrastructure data) are communicatively connected to a site data collection 1106. In this example, the site data collection 1106 includes a data agent 1108 that receives data input 1114 (from the live customer network 1102) that is decrypted by a decryption component 1110. The data agent 1108 outputs the decrypted data to a parser component 1112, which provides the parsed data to a database 1118. The external data 1104 is provided to a parser 1116, which also provides the parsed data to a database 1118.

[0133] Data from the database 1118 of the site data collection 1106 is provided to the identification 1124 component in the air measurement evaluation component 1120. In this example, the identification component 1124 identifies robust and reliable node ARP pairs. The air measurement evaluation component also includes an extraction component 1122. The extraction component 1122 extracts synchronized node ARP data from the source network.

[0134] In this example, the over-the-air measurement evaluation component 1120 also includes a cross knowledge transfer component 1126. The cross knowledge transfer component 1126 includes components for: generating 1128 representations for reliable node ARP pairs; training 1130 an ML model to learn similarities between node ARP pairs; generating 1132 node ARP pairs in a target network; and synchronized network topology generation 1134. The generated synchronized network topology is provided to the target network 1136.

[0135] In this example, the target network 1136 includes an activated over-the-air synchronization feature 1138 (which receives the generated synchronization network topology, such as Figure 11 ); and a component 1140 that enables determining (eg, predicting) reliable pairs for a target network.

[0136] Figure 12 An example of a communication system 1200 is shown in accordance with some embodiments.

[0137] In this example, a communication system 1200 includes a telecommunications network 1202, which includes an access network 1204 (such as a radio access network (RAN)) and a core network 1206, which includes one or more core network nodes 1208. The access network 1204 includes one or more access network nodes, such as network nodes 1210a and 1210b (one or more of which may be generally referred to as network nodes 1210), or any other similar third generation partnership project (3GPP) access nodes or non-3GPP access points. The network nodes 1210 facilitate direct or indirect connection of user equipment (UE), for example, by connecting UEs 1212a, 1212b, 1212c, and 1212d (one or more of which may be generally referred to as UE 1212) to the core network 1206 via one or more wireless connections.

[0138] Example wireless communications via wireless connections include sending 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. Furthermore, in various embodiments, the communication system 1200 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 a wired or wireless connection. The communication system 1200 may include and / or interface with any type of communication, telecommunication, data, cellular, radio network, and / or other similar types of systems.

[0139] The UE 1212 may be any of a variety of communication devices, including wireless devices that are arranged, configured, and / or operable to communicate wirelessly with the network node 1210 and other communication devices. Similarly, the network node 1210 is arranged, capable, configured, and / or operable to communicate directly or indirectly with the UE 1212 and / or with other network nodes or devices in the telecommunications network 1202 to enable and / or provide network access, such as wireless network access, and / or perform other functions, such as management in the telecommunications network 1202.

[0140] In the depicted example, core network 1206 connects network node 1210 to one or more hosts, such as host 1216. These connections can be direct or indirect via one or more intermediate networks or devices. In other examples, the network node can be directly coupled to the host. Core network 1206 includes one or more core network nodes (e.g., core network node 1208) constructed using hardware and software components. The features of these components can be substantially similar to those described with respect to the UE, network node, and / or host, so that the description generally applies to the corresponding components of core network node 1208. Example core network nodes include functionality of one or more of the following: a mobile switching center (MSC), a mobility management entity (MME), a home subscriber server (HSS), an access and mobility management function (AMF), a session management function (SMF), an authentication server function (AUSF), a subscription identifier dehiding function (SIDF), a unified data management (UDM), a security edge protection proxy (SEPP), a network exposure function (NEF), and / or a user plane function (UPF).

[0141] The host 1216 may be owned or controlled by a service provider that is different from the operator or provider of the access network 1204 and / or the telecommunications network 1202, and may be operated by or on behalf of the service provider. The host 1216 may host various applications to provide one or more services. Examples of such applications include live and pre-recorded audio / video content, data collection services such as retrieving and compiling data about various environmental conditions detected by multiple UEs, analytical functions, social media, functions for controlling or otherwise interacting with remote devices, functions for alarm and monitoring centers, or any other such functions implemented by a server.

[0142] Overall, Figure 12 The communication system 1200 enables connectivity between UEs, network nodes, and hosts. In this sense, the communication system can be configured to operate according to predefined rules or procedures, such as specific standards, including but 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 next-generation standards (e.g., 6G); Wireless Local Area Network (WLAN) standards such as the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standard (WiFi); and / or any other suitable wireless communication standards, such as 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.

[0143] In some examples, telecommunication network 1202 is a cellular network that implements 3GPP standardized features. Thus, telecommunication network 1202 can support network slicing to provide different logical networks to different devices connected to telecommunication network 1202. For example, telecommunication network 1202 can provide ultra-reliable low-latency communication (URLLC) services to some UEs, enhanced mobile broadband (eMBB) services to other UEs, and / or massive machine type communication (mMTC) / massive IoT services to yet other UEs.

[0144] In some examples, UE 1212 is configured to send and / or receive information without direct human interaction. For example, the UE can be designed to send information to access network 1204 according to a predetermined schedule when triggered by an internal or external event, or in response to a request from access network 1204. In addition, the UE can be configured to operate in a single RAT or multi-RAT or multi-standard mode. For example, the UE can operate using any one or a combination of Wi-Fi, NR (New Radio), and LTE, i.e., be configured for multi-radio dual connectivity (MR-DC), such as E-UTRAN (Evolved UMTS Terrestrial Radio Access Network) New Radio Dual Connectivity (EN-DC).

[0145] In this example, hub 1214 communicates with access network 1204 to facilitate indirect communication between one or more UEs (e.g., UE 1212c and / or 1212d) and a network node (e.g., network node 1210b). In some examples, hub 1214 can be a controller, router, content source and analysis, or any other communication device described herein with respect to a UE. For example, hub 1214 can be a broadband router that enables a UE to access core network 1206. As another example, hub 1214 can be a controller that sends commands or instructions to one or more actuators in a UE. The commands or instructions can be received from a UE, network node 1210, or received through executable code, scripts, processes, or other instructions in hub 1214. As another example, hub 1214 can be a data collector that acts as a temporary storage for UE data, and in some embodiments, can perform analysis or other processing of the data. As another example, hub 1214 can be a content source. For example, for a UE that is a VR headset, display, speaker, or other media delivery device, the hub 1214 can retrieve VR assets, video, audio, or other media or data related to sensory information via a network node, and then the hub 1214 provides it directly to the UE, provides it to the UE after performing local processing, and / or provides it to the UE after adding additional local content. In another example, the hub 1214 acts as a proxy server or orchestrator for the UE, especially when one or more UEs are low-energy IoT devices.

[0146] The hub 1214 may have a constant / persistent or intermittent connection to the network node 1210b. The hub 1214 may also allow different communication schemes and / or scheduling between the hub 1214 and UEs (e.g., UEs 1212c and / or 1212d) and between the hub 1214 and the core network 1206. In other examples, the hub 1214 is connected to the core network 1206 and / or one or more UEs via a wired connection. In addition, the hub 1214 may be configured to connect to an M2M service provider via the access network 1204 and / or to another UE via a direct connection. In some scenarios, the UE may establish a wireless connection with the network node 1210 while still being connected via the hub 1214 via a wired or wireless connection. In some embodiments, the hub 1214 may be a dedicated hub, i.e., a hub whose primary function is to route communications from the network node 1210b to the UE / from the UE to the network node 1210a. In other embodiments, the hub 1214 may be a non-dedicated hub, ie, a device operable to route communications between UEs and the network node 1210b, but also operable as a communications origin and / or destination for certain data channels.

[0147] Figure 13 UE 13200 according to some embodiments is shown. As used herein, UE refers to a device capable of, configured, arranged and / or operable to wirelessly communicate with a network node and / or other UEs. Examples of UE include, but are not limited to, smartphones, mobile phones, cellular phones, voice over IP (VoIP) phones, wireless local loop phones, desktop computers, personal digital assistants (PDAs), wireless cameras, game consoles or devices, music storage devices, playback devices, wearable terminal devices, wireless endpoints, mobile stations, tablet computers, notebook computers, notebook embedded devices (LEEs), notebook mounted devices (LMEs), smart devices, wireless customer premises equipment (CPE), vehicle-mounted or vehicle-embedded / integrated wireless devices, etc. Other examples include any UE identified by the Third Generation Partnership Project (3GPP), including narrowband Internet of Things (NB-IoT) UEs, machine type communication (MTC) UEs, and / or enhanced MTC (eMTC) UEs.

[0148] A UE may support device-to-device (D2D) communications, for example, by implementing 3GPP standards for sidelink communications, dedicated short-range communications (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 owning and / or operating the associated device. Rather, a UE may represent a device that is intended to be sold to or operated by a human user, but which may not be associated with a specific human user or may not initially be associated (e.g., a smart sprinkler controller). Alternatively, a UE may represent a device that is not intended to be sold to or operated by an end user, but which may be associated with or operated for the benefit of a user (e.g., a smart meter).

[0149] UE 13200 includes a processing circuit 13202 operatively coupled to an input / output interface 13206, a power supply 13208, a memory 13210, a communication interface 13212, and / or any other components, or any combination thereof, via a bus 13204. Some UEs may utilize Figure 13 All or a subset of the components shown in . The level of integration between components may vary from UE to UE. In addition, some UEs may contain multiple instances of a component, such as multiple processors, memories, transceivers, transmitters, receivers, etc.

[0150] Processing circuitry 13202 is configured to process instructions and data and may be configured to implement any sequential state machine operable to execute instructions stored as a machine-readable computer program in memory 13210. Processing circuitry 13202 may be implemented as one or more hardware-implemented state machines (e.g., in discrete logic, a field programmable gate array (FPGA), an application-specific integrated circuit (ASIC), etc.); programmable logic and appropriate firmware; one or more stored computer programs, general-purpose processors such as microprocessors or digital signal processors (DSPs), and appropriate software; or any combination thereof. For example, processing circuitry 13202 may include multiple central processing units (CPUs).

[0151] In this example, the input / output interface 13206 can be configured to provide one or more interfaces to an input device, an output device, or one or more input and / or output devices. Examples of output devices include: a speaker, a sound card, a video card, a display, a monitor, a printer, an actuator, a transmitter, a smart card, another output device, or any combination thereof. An input device can allow a user to capture information into the UE 13200. Examples of input devices include: a touch-sensitive or presence-sensitive display, a camera (e.g., a digital camera, a digital video camera, a webcam, etc.), a microphone, a sensor, a mouse, a trackball, a directional pad, a trackpad, a scroll wheel, a smart card, etc. A presence-sensitive display can include a capacitive or resistive touch sensor to sense input from the user. The sensor can be, for example, 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. The output device can use the same type of interface port as the input device. For example, a Universal Serial Bus (USB) port can be used to provide both input and output devices.

[0152] In some embodiments, the power supply 13208 is configured as a battery or battery pack. Other types of power sources may be used, such as an external power source (e.g., a power outlet), a photovoltaic device, or a battery. The power supply 13208 may also include power circuitry for delivering power from the power supply 13208 itself and / or an external power source to the various components of the UE 13200 via an input circuit or an interface such as a power cable. For example, the delivered power may be used to charge the power supply 13208. The power circuitry may perform any formatting, conversion, or other modifications to the power from the power supply 13208 to make the power suitable for the respective components of the UE 13200 being powered.

[0153] The memory 13210 may be or be configured to include a memory such as a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic disk, an optical disk, a hard disk, a removable cassette tape, a flash drive, etc. In one example, the memory 13210 includes one or more application programs 13214, such as an operating system, a web browser application, a widget, a widget engine, or other applications, and corresponding data 13216. The memory 13210 may store any one of a variety of operating systems or a combination of operating systems for use by the UE 13200.

[0154] The memory 13210 may be configured to include multiple physical drive units, such as a redundant array of independent disks (RAID), flash memory, a USB flash drive, an external hard drive, a thumb drive, a pen drive, a key drive, a high-density digital versatile disk (HD-DVD) optical drive, an internal hard drive, a Blu-ray disc drive, a holographic digital data storage (HDDS) optical drive, an external mini dual in-line memory module (DIMM), synchronous dynamic random access memory (SDRAM), an external micro DIMM SDRAM, a smart card memory, such as a 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 an ISIM, other memory, or any combination thereof. The UICC may be, for example, an embedded UICC (eUICC), an integrated UICC (iUICC), or a removable UICC commonly referred to as a "SIM card." The memory 13210 may allow the UE 13200 to access instructions, applications, and the like stored on a transient or non-transient storage medium to offload or upload data. An article of manufacture, such as an article of manufacture utilizing a communication system, may be tangibly embodied as or in memory 13210 , which may be or may include a device-readable storage medium.

[0155] The processing circuit 13202 can be configured to communicate with an access network or other network using a communication interface 13212. The communication interface 13212 may include one or more communication subsystems and may include an antenna 13222 or be communicatively coupled to an antenna 13222. The communication interface 13212 may include one or more transceivers used for communication, such as by communicating with one or more remote transceivers of another device capable of wireless communication (e.g., a network node in an access network or another UE). Each transceiver may include a transmitter 13218 and / or a receiver 13220 suitable for providing network communication (e.g., optical, electrical, frequency allocation, etc.). In addition, the transmitter 13218 and the receiver 13220 may be coupled to one or more antennas (e.g., antenna 13222) and may share circuit components, software, or firmware, or alternatively be implemented separately.

[0156] In the illustrated embodiment, the communication functionality of the communication interface 13212 may include cellular communication, Wi-Fi communication, LPWAN communication, data communication, voice communication, multimedia communication, short-range communication (such as Bluetooth), near-field communication, location-based communication (such as using a global positioning system (GPS) to determine location), another similar communication functionality, or any combination thereof. Communication may be implemented according to one or more communication protocols and / or standards, such as IEEE 802.11, code division multiple 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 network (SONET), asynchronous transfer mode (ATM), QUIC, hypertext transfer protocol (HTTP), etc.

[0157] Regardless of the type of sensor, the UE can provide an output of the data captured by its sensor via a wireless connection to a network node through its communication interface 13212. The data captured by the UE's sensor can be transmitted to the network node via another UE via a wireless connection. The output can be periodic (e.g., once every 15 minutes if it reports the sensed temperature), random (e.g., to balance the load of reports from several sensors), in response to a trigger event (e.g., when moisture is detected, an alarm 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).

[0158] As another example, a UE may include an actuator, motor, or switch associated with a communication interface configured to receive wireless input from a network node via a wireless connection. In response to the received wireless input, the state of the actuator, motor, or switch may change. For example, the UE may include a motor that adjusts the control surfaces or rotors of a drone in flight based on the received input, or adjusts a robotic arm performing a medical procedure based on the received input.

[0159] When the UE is in the form of an Internet of Things (IoT) device, it can be a device used in one or more application areas including, but not limited to, urban wearable technology, extended industrial applications, and healthcare. Non-limiting examples of such IoT devices are devices that are or are embedded in connected refrigerators or freezers, televisions, connected lighting devices, electricity meters, robotic vacuum cleaners, voice-controlled smart speakers, home security cameras, motion detectors, thermostats, smoke detectors, door / window sensors, flood / water vapor sensors, electric door locks, connected doorbells, air conditioning systems such as heat pumps, autonomous vehicles, surveillance systems, weather monitoring devices, parking monitoring devices, electric vehicle charging stations, smart watches, fitness trackers, head-mounted displays for augmented reality (AR) or virtual reality (VR), wearable devices for tactile enhancement or sensory enhancement, sprinklers, animal or item tracking devices, sensors for monitoring plants or animals, industrial robots, unmanned aerial vehicles (UAVs), and any type of medical device, such as a heart rate monitor or a remotely controlled surgical robot. In addition to incorporating such Figure 13 In addition to the other components described for the UE 13200 shown in FIG, a UE in the form of an IoT device may further include circuitry and / or software depending on the intended application of the IoT device.

[0160] As another specific example, in an IoT scenario, a UE can represent a machine or other device that performs monitoring and / or measurement, and transmits the results of such monitoring and / or measurement to another UE and / or a network node. In this case, the UE can be an M2M device, which in the 3GPP context can be referred to as an MTC device. As a specific example, a UE can implement the 3GPP NB-IoT standard. In other scenarios, a UE can represent a vehicle, such as a car, bus, truck, ship, or airplane, or other device capable of monitoring and / or reporting its operating status or other functions associated with its operation.

[0161] In practice, any number of UEs can be used together for a single use case. For example, the first UE can be a drone or be integrated into a drone and provide the drone's speed information (obtained by a speed sensor) to a second UE that is a remote control for operating the drone. When the user implements changes from the remote control, the first UE can adjust the throttle on the drone (e.g., by controlling an actuator) to increase or decrease the speed of the drone. The first UE and / or the second UE can also include more than one of the above functions. For example, the UE can include a sensor and an actuator and handle the communication of data for both the speed sensor and the actuator.

[0162] Figure 14 A network node 14300 according to some embodiments is shown. As used herein, a network node refers to a device capable of, configured, arranged, and / or operable to communicate directly or indirectly with a UE and / or other network nodes or devices in a telecommunications 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 Node Bs (gNBs)).

[0163] Base stations can be categorized based on the amount of coverage they provide (or, in other words, their transmit power level) and thus may be referred to as femto base stations, pico base stations, micro base stations, or macro base stations, depending on the amount of coverage provided. A base station may be a relay node or a relay donor node that controls a relay. A network node may also include one or more (or all) components of a distributed radio base station, such as a centralized digital unit and / or a remote radio unit (RRU), sometimes referred to as a remote radio head (RRH). Such a remote radio unit may or may not be integrated with an antenna as an antenna-integrated radio. Components of a distributed radio base station may also be referred to as nodes in a distributed antenna system (DAS).

[0164] Other examples of network nodes include: multi-transmission point (multi-TRP) 5G access node, multi-standard radio (MSR) equipment (such as MSR BS), network controller (such as radio network controller (RNC) or base station controller (BSC)), base transceiver station (BTS), transmission point, transmission node, multi-cell / multicast coordination entity (MCE), operation and maintenance (O&M) node, operation support system (OSS) node, self-organizing network (SON) node, positioning node (e.g., evolved serving mobile positioning center (E-SMLC)) and / or minimization of drive tests (MDT).

[0165] Network node 14300 includes processing circuitry 14302, memory 14304, a communication interface 14306, and a power supply 14308. Network node 14300 may be composed of multiple physically separate components (e.g., a NodeB component and an RNC component, or a BTS component and a BSC component, etc.), each of which may have its own corresponding components. In certain scenarios where network node 14300 includes multiple separate components (e.g., BTS and BSC components), one or more separate components may be shared across multiple network nodes. For example, a single RNC may control multiple NodeBs. In such scenarios, in certain instances, each unique NodeB and RNC pair may be considered a single, separate network node. In some embodiments, network node 14300 may be configured to support multiple radio access technologies (RATs). In such embodiments, some components may be replicated (e.g., separate memory 14304 for different RATs) and some components may be reused (e.g., the same antenna 14310 may be shared by different RATs). The network node 14300 may also include various sets of components shown for different wireless technologies, such as GSM, WCDMA, LTE, NR, WiFi, Zigbee, Z-wave, LoRaWAN, radio frequency identification (RFID), or Bluetooth wireless technologies, integrated into the network node 14300. These wireless technologies may be integrated into the same or different chips or chipsets and other components within the network node 14300.

[0166] The processing circuit 14302 may include one or more combinations of: a microprocessor, a controller, a microcontroller, a central processing unit, a digital signal processor, an application specific integrated circuit, a field programmable gate array, or any other suitable computing device, resource, or combination of hardware, software and / or encoded logic, which is used to provide network node 14300 functionality alone or in combination with other network node 14300 components (such as memory 14304).

[0167] In some embodiments, processing circuitry 14302 comprises a system on a chip (SOC). In some embodiments, processing circuitry 14302 comprises one or more of radio frequency (RF) transceiver circuitry 14312 and baseband processing circuitry 14314. In some embodiments, radio frequency (RF) transceiver circuitry 14312 and baseband processing circuitry 14314 may be on separate chips (or chipsets), boards, or units, such as a radio unit and a digital unit. In alternative embodiments, some or all of RF transceiver circuitry 14312 and baseband processing circuitry 14314 may be on the same chip, chipset, board, or unit.

[0168] Memory 14304 may include any form of volatile or non-volatile computer-readable memory, including, but not limited to, persistent memory, solid-state memory, remotely mounted memory, magnetic media, optical media, random access memory (RAM), read-only memory (ROM), mass storage media (e.g., a hard disk), removable storage media (e.g., 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 storage device that stores instructions, data, and / or information that can be used by processing circuit 14302. Memory 14304 may store any suitable instructions, data, or information, including computer programs, software, applications (including one or more of logic, rules, code, tables, and / or other instructions that can be executed by processing circuit 14302 and utilized by network node 14300). Memory 14304 may be used to store any calculations performed by processing circuit 14302 and / or any data received via communication interface 14306. In some embodiments, processing circuitry 14302 and memory 14304 are integrated.

[0169] Communication interface 14306 is used for wired or wireless communication of signaling and / or data between network nodes, access networks and / or UEs. As shown, communication interface 14306 includes port / terminal 14316 for, for example, sending data to and receiving data from the network via a wired connection. Communication interface 14306 also includes radio front-end circuitry 14318, which can be coupled to antenna 14310, or in some embodiments, can be part of antenna 14310. Radio front-end circuitry 14318 includes filter 14320 and amplifier 14322. Radio front-end circuitry 14318 can be connected to antenna 14310 and processing circuitry 14302. Radio front-end circuitry can be configured to condition signals transmitted between antenna 14310 and processing circuitry 14302. Radio front-end circuitry 14318 can receive digital data to be sent to other network nodes or UEs via a wireless connection. The radio front-end circuit 14318 can use a combination of filters 14320 and / or amplifiers 14322 to convert the digital data into a radio signal with appropriate channel and bandwidth parameters. The radio signal can then be transmitted via the antenna 14310. Similarly, when receiving data, the antenna 14310 can collect the radio signal, which is then converted into digital data by the radio front-end circuit 14318. The digital data can be passed to the processing circuit 14302. In other embodiments, the communication interface may include different components and / or different combinations of components.

[0170] In certain alternative embodiments, the network node 14300 does not include a separate radio front-end circuit 14318; instead, the processing circuit 14302 includes the radio front-end circuit and is connected to the antenna 14310. Similarly, in some embodiments, all or some of the RF transceiver circuit 14312 is part of the communication interface 14306. In other embodiments, the communication interface 14306 includes one or more ports or terminals 14316, the radio front-end circuit 14318, and the RF transceiver circuit 14312 as part of a radio unit (not shown), and the communication interface 14306 communicates with the baseband processing circuit 14314, which is part of a digital unit (not shown).

[0171] Antenna 14310 may include one or more antennas or antenna arrays configured to transmit and / or receive wireless signals. Antenna 14310 may be coupled to radio front-end circuitry 14318 and may be any type of antenna capable of wirelessly transmitting and receiving data and / or signals. In some embodiments, antenna 14310 is separate from network node 14300 and may be connected to network node 14300 via an interface or port.

[0172] Antenna 14310, communication interface 14306, and / or processing circuit 14302 may be configured to implement any receiving operations and / or certain obtaining operations performed by a network node as described herein. Any information, data, and / or signals may be received from a UE, another network node, and / or any other network device. Similarly, antenna 14310, communication interface 14306, and / or processing circuit 14302 may be configured to implement any transmitting operations performed by a network node as described herein. Any information, data, and / or signals may be transmitted to a UE, another network node, and / or any other network device.

[0173] The power supply 14308 provides power to the various components of the network node 14300 in a form suitable for the respective components (e.g., at the voltage and current levels required by each respective component). The power supply 14308 may also include or be coupled to power management circuitry to provide power to the components of the network node 14300 for implementing the functionality described herein. For example, the network node 14300 may be connected to an external power source (e.g., an electrical grid, a power outlet) via an input circuit or an interface such as a cable, whereby the external power source provides power to the power circuitry of the power supply 14308. As another example, the power supply 14308 may include a power source in the form of a battery or battery pack connected to or integrated into the power circuitry. The battery may provide backup power if the external power source fails.

[0174] An embodiment of the network node 14300 may include Figure 14Additional components beyond those shown may be used to provide certain aspects of the functionality of the network node, including any functionality described herein and / or any functionality required to support the subject matter described herein. For example, the network node 14300 may include a user interface device to allow information to be input into the network node 14300 and to allow information to be output from the network node 14300. This may allow a user to perform diagnostic, maintenance, repair, and other management functions on the network node 14300.

[0175] Figure 15 is a block diagram of a host 15400 according to various aspects described herein, which may be Figure 12 As used herein, host 15400 may be or may include various hardware and / or software combinations, including standalone servers, blade servers, cloud-enabled servers, distributed servers, virtual machines, containers, or processing resources in a server farm. Host 15400 may provide one or more services to one or more UEs.

[0176] Host 15400 includes processing circuitry 15402 operatively coupled to input / output interface 15406, network interface 15408, power supply 15410, and memory 15412 via bus 15404. Other components may be included in other embodiments. The features of these components may be similar to those described with respect to previous figures (such as Figure 13 and Figure 14 ) devices are substantially similar so that their descriptions are generally applicable to corresponding components of the host 15400.

[0177] Memory 15412 may include one or more computer programs, including one or more host applications 15414 and data 15416, which may include user data, such as data generated by a UE for host 15400 or data generated by host 15400 for a UE. Embodiments of host 15400 may utilize only a subset or all of the components shown. Host application 15414 may be implemented in a container-based architecture and may provide support for video codecs (e.g., Versatile Video Coding (VVC), High Efficiency Video Coding (HEVC), Advanced Video Coding (AVC), MPEG, VP9) and audio codecs (e.g., FLAC, Advanced Audio Coding (AAC), MPEG, G.711), including transcoding for multiple different classes, types, or implementations of UEs (e.g., mobile phones, desktop computers, wearable display systems, head-up display systems). Host application 15414 may also provide user authentication and permission checks and may periodically report health, routing, and content availability to a central node (such as a device in the core network or at the edge). Thus, the host 15400 can select and / or instruct different hosts for over-the-top services to the UE. The host application 15414 can support various protocols, such as HTTP Live Streaming (HLS), Real-Time Messaging Protocol (RTMP), Real-Time Streaming Protocol (RTSP), Dynamic Adaptive Streaming over HTTP (MPEG-DASH), etc.

[0178] Figure 16 is a block diagram illustrating a virtualization environment 16500 in which functionality implemented by some embodiments may be virtualized. In the present context, virtualization means creating a virtual version of a device or apparatus, which may include virtualized hardware platforms, storage devices, and network resources. As used herein, virtualization may be applied to any device described herein or components thereof, and relates to an implementation in which at least a portion of functionality is implemented as one or more virtual components. Some or all of the functionality described herein may be implemented as virtual components performed by one or more virtual machines (VMs) implemented in one or more virtual environments 16500 hosted by one or more hardware nodes (such as hardware computing devices operating as network nodes, UEs, core network nodes, or hosts). In addition, in embodiments where a virtual node does not require a radio connection (e.g., a core network node or host), the node may be fully virtualized.

[0179] Application 16502 (which may alternatively be referred to as a software instance, a virtual appliance, a network function, a virtual node, a virtual network function, etc.) runs in a virtualized environment Q400 to implement some features, functions and / or benefits of some embodiments disclosed herein.

[0180] The hardware 16504 includes processing circuitry, memory storing instructions and / or software executable by the hardware processing circuitry, and / or other hardware devices as described herein, such as network interfaces, input / output interfaces, and the like. The software can be executed by the processing circuitry to instantiate one or more virtualization layers 16506 (also referred to as hypervisors or virtual machine monitors (VMMs)), provide VMs 16508a and 16508b (one or more of which may be generally referred to as VMs 16508), and / or implement any functionality, features, and / or benefits associated with some embodiments described herein. The virtualization layer 16506 can present a virtual operating platform that appears to be network hardware to the VMs 16508.

[0181] VM 16508 includes virtual processing, virtual memory, virtual networks or interfaces, and virtual storage, and can be run by a corresponding virtualization layer 16506. Different embodiments of instances of virtual devices 16502 can be implemented on one or more of VMs 16508 and can be implemented in different ways. Hardware virtualization is sometimes referred to as network function virtualization (NFV). NFV can be used to consolidate many network device types onto industry-standard high-volume server hardware, physical switches, and physical storage, which can be located in data centers and customer premises equipment.

[0182] In the context of NFV, VMs 16508 can be software implementations of physical machines that run programs as if they were executed on a physical, non-virtualized machine. Each VM 16508 and the portion of hardware 16504 on which it executes, whether dedicated to that VM and / or shared with other VMs, form a separate virtual network element. Still in the context of NFV, a virtual network function is responsible for handling specific network functions running in one or more VMs 16508 on hardware 16504, corresponding to applications 16502.

[0183] Hardware 16504 can be implemented in a standalone network node with general or specific components. Hardware 16504 can implement some functions via virtualization. Alternatively, hardware 16504 can be part of a larger hardware cluster (e.g., as in a data center or CPE), where many hardware nodes work together and are managed via management and orchestration 16510, which, among other things, oversees the lifecycle management of applications 16502. In some embodiments, hardware 16504 is coupled to one or more radio units, each of which includes one or more transmitters and one or more receivers that can be coupled to one or more antennas. The radio units can communicate directly with other hardware nodes via one or more appropriate network interfaces, and can be used in conjunction with virtual components to provide radio capabilities to virtual nodes, such as radio access nodes or base stations. In some embodiments, a control system QQ512 can be used to provide some signaling, which can optionally be used for communication between hardware nodes and radio units.

[0184] refer to Figure 17 , the computing device can be node 17100.

[0185] In some embodiments, the computing device 17100 may be an electronic device that can be communicatively connected to other electronic devices on a network (e.g., other computing devices, UEs, radio base stations, etc.). In certain embodiments, the node 17100 may include radio access features that provide wireless radio network access to other electronic devices such as UEs (e.g., "radio access computing device" may refer to such computing devices). For example, the node 17100 may be a base station, such as an eNodeB in LTE, a NodeB in Wideband Code Division Multiple Access (WCDMA), or other types of base stations, as well as a radio network controller (RNC), a base station controller (BSC), or other types of control nodes. Figure 17 As shown in FIG, example node 17100 includes a processor 17101, a memory 17102, an interface 17103, and an antenna 17104. These components can work together to provide various computing device functions as disclosed herein.

[0186] Processor 17101 may be a microprocessor, a controller, a microcontroller, a central processing unit, a digital signal processor, an application specific integrated circuit, a field programmable gate array, any other type of electronic circuit, or any combination of one or more of the foregoing. Processor 17101 may include one or more processor cores. In certain embodiments, some or all of the functionality provided by node 17100 described herein may be implemented by executing software instructions by processor 17101 alone or in combination with other node 17100 components (such as memory 17102).

[0187] Memory 17102 may store code (which consists of software instructions, sometimes referred to as computer program code or computer program) and / or data using non-transitory machine-readable (e.g., computer-readable) media, such as machine-readable storage media (e.g., magnetic disks, optical disks, solid-state drives, read-only memory (ROM), flash memory devices, phase-change memory) and machine-readable transmission media (e.g., electrical, optical, radio, acoustic, or other forms of propagated signals, such as carrier waves, infrared signals). For example, memory 17102 may include non-volatile memory that contains code to be executed by processor 17101. If memory 17102 is non-volatile, the code and / or data stored therein may persist even when the computing device is turned off (when power is removed). In some instances, when node 17100 is turned on, the portion of the code to be executed by processor 17101 may be copied from the non-volatile memory to the volatile memory (e.g., dynamic random access memory (DRAM), static random access memory (SRAM)) of node 17100.

[0188] Interface 17103 can be used for wired and / or wireless communication of signaling and / or data to or from node 17100. For example, interface 17103 can implement any formatting, encoding or translation to allow node 17100 to send and receive data via wired and / or wireless connection. In some embodiments, interface 17103 may include a radio circuit capable of receiving data from other devices in the network via a wireless connection and / or sending data to other devices via a wireless connection. The radio circuit may include a transmitter, receiver and / or transceiver suitable for radio frequency communication. The radio circuit can convert digital data into a radio signal with appropriate parameters (e.g., frequency, timing, channel, bandwidth, etc.). The radio signal can then be sent to the appropriate recipient via antenna 17104. In some embodiments, interface 17103 may include a network interface controller (NIC), also known as a network interface card, network adapter, local area network (LAN) adapter or physical network interface. The NIC can facilitate connecting node 17100 to other devices to allow them to communicate via wires by plugging a cable into a physical port connected to the NIC. As explained above, in certain embodiments, the processor 17101 may represent a portion of the interface 17103 , and some or all functionality described as being provided by the interface X103 may be more specifically provided by the processor 17101 .

[0189] For the sake of simplicity in describing certain aspects and features of the node 17100 disclosed herein, the components of the node 17100 are each depicted as separate boxes within a single larger box. However, in practice, one or more of the components shown in the example node 17100 may include multiple different physical elements (e.g., the interface 17103 may include terminals for coupling wires for a wired connection and a radio transceiver for a wireless connection).

[0190] Therefore, the method of the disclosed solution described herein can be implemented in node 17100 by a computer program, which includes instructions that, when executed on at least one processor, cause the at least one processor to perform actions according to any one of the above features and embodiments, where appropriate.

[0191] While the modules are shown as being implemented in software stored in memory 17102, other embodiments implement part or all of each of these modules in hardware.

[0192] Figure 18 1 is a schematic diagram illustrating an implementation of a computing device 18000 in the cloud. For example, the computing device 18000 may be a server, a distributed base station, a site data collection node, a synchronization feature evaluation node, and / or a node for target deployment. Figure 18 As shown in FIG, example computing device 18000 includes a processor 18101, a memory 18102, an interface 18103, and an antenna 18104. These components can work together to provide the various computing device functions disclosed herein.

[0193] The processor 18101 may be a microprocessor, a controller, a microcontroller, a central processing unit, a digital signal processor, an application specific integrated circuit, a field programmable gate array, any other type of electronic circuit, or any combination of one or more of the foregoing. The processor 18101 may include one or more processor cores. In certain embodiments, some or all of the functionality provided by the computing device 18000 described herein may be implemented by the processor 18101 executing software instructions alone or in combination with other computing device 18000 components (such as memory 18102).

[0194] Memory 18102 may store code (which consists of software instructions, sometimes referred to as computer program code or computer program) and / or data using non-transitory machine-readable (e.g., computer-readable) media, such as machine-readable storage media (e.g., magnetic disks, optical disks, solid-state drives, read-only memory (ROM), flash memory devices, phase-change memory) and machine-readable transmission media (e.g., electrical, optical, radio, acoustic, or other forms of propagated signals, such as carrier waves, infrared signals). For example, memory 18102 may include non-volatile memory that contains code to be executed by processor 18101. If memory 18102 is non-volatile, the code and / or data stored therein may persist even when the computing device is turned off (when power is removed). In some instances, when computing device 18000 is turned on, the portion of the code to be executed by processor 18101 may be copied from the non-volatile memory to volatile memory (e.g., dynamic random access memory (DRAM), static random access memory (SRAM)) of computing device 18000.

[0195] Interface 18103 can be used for wired and / or wireless communication of signaling and / or data to or from computing device 18000. For example, interface 18103 can implement any formatting, encoding or translation to allow computing device 18000 to send and receive data via wired and / or wireless connection. In some embodiments, interface 18103 may include a radio circuit capable of receiving data from other devices in the network via a wireless connection and / or sending data to other devices via a wireless connection. The radio circuit may include a transmitter, receiver and / or transceiver suitable for radio frequency communication. The radio circuit can convert digital data into a radio signal with appropriate parameters (e.g., frequency, timing, channel, bandwidth, etc.). The radio signal can then be sent to the appropriate recipient via antenna 18104. In some embodiments, interface 18103 may include a network interface controller (NIC), also known as a network interface card, network adapter, local area network (LAN) adapter or physical network interface. The NIC can facilitate connecting the computing device 18000 to other devices, allowing them to communicate via wires by plugging a cable into a physical port connected to the NIC. As explained above, in certain embodiments, the processor 18101 can represent a portion of the interface 18103, and some or all of the functionality described as being provided by the interface can be more specifically provided by the processor 18101.

[0196] For simplicity in describing certain aspects and features of the computing device 18000 disclosed herein, the components in the computing device 18000 are each depicted as separate blocks within a single larger block. However, in practice, one or more of the components shown in the example computing device 18000 may include multiple different physical elements (e.g., the interface 18103 may include an air interface).

[0197] Thus, the method of the disclosed solution described herein may be implemented in a computing device 18000 by a computer program comprising instructions which, when executed on at least one processor, cause the at least one processor to perform actions according to any one of the above features and embodiments, where appropriate.

[0198] While the modules are shown as being implemented in software stored in memory 18102, other embodiments implement part or all of each of these modules in hardware.

[0199] Figure 19 Two specific examples of how a computing device 18000 may be implemented in certain embodiments of the present disclosure are shown, including: 1) a special-purpose computing device 19502 using custom processing circuitry, such as an application-specific integrated circuit (ASIC) and a proprietary operating system (OS); and 2) a general-purpose computing device 19504 using a common off-the-shelf (COTS) processor and a standard OS that has been configured to provide one or more features or functionality disclosed herein.

[0200] Special-purpose computing device 19502 includes hardware 19510, including a processor 19512, an interface 19516, and memory 19518, which stores software 19520. In one embodiment, software 19520 implements the modules described with respect to the previous figures. During operation, software 19520 can be executed by hardware 19510 to instantiate a set of one or more software instances 19522. Each software instance in software instances 19522 and the portion of hardware 19510 on which it executes (whether dedicated to that software instance, utilizing a portion of available physical resources (e.g., a processor core), and / or a time slice of hardware shared in time by that software instance and other software instances 19522) form a separate virtual network element 19530A-R. Thus, in the presence of multiple virtual network elements 19530A-R, each virtual network element operates as one of the computing devices described in the previous figures.

[0201] Back to Figure 19, an example general purpose computing device 19504 includes hardware 19540, which includes a set of one or more processors 19542 (which are typically COTS processors) and interfaces 19546, and memory 19548 in which software 19550 is stored. During operation, the processor 19542 executes the software 19550 to instantiate one or more sets of one or more applications 19564A-R. Although some embodiments do not implement virtualization, alternative embodiments may use different forms of virtualization. For example, in some alternative embodiments, the virtualization layer 19554 represents the kernel of an operating system (or a shim executed on top of a base operating system) that allows the creation of multiple instances 19562A-R, called software containers, each of which can be used to execute one (or more) of the multiple sets of applications 19564A-R. In this embodiment, software containers 19562A-R (also known as virtualization engines, virtual private servers, or virtualization jails) are user spaces (typically virtual memory spaces) that can be isolated from each other and from the kernel space in which the operating system runs. In some embodiments, a set of applications running in a given user space can be prevented from accessing the memory of other processes unless explicitly allowed. In other such optional embodiments, the virtualization layer 19554 can represent a hypervisor (sometimes called a virtual machine monitor (VMM)) or a manager that executes on top of a host operating system; and each set of applications 19564A-R can run on top of a guest operating system within an instance 19562A-R called a virtual machine (in some cases, a virtual machine can be considered a tightly isolated form of software container run by the hypervisor). In some embodiments, one, some, or all applications are implemented as a monolithic kernel, which can be generated by directly compiling only a limited set of libraries (e.g., from a library operating system (LibOS) that includes libraries / drivers for OS services) with the application, which provide specific OS services required by the application. Since a single kernel can be implemented to run directly on hardware 19540, directly on a hypervisor (in which case the single kernel is sometimes described as running within a LibOS virtual machine), or in a software container, embodiments can be implemented entirely with a single kernel running directly on a hypervisor represented by virtualization layer 19554, a single kernel running within a software container represented by instances 19562A-R, or as a single kernel and a combination of the above techniques (e.g., a single kernel and a virtual machine both running directly on a hypervisor, a single kernel and multiple sets of applications running in different software containers).

[0202] The instantiation and virtualization (if implemented) of one or more sets of one or more applications 19564A-R are collectively referred to as software instances 19552. Each set of applications 19564A-R, the corresponding virtualization construct (e.g., instance 19562A-R) (if implemented), and the portion of the hardware 19540 on which they execute (whether it is hardware dedicated to that execution and / or a time slice of hardware temporally shared by the software containers 19562A-R) form a separate virtual network element 19560A-R.

[0203] Virtual network elements 19560A-R implement similar functionality as virtual network elements 19530A-R. Such virtualization of hardware 19540 is sometimes referred to as network function virtualization (NFV). Thus, NFV can be used to consolidate many network device types onto industry-standard, high-volume server hardware, physical switches, and physical storage, which may be located, for example, in data centers and customer premises equipment (CPE). However, different embodiments of the present invention may implement one or more of the software containers 19562A-R differently. Although embodiments of the present invention are shown as each instance 19562A-R corresponding to one VNE 19560A-R, alternative embodiments may implement such correspondence at a finer level of granularity; it should be understood that the techniques described herein with reference to the correspondence of instances 19562A-R to VNEs also apply to embodiments in which such a finer level of granularity and / or a single kernel is used.

[0204] Figure 19 A third exemplary ND implementation in is a hybrid computing device 19506 that includes both a custom ASIC / proprietary OS and a COTS processor / standard OS in a single ND or a single card within an ND. In certain embodiments of such a hybrid computing device, a platform virtual machine (VM), such as a VM that implements the functionality of the dedicated computing device 19502, can provide paravirtualization for the hardware present in the hybrid computing device 19506.

[0205] Although the computing devices (e.g., UE, network node, host) described herein may include the combinations of hardware components shown, other embodiments may include computing devices with different combinations of components. It should be understood that these computing devices may include any suitable combination of hardware and / or software required to implement the tasks, features, functions, and methods disclosed herein. The determination, calculation, acquisition, or similar operations described herein may be implemented by processing circuitry that may process information by, for example, converting the obtained information into other information, comparing the obtained information or the converted information with information stored in the network node, and / or performing one or more operations based on the obtained information or the converted information to process the information, and making a determination as a result of the processing. In addition, although the components are described as a single box within a larger box, or a single box nested within multiple boxes, in practice, a computing device may include multiple different physical components that constitute a single illustrated component, and functionality may be divided between separate components. For example, a communication interface may be configured to include any of the components described herein, and / or the functionality of a component may be divided between the processing circuitry and the communication interface. In another example, the non-computationally intensive functions of any such component may be implemented in software or firmware, and the computationally intensive functions may be implemented in hardware.

[0206] In certain embodiments, some or all of the functionality described herein may be provided by processing circuitry that executes instructions stored in a 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 processing circuitry without executing instructions stored on a separate or discrete device-readable storage medium, such as in a hardwired manner. In any of these particular embodiments, the processing circuitry may be configured to implement the described functionality regardless of whether instructions stored on a non-transitory computer-readable storage medium are executed. The benefits provided by such functionality are not limited to separate processing circuitry or other components of a computing device, but are enjoyed by the computing device as a whole and / or by end users and wireless networks generally.

[0207] Further definitions and examples are discussed below.

[0208] In the above description of various embodiments of the present invention, it should be understood that the terms used herein are only used to describe specific embodiments and are not intended to limit the present invention. Unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as those generally understood by those of ordinary skill in the art to which the present invention belongs. It will be further understood that terms (such as those defined in commonly used dictionaries) should be interpreted as having a meaning consistent with their meaning in the context of this specification and the related art, and unless clearly defined herein, will not be interpreted in an idealized or overly formalized sense.

[0209] When an element is referred to as being "connected," "coupled," "responsive," or variations thereof to another element, it may be directly connected, coupled, or responsive to the other element, or there may be intermediate elements. In contrast, when an element is referred to as being "directly connected," "directly coupled," "directly responsive," or variations thereof to another element, there are no intermediate elements. Similar numbers always refer to similar elements. In addition, "coupling," "connecting," "responsive," or variations thereof as used herein may include wireless coupling, connection, or response. As used herein, the singular forms "a," "an," and "the" are also intended to include the plural forms unless the context clearly indicates otherwise. For the sake of brevity and / or clarity, well-known functions or configurations may not be described in detail. The term "and / or" (abbreviated as " / ") includes any and all combinations of one or more associated listed items.

[0210] It should be understood that although the terms first, second, third, etc. may be used herein to describe various elements / operations, these elements / operations should not be limited by these terms. These terms are only used to distinguish one element / operation from another element / operation. Therefore, without departing from the teachings of the present invention, the first element / operation in some embodiments may be referred to as the second element / operation in other embodiments. Throughout the specification, the same reference numerals or the same reference numbers represent the same or similar elements.

[0211] As used herein, the terms "comprises," "comprising," "containing," "having," or variations thereof are open-ended and include one or more stated features, integers, elements, steps, components, or functions, but do not preclude the presence or addition of one or more other features, integers, elements, steps, components, functions, or groups thereof. Furthermore, as used herein, the commonly used abbreviation "such as," derived from the Latin phrase "exempli gratia," may be used to introduce or specify one or more general examples of previously mentioned items without intending to limit such items. The commonly used abbreviation "namely," derived from the Latin phrase "id est," may be used to specify a specific item from a more general statement.

[0212] Example embodiments are described herein with reference to block diagrams and / or flowchart illustrations of computer-implemented methods, apparatus (systems and / or devices), and / or computer program products. It should be understood that the blocks in the block diagrams and / or flowchart illustrations, and combinations of blocks in the block diagrams and / or flowchart illustrations, may be implemented by computer program instructions implemented by one or more computer circuits. These computer program instructions may be provided to a processor circuit of a general-purpose computer circuit, a special-purpose computer circuit, and / or other programmable data processing circuit to produce a machine such that the instructions executed by the processor of the computer and / or other programmable data processing device transform and control transistors, values stored in memory locations, and other hardware components within such circuits to implement the block diagrams and / or flowchart illustrations. Figure 1 The functions / acts specified in one or more blocks are used to create a component (function) and / or structure for implementing the functions / acts specified in the block diagram and / or flowchart block.

[0213] These computer program instructions may also be stored in a tangible computer-readable medium, which may direct a computer or other programmable data processing device to implement functions in a specific manner, so that the instructions stored in the computer-readable medium generate instructions including implementing block diagrams and / or process diagrams. Figure 1 Therefore, the embodiments of the present inventive concept may be embodied in hardware and / or in software (including firmware, resident software, microcode, etc.) running on a processor (such as a digital signal processor), which may be collectively referred to as a "circuit", "module" or variations thereof.

[0214] It should also be noted that, in some optional implementations, the function / action recorded in the frame may be different from the order recorded in the flow chart. For example, depending on the function / action involved, the two frames shown in succession can actually be performed substantially simultaneously, or these frames can sometimes be performed in reverse order. In addition, the function of a given frame of a flow chart and / or block diagram can be divided into multiple frames and / or the function of two or more frames of a flow chart and / or block diagram can be at least partially integrated. Finally, without departing from the scope of the inventive concept, other frames can be added / inserted between the frames shown, and / or frames / operations can be omitted. In addition, although some diagrams include arrows on communication paths to illustrate the main direction of communication, it should be understood that communication can occur in the direction opposite to the arrows shown.

[0215] Many changes and modifications may be made to the embodiments without substantially departing from the principles of the present invention. All of these changes and modifications are intended to be included within the scope of the present invention. Therefore, the subject matter disclosed above should be considered illustrative, not restrictive, and the examples of the embodiments are intended to cover all such modifications, enhancements, and other embodiments that fall within the spirit and scope of the present invention. Therefore, to the maximum extent permitted by law, the scope of the present invention will be determined by the broadest permissible interpretation of the present disclosure, including the examples of the embodiments and their equivalents, and should not be limited or restricted by the foregoing detailed description.

Claims

1. A computer-implemented method, performed by a computing device, for selecting an over-the-air network synchronization topology in a planned target network, the method comprising: Identifying (206) a plurality of first antenna reference point (ARP) pairs in a source network; generating (208) a first representation of the plurality of first ARP pairs in the source network; generating (218) a second representation of a plurality of second ARP pairs in the planned target network; using (220) a trained machine learning (ML) model to classify respective pairs of the second representation of the plurality of second ARP pairs in the planned target network as at least one of similar and dissimilar, the trained ML model including transferred knowledge of learned similarities between the plurality of first ARP pairs in the first representation; as well as Based on identifying a minimum number of ARPs from the corresponding pairs classified as similar that meet a criterion, a subset of the second ARP pairs is selected (222), wherein the selected subset defines the over-the-air network synchronization topology in the planned target network.

2. The method according to claim 1, wherein The knowledge includes antenna data and radio environment data from the source network.

3. The method according to any one of claims 1 to 2, wherein Identifying (206) the plurality of first ARP pairs includes: (i) marking a plurality of historical time series measurements as at least one of reliable and unreliable, the plurality of historical time series measurements including time alignment error data for pairs of the plurality of first ARP pairs, (ii) identifying and marking unreliable ARP pairs and reliable ARP pairs, and (iii) using the marked ARP pairs to select pairs of the second ARP pairs.

4. The method according to claim 3, wherein: The identifying (206) also includes: (i) identifying multiple cycles in the first representation of the multiple first ARP pairs in the source network, wherein a cycle includes a time difference between corresponding ARPs in a set of three or more ARPs, ii) calculating a sum of the time differences for each cycle in the source network, (iii) comparing the sum of the time differences for each cycle to a specified cycle threshold, (iv) when a cycle meets the cycle threshold, marking the cycle as acceptable and marking multiple edges in the cycle as reliable, (v) when a cycle does not meet the cycle threshold, marking the cycle as unacceptable and marking multiple edges in the cycle as unreliable, and (vi) if an edge exists in both an unacceptable cycle and an acceptable cycle, the edge is considered reliable.

5. The method according to any one of claims 1 to 4, wherein Generating (208) a first representation of the plurality of first ARP pairs in the source network comprises: (i) accessing data for an ARP pair in the first ARP pairs, the data comprising cell configuration, synchronization key performance indicators (KPIs) comprising air time alignment measurements, and spatial context data, (ii) generating embeddings using the accessed data, and (iii) concatenating the embeddings for the representation using a latent representation vector.

6. The method according to any one of claims 1 to 5, further comprising: An ML model is trained (210) to learn to distinguish corresponding pairs of the plurality of first ARP pairs in the first representation, the corresponding pairs to be compared to verify similarity of corresponding second pairs in the planned target network.

7. The method according to claims 1 to 6, wherein: Selecting (222) a subset of the second ARP pairs includes: (i) generating the plurality of first ARP pairs in the source network that are identified as similar or dissimilar, (ii) training the ML model on the identified first ARP pairs based on the generated plurality of first ARP pairs identified as similar or dissimilar, (iii) using the trained ML model to generate a plurality of second ARP pairs in the planned target network, and comparing the generated plurality of second ARP pairs with the first plurality of first ARP pairs in the source network that are identified as similar or dissimilar, and (iv) marking corresponding pairs of the plurality of second ARP pairs in the planned target network as at least one of candidate reliable pairs and candidate unreliable pairs.

8. The method according to any one of claims 1 to 7, wherein ARP is included in the node, and wherein selecting (222) a subset of the second ARP pairs that defines the over-the-air synchronization topology in the planned target network includes: (i) constructing a graph of the candidate reliable pairs, (ii) identifying a minimum number of nodes that reach all other nodes in the graph, and (iii) marking the minimum number of nodes as the selected subset of the second ARP pairs.

9. The method according to any one of claims 1 to 8, wherein Identifying (206) the plurality of first ARP pairs in the source network is based on the method further comprising: accessing (200) data from the source network, including configuration management and performance data; and Data is extracted (202) from the accessed data, including at least one of: cell configuration, location of antennas in the ARP, signal quality, signal path loss, and time alignment error measurement for each of the plurality of first ARP pairs.

10. The method according to any one of claims 1 to 9, wherein The multiple second ARP pairs in the planned target network are identified based on the method further comprising: accessing (212) data for the planned target network, including configuration management and performance data; and Data (214) is extracted from the accessed data, including at least one of the following for each of the plurality of second ARP pairs: a cell configuration, a location of an antenna in the ARP, an expected signal quality, and an expected signal path loss.

11. The method according to any one of claims 1 to 10, wherein Generating (220) a second representation of the plurality of second ARP pairs classified as similar in the planned target network is based on the method further comprising: Spatial infrastructure data of the geographic area of the planned target network is extracted (204, 216), including at least one of the following: building footprints, points of interest, land use information, and terrain type.

12. The method according to any one of claims 3 to 11, wherein Identifying (206) the first plurality of ARP pairs in the source network is identified from a plurality of historical time series measurements including time alignment error data for pairs of the first plurality of ARP pairs in the source network.

13. The method according to any one of claims 6 to 12, wherein The trained ML model includes at least one neural network that learns to distinguish corresponding pairs of the plurality of first ARP pairs in the first representation, which are then compared to verify similarity of corresponding second pairs in the planned target network.

14. The method according to claim 13, wherein The at least one neural network includes a Siamese neural network, which includes two identical deep neural networks, and training the ML model to learn similarities between corresponding pairs of the multiple ARP pairs in the first representation includes: (i) comparing corresponding latent representation vectors of two ARP pairs from the multiple first ARP pairs in the first representation to obtain a similarity score, (ii) based on the comparison, the similarity score includes a score identifying the two ARP pairs as similar when the two ARP pairs share the same label, and (iii) based on the comparison, the similarity score includes a score identifying the two ARP pairs as dissimilar when the two ARP pairs have different labels.

15. The method according to any one of claims 1 to 14, wherein The computing device includes one of the following: a centralized computing device communicatively connected to the source network and the planned target network, and a distributed cloud-based computing device, the distributed cloud-based computing device including one or more of the following modules: (i) a data collection module, (ii) a transmission network synchronization feature evaluation module, and (iii) a cross-network knowledge transfer module.

16. The method according to any one of claims 1 to 15, wherein The criteria include that the minimum number of ARPs from the corresponding pair classified as similar reach a remaining number of ARPs from the corresponding pair within a specified hop distance.

17. The method according to any one of claims 1 to 16, further comprising: An anchor node in the planned target network is identified (224) from the subset of the second ARP pairs.

18. A computing device (502, 506, 510, 17100, 18000), configured to select an air network synchronization topology in a planned target network, the computing device comprising: Processing circuits (14302, 17101, 18101, 19512, 19542); A memory (14304, 17102, 18102, 19518, 19548) coupled to the processing circuit, wherein the memory includes instructions that, when executed by the processing circuit, cause the computing device to perform operations including: Identifying a plurality of first antenna reference point (ARP) pairs in a source network; generating a first representation of the plurality of first ARP pairs in the source network; generating a second representation of a plurality of second ARP pairs in the planned target network; using a trained machine learning (ML) model to classify respective pairs from the second representation of the plurality of second ARP pairs in the planned target network as at least one of similar and dissimilar, the trained ML model including transferred knowledge of learned similarities between the plurality of first ARP pairs in the first representation; and Based on identifying a minimum number of ARPs from the respective pairs classified as similar that satisfy a criterion, a subset of the second ARP pairs is selected, wherein the selected subset defines the over-the-air network synchronization topology in the planned target network.

19. The computing device of claim 18, wherein: The memory includes instructions that, when executed by the processing circuitry, cause the computing device to perform further operations including any of the operations of any of claims 2 to 16.

20. A computing device (502, 506, 510, 17100, 18000) configured to select an air network synchronization topology in a planned target network, the computing device being adapted to perform operations comprising: Identifying a plurality of first antenna reference point (ARP) pairs in a source network; generating a first representation of the plurality of first ARP pairs in the source network; generating a second representation of a plurality of second ARP pairs in the planned target network; using a trained machine learning (ML) model to classify respective pairs from the second representation of the plurality of second ARP pairs in the planned target network as at least one of similar and dissimilar, the trained ML model including transferred knowledge of learned similarities between the plurality of first ARP pairs in the first representation; as well as Based on identifying a minimum number of ARPs from the respective pairs classified as similar that satisfy a criterion, a subset of the second ARP pairs is selected, wherein the selected subset defines the over-the-air network synchronization topology in the planned target network.

21. The computing device of claim 20, adapted to perform the further operations of any one of claims 2 to 17.

22. A computer program comprising program code to be executed by a processing circuit (14302, 17101, 18101, 19512, 19542) of a computing device (502, 506, 510, 17100, 18000), the computing device being configured to select an over-the-air network synchronization topology in a planned target network, whereby execution of the program code causes the computing device to perform operations comprising: Identifying a plurality of first antenna reference point (ARP) pairs in a source network; generating a first representation of the plurality of first ARP pairs in the source network; generating a second representation of a plurality of second ARP pairs in the planned target network; using a trained machine learning (ML) model to classify respective pairs from the second representation of the plurality of second ARP pairs in the planned target network as at least one of similar and dissimilar, the trained ML model including transferred knowledge of learned similarities between the plurality of first ARP pairs in the first representation; as well as Based on identifying a minimum number of ARPs from the respective pairs classified as similar that satisfy a criterion, a subset of the second ARP pairs is selected, wherein the selected subset defines the over-the-air network synchronization topology in the planned target network.

23. The computer program according to claim 22, wherein Execution of the program code causes the computing device to perform the operations according to any one of claims 2 to 17.

24. A computer program product comprising a non-transitory storage medium (14304, 17102, 18102, 19518, 19548) comprising program code to be executed by processing circuitry (14302, 17101, 18101, 19512, 19542) of a computing device (502, 506, 510, 17100, 18000), whereby execution of the program code causes the computing device to perform operations comprising: Identifying a plurality of first antenna reference point (ARP) pairs in a source network; generating a first representation of the plurality of first ARP pairs in the source network; generating a second representation of a plurality of second ARP pairs in the planned target network; using a trained machine learning (ML) model to classify respective pairs from the second representation of the plurality of second ARP pairs in the planned target network as at least one of similar and dissimilar, the trained ML model including transferred knowledge of learned similarities between the plurality of first ARP pairs in the first representation; as well as A subset of the second ARP pairs is selected based on identifying a minimum number of ARPs from the corresponding pairs classified as similar that meet a criterion, wherein the selected subset defines an over-the-air network synchronization topology in the planned target network.

25. The computer program product of claim 24, wherein: Execution of the program code causes the computing device to perform the operations according to any one of claims 2 to 17.

Images