US-12621038-B2 - Optimizing beam-search time and beam selection in wireless networks

US12621038B2US 12621038 B2US12621038 B2US 12621038B2US-12621038-B2

Abstract

Optimizing or improving the operations of selecting a beam pair for communication and updating the beam pair are disclosed. A model is trained such that a beam pair can be selected based on a geolocation of user equipment. Once a beam pair is selected, the node and/or user equipment are configured such that subsequent transmissions use the selected beam pair. The update time is determined using the user equipment's location and velocity. The update period is based on a time required for the user equipment to reach an area associated with a different beam pair as determined using a decision function of the beam pair selection model.

Inventors

Ludwing Ferney Marenco Camacho
Luiz Eduardo Hupalo
Felipe Agusto Pereira de Figueiredo
Cristiani Vilela Ribeiro Gumarães
Jaumir Valença da Silveira Junior

Assignees

DELL PRODUCTS L.P.

Dates

Publication Date: 20260505
Application Date: 20240125

Claims (18)

1 . A method comprising: receiving coordinates from a user equipment at a beam pair configuration system integrated with a node of a network; performing a beam selection operation to select a most likely beam pair using a beam selection engine that is trained to predict the most likely beam pair for communicating with the user equipment, wherein an input to the beam selection engine includes the coordinates of the user equipment; configuring the node and the user equipment to communicate using the most likely beam pair during a next transmission; and determining an update time based on the coordinates of the user equipment and a velocity of the user equipment, wherein determining the update time includes determining contours of areas associated with beam pairs using a decision function of the beam selection engine and determining a distance to a closest contour, wherein the update time is a time required for the user equipment to reach the closest contour.
2 . The method of claim 1 , wherein the coordinates comprise geolocation data or GPS (Global Positioning System) data.
3 . The method of claim 1 , wherein the beam selection operation is performed using only the coordinates of the user equipment.
4 . The method of claim 1 , further comprising receiving the velocity of the user.
5 . The method of claim 1 , further comprising performing a next beam selection operation to select a next beam pair when the update time expires or when the velocity changes by more than a threshold amount.
6 . The method of claim 1 , wherein the contours are determined by mapping coordinates around the node to a probability function that predicts what beam pair is most likely to occur in a given coordinate, wherein contours correspond to coordinates with low probabilities.
7 . The method of claim 1 , further comprising collecting data for training the beam selection engine, the collected data including beam pairs, at least one signal quality characteristic, and user equipment coordinates, wherein the data is collected by dividing a range of the node into annuli and dividing each annulus into sectors.
8 . The method of claim 7 , wherein the signal quality characteristic is one of a power spectral density, a signal strength, a noise level, or a signal to noise ratio.
9 . The method of claim 1 , further comprising configuring the node and the user equipment to communicate using the most likely beam pair, wherein the update time is dynamic and wherein the beam selection engine includes one of a multilayer perceptron, a random forest, or a K nearest neighbor model.
10 . A non-transitory storage medium having stored therein instructions that are executable by one or more hardware processors to perform operations comprising: receiving coordinates from a user equipment at a beam pair configuration system integrated with a node of a network; performing a beam selection operation to select a most likely beam pair using a beam selection engine that is trained to predict the most likely beam pair for communicating with the user equipment, wherein an input to the beam selection engine includes the coordinates of the user equipment; configuring the node and the user equipment to communicate using the most likely beam pair during a next transmission; and determining an update time based on the coordinates of the user equipment and a velocity of the user equipment, wherein determining the update time includes determining contours of areas associated with beam pairs using a decision function of the beam selection engine and determining a distance to a closest contour, wherein the update time is a time required for the user equipment to reach the closest contour.
11 . The non-transitory storage medium of claim 10 , wherein the coordinates comprise geolocation data or GPS (Global Positioning System) data.
12 . The non-transitory storage medium of claim 10 , wherein the beam selection operation is performed using only the coordinates of the user equipment.
13 . The non-transitory storage medium of claim 10 , further comprising receiving the velocity of the user.
14 . The non-transitory storage medium of claim 10 , further comprising performing a next beam selection operation to select a next beam pair when the update time expires or when the velocity changes by more than a threshold amount.
15 . The met non-transitory storage medium hod of claim 10 , wherein the contours are determined by mapping coordinates around the node to a probability function that predicts what beam pair is most likely to occur in a given coordinate, wherein contours correspond to coordinates with low probabilities.
16 . The non-transitory storage medium of claim 10 , further comprising collecting data for training the beam selection engine, the collected data including beam pairs, at least one signal quality characteristic, and user equipment coordinates, wherein the data is collected by dividing a range of the node into annuli and dividing each annulus into sectors.
17 . The non-transitory storage medium of claim 16 , wherein the signal quality characteristic is one of a power spectral density, a signal strength, a noise level, or a signal to noise ratio.
18 . The non-transitory storage medium of claim 10 , further comprising configuring the node and the user equipment to communicate using the most likely beam pair, wherein the update time is dynamic and wherein the beam selection engine includes one of a multilayer perceptron, a random forest, or a K nearest neighbor model.

Description

FIELD OF THE INVENTION Embodiments of the present invention generally relate to improving beam search time and beam selection in wireless networks. More particularly, at least some embodiments of the invention relate to systems, hardware, software, computer-readable media, and methods for improving network performance by reducing latency and overhead associated with beam searching and beam selecting in wireless networks. BACKGROUND In order to obtain greater bandwidths and transmit at higher rates, particularly for frequencies above 24 GigaHertz (GHz), it is useful to focus transmitted power in specific directions to overcome signal attenuation. This is often achieved using beamforming techniques. Beamforming techniques may include searching for a beam pair that improves or maximizes the received power. However, the time required to search for the optimum beam pair or even a suitable beam pair is significant. In mmWave frequencies, for example, beam-searching is a complex and time consuming process. Conventional beam pair selection techniques, such as a periodic exhaustive search, may not be able to adapt quickly to changing network conditions and may lead to suboptimal beamforming execution and reduced network performance due to consumption of time and resources that could otherwise be directed to data transmission. BRIEF DESCRIPTION OF THE DRAWINGS In order to describe the manner in which at least some of the advantages and features of the invention may be obtained, a more particular description of embodiments of the invention will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings. Understanding that these drawings depict only typical embodiments of the invention and are not therefore to be considered to be limiting of its scope, embodiments of the invention will be described and explained with additional specificity and detail through the use of the accompanying drawings, in which: FIG. 1 discloses aspects of a network such as a cellular network; FIG. 2 discloses aspects of collecting data related to optimal beam pairs in a network; FIG. 3 discloses aspects of training a model to predict a beam pair for use by a node and a user equipment; FIG. 4 discloses aspects of searching for and selecting a beam pair in a network; FIG. 5 discloses aspects of selecting or determining a beam selection period or a beam selection update period; FIG. 6 discloses aspects of a method for determining a beam selection period; FIG. 7 discloses aspects of a beamforming operation in a system that includes a beam selection engine and a beam period engine; FIG. 8A discloses aspects of parameters used to generate or collect a dataset for training a model related to selecting a beam pair in a network; FIG. 8B discloses aspects of data sampling; FIG. 8C discloses aspects of a final data set collected from an environment and related to selecting a beam pair in a network; FIG. 8D discloses aspects of a decision function for a trained model; FIG. 8E discloses aspects of a comparison between selecting a beam pair using an exhaustive search method and a machine learning assisted beam selection method; FIG. 8F discloses aspects of a comparison between determining a beam selection update period in the context of exhaustive search and in the context of machine learning assisted beam pair selection; FIG. 8G discloses aspect of gains in terms of throughput; and FIG. 9 discloses aspects of a computing device, system, or entity. DETAILED DESCRIPTION OF SOME EXAMPLE EMBODIMENTS Embodiments of the present invention generally relate to improving beam pair searching time and beam pair selection in wireless networks. More particularly, at least some embodiments of the invention relate to systems, hardware, software, computer-readable media, and methods for improving network performance by reducing latency and overhead while improving energy efficiency in the context of beam-searching and beam pair selection in networks. Embodiments of the invention further relate to improving or optimizing a beam pair update period. In networks including cellular networks, beam pairs are used by a node (e.g., a gNodeB (gNB)) to communicate with a user equipment (UE). Beam pairs are often used in the context of beamforming technologies. However, it is often necessary to change the beam pair used for communication due in part to movement of the user equipment in the environment. Conventionally, an exhaustive beam pair search is performed and a specific beam pair is selected after performing the exhaustive search. The beam pair search is performed according to a beam pair update period. Performing the beam pair search and selection operation periodically (according to the beam pair update period) can impact the time available for data transmission using the selected beam pair. Embodiments of the invention advantageously improve or optimize these operations to reduce latencies and allow more time to be used for data