US-12628121-B2 - Obtaining machine learning (ML) models for secondary method of orientation detection in user equipment (UE)

Abstract

Various techniques are provided for receiving, by a base station (BS) from a user equipment (UE), a communication including a feature vector, storing, by the BS, a dataset including one or more feature vectors associated with the UE, communicating, by the BS to a network device, the dataset associated with the UE, receiving, by the BS from the network device, a machine learning (ML) model, the ML model being trained, using the dataset, to detect UE orientation, and communicating, by the BS to the UE, the trained ML model.

Inventors

• Samad ALI
• Sofonias Hailu
• Satya Krishna JOSHI
• Rauli Jarkko Kullervo JÄRVELÄ

Assignees

• NOKIA TECHNOLOGIES OY

Dates

Publication Date

20260512

Application Date

20210309

Claims (17)

1. 1 . An apparatus comprising: at least one processor; and at least one memory including computer program code; the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to: receive, from a user equipment (UE), a communication including a feature vector; store a dataset including one or more vectors associated with the UE; communicate, to a network device, the dataset associated with the UE; receive, from the network device, a machine learning (ML) model, the ML model being trained, using the dataset, to detect UE orientation; and communicate, to the UE, the trained ML model, wherein the apparatus is further caused to: receive, from the UE, a communication indicating a sensor failure at the UE; and receive, from the UE, a communication including a request for the trained ML model, wherein the receiving of the trained ML model from the network device is in response to a request for the trained ML model based on the communication of the request for the trained model from the UE.
2. 2 . The apparatus of claim 1 , wherein the dataset includes UE information associated with orientation.
3. 3 . The apparatus of claim 1 , wherein the feature vector includes information associated with at least one of Reference Signal Received Power (RSRP) measurements, Signal to Interference and Noise Ratio measurements, serving mobile terminal panel, serving receiver beam, serving transmitter beam(s), or UE orientation.
4. 4 . The apparatus of claim 1 , wherein the feature vector includes at least one of RSRPs of strongest beams to the BS, selected UE panel, receiver beam, or UE orientation per each reported RSRP.
5. 5 . The apparatus of claim 1 , wherein the trained ML model maps a set of RSRP, selected UE panel and selected receiver beam, and serving transmitter beam(s) to the UE orientation.
6. 6 . The apparatus of claim 1 , wherein the communication including the request for the trained ML model received from the UE includes UE device type information and UE sensor type information.
7. 7 . The apparatus of claim 1 , wherein the request for the trained ML model communicated to the network device includes UE device type information and UE sensor type information.
8. 8 . The apparatus of claim 1 , wherein the apparatus further caused to: select the trained ML model; and evaluate an accuracy of the trained ML model based a test dataset, wherein the communicating of the trained ML model is in response to determining the trained ML model meets an accuracy criterion.
9. 9 . The apparatus of claim 1 , wherein the apparatus further caused to: receive, from the UE, a communication including UE orientation determined using the trained ML model.
10. 10 . An apparatus comprising: at least one processor; and at least one memory including computer program code; the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to: generate a feature vector based on measurements performed by the apparatus; communicate, to a base station (BS), a message including the feature vector; and receive, from the BS, a machine learning (ML) model, the ML model being trained, using the feature vector, to detect an orientation of the apparatus, further causing the apparatus to: determine at least one sensor failure, the sensor being associated with determining orientation of the apparatus; communicate, to the BS, a message indicating the sensor failure at the apparatus; and communicate, to the BS, a message including a request for the trained ML model.
11. 11 . The apparatus of claim 10 , wherein the feature vector includes information associated with at least one of Reference Signal Received Power (RSRP) measurements, Signal to Interference and Noise Ratio measurements, serving mobile terminal panel, serving receiver beam, serving transmitter beam(s), or orientation of the apparatus.
12. 12 . The apparatus of claim 10 , wherein the feature vector includes at least one of RSRPs of strongest beams to the BS, selected panel, receiver beam, or orientation of the apparatus per each reported RSRP.
13. 13 . The apparatus of claim 10 , wherein the trained ML model maps a set of RSRP, selected panel, selected receiver beam and serving transmitter beam(s) to the orientation of the apparatus.
14. 14 . The apparatus of claim 10 , wherein the communication including the request for the trained ML model includes device type information of the apparatus and sensor type information of the apparatus.
15. 15 . The apparatus of claim 10 , further causing the apparatus to communicate, to the BS, a message including orientation of the apparatus determined using the trained ML model.
16. 16 . A method comprising generating, by a user equipment (UE), a feature vector based on UE measurements; communicating, by the UE to a base station (BS), a message including the feature vector; and receiving, by the UE from the BS, a machine learning (ML) model, the ML model being trained, using the feature vector, to detect UE orientation, further comprising: determining, by the UE, at least one sensor failure, the sensor being associated with determining UE orientation; communicating, by the UE to the BS, a message indicating the sensor failure at the UE; and communicating, by the UE to the BS, a message including a request for the trained ML model.
17. 17 . The method of claim 16 , wherein the trained ML model maps a set of RSRP, selected UE panel and selected receiver beam, and serving transmitter beam(s) to the UE orientation.

Description

RELATED APPLICATION This application was originally filed as PCT Application No. PCT/US2021/070255 on Mar. 9, 2021, which is incorporated herein by reference in its entirety. TECHNICAL FIELD This description relates to wireless communications. BACKGROUND A communication system may be a facility that enables communication between two or more nodes or devices, such as fixed or mobile communication devices. Signals can be carried on wired or wireless carriers. An example of a cellular communication system is an architecture that is being standardized by the 3rd Generation Partnership Project (3GPP). A recent development in this field is often referred to as the long-term evolution (LTE) of the Universal Mobile Telecommunications System (UMTS) radio-access technology. E-UTRA (evolved UMTS Terrestrial Radio Access) is the air interface of 3GPP's Long Term Evolution (LTE) upgrade path for mobile networks. In LTE, base stations or access points (APs), which are referred to as enhanced Node AP (eNBs), provide wireless access within a coverage area or cell. In LTE, mobile devices, or mobile stations are referred to as user equipments (UE). LTE has included a number of improvements or developments. Aspects of LTE are also continuing to improve. 5G New Radio (NR) development is part of a continued mobile broadband evolution process to meet the requirements of 5G, similar to earlier evolution of 3G and 4G wireless networks. 5G is also targeted at the new emerging use cases in addition to mobile broadband. A goal of 5G is to provide significant improvement in wireless performance, which may include new levels of data rate, latency, reliability, and security. 5G NR may also scale to efficiently connect the massive Internet of Things (IoT) and may offer new types of mission-critical services. For example, ultra-reliable and low-latency communications (URLLC) devices may require high reliability and very low latency. SUMMARY According to an example embodiment, a method may include receiving, by a base station (BS) from a user equipment (UE), a communication including a feature vector, storing, by the BS, a dataset including one or more feature vectors associated with the UE, communicating, by the BS to a network device, the dataset associated with the UE, receiving, by the BS from the network device, a machine learning (ML) model, the ML model being trained, using the dataset, to detect UE orientation, and communicating, by the BS to the UE, the trained ML model. Implementations can include one or more of the following features. For example, the dataset can include UE information associated with orientation. The feature vector can include information associated with at least one of Reference Signal Received Power (RSRP) measurements, Signal to Interference and Noise Ratio (SINR) measurements, serving mobile terminal (MT) panel, serving receiver (Rx) beam, serving transmitter (Tx) beam(s), and UE orientation. The feature vector can include at least one of RSRPs of strongest beams to the BS, selected UE panel, Rx beam, and UE orientation per each reported RSRP. The ML model maps a set of RSRP, selected UE panel and selected Rx beam, and serving Tx beam(s) to the UE orientation. The method can further include receiving, by the BS from the UE, a communication indicating a sensor failure at the UE, and receiving, by the BS from the UE, a communication including a request for the trained ML model, wherein the receiving of the trained ML model by the BS from the network device is in response to a request for the ML model based on the communication of the request for the trained model from the UE. The communication including the request for the trained ML model received from the UE can include UE device type information and UE sensor type information. The request for the trained ML model communicated to the network device can include UE device type information and UE sensor type information. The method can further include selecting, by the BS, the ML model, and evaluating, by the BS, an accuracy of the trained ML model based a test dataset, wherein the communicating of the trained ML model by the BS is in response to determining the trained ML model meets an accuracy criterion. The method can further include receiving, by the BS from the UE, a communication including UE orientation determined using the ML model. According to another example embodiment, a method may include generating, by a user equipment (UE), a feature vector based on UE measurements, communicating, by the UE to a base station (BS), a message including the feature vector, and receiving, by the UE from the BS, a machine learning (ML) model, the ML model being trained, using the feature vector, to detect UE orientation. Implementations can include one or more of the following features. For example, the message can further include UE information associated with orientation. The feature vector can include information associated with at least one of Reference Signal R