Apparatus and method for los training dataset imbalance detection

Abstract

A method for a UE of a wireless communication system is provided, the method comprising: detecting an imbalance in a LOS training dataset used to train a line-of-sight, LOS, detector; determining a minority class of the LOS training dataset associated with the detected imbalance; transmitting a request message to a network element of the wireless communication system, the request message indicating the determined minority class and requesting a measurement activation for one or more measurements to be performed by the UE to obtain one or more additional signal samples for the minority class; and receiving an activation message from the network element, the activation message causing the UE to perform the one or more measurements during a configured observation window.

Description

Technical Field

[0001] This invention relates to communications. Background Technology

[0002] User equipment (UE) can measure and report a metric called Time of Arrival (TOA) to the wireless communication network. TOA can be used to calculate the distance between the UE and network nodes such as base stations. However, if TOA is measured incorrectly, the distance may be miscalculated. Therefore, providing further solutions aimed at improving the accuracy of TOA measurement may be beneficial. Summary of the Invention

[0003] According to one aspect, the subject matter of the independent claim is provided.

[0004] According to one aspect, an apparatus is provided, comprising components for performing the following operations: detecting an imbalance in a LOS training dataset used to train a line-of-sight LOS detector by a user equipment (UE) of a wireless communication system; determining a minority class of the LOS training dataset associated with the detected imbalance; transmitting a request message to a network element of the wireless communication system, the request message indicating the determined minority class and requesting measurement activation for the UE to perform one or more measurements to obtain one or more additional signal samples for the minority class; and receiving an activation message from the network element, the activation message causing the UE to perform one or more measurements during a configured observation window.

[0005] According to one aspect, an apparatus is provided, comprising components for performing the following operations: receiving a request message from a user equipment (UE) of a wireless communication system by a network element of the wireless communication system, the request message indicating a minority class associated with an imbalance in a LOS training dataset used to train a line-of-sight LOS detector, and requesting measurement activation to be performed by the UE to obtain one or more additional signal samples for the minority class; and transmitting an activation message to the UE, the activation message causing the UE to perform one or more measurements during a configured observation window.

[0006] According to one aspect, a method for a user equipment (UE) in a wireless communication system is provided, the method comprising: detecting an imbalance in a LOS training dataset used to train a line-of-sight (LOS) detector; determining a minority class of the LOS training dataset associated with the detected imbalance; transmitting a request message to a network element of the wireless communication system, the request message indicating the determined minority class and requesting measurement activation to be performed by the UE to obtain one or more additional signal samples for the minority class; and receiving an activation message from the network element, the activation message causing the UE to perform one or more measurements during a configured observation window.

[0007] According to one aspect, a method for a network element of a wireless communication system is provided, the method comprising: receiving a request message from a user equipment (UE) of the wireless communication system, the request message indicating a minority class associated with an imbalance in a LOS training dataset used to train a line-of-sight (LOS) detector, and requesting measurement activation to be performed by the UE to obtain one or more additional signal samples for the minority class; and transmitting an activation message to the UE, the activation message causing the UE to perform one or more measurements during a configured observation window.

[0008] According to one aspect, a computer program is provided, the computer program including instructions for causing a device to perform the following operations: detecting an imbalance in a LOS training dataset used to train a line-of-sight LOS detector by a user equipment (UE) of a wireless communication system; determining a minority class of the LOS training dataset associated with the detected imbalance; transmitting a request message to a network element of the wireless communication system, the request message indicating the determined minority class and requesting measurement activation to be performed by the UE to obtain one or more additional signal samples for the minority class; and receiving an activation message from the network element, the activation message causing the UE to perform one or more measurements during a configured observation window.

[0009] According to one aspect, a computer program is provided, the computer program including instructions for causing a device to perform the following operations: receiving a request message from a user equipment (UE) of a wireless communication system by a network element of the wireless communication system, the request message indicating a minority class associated with an imbalance in a LOS training dataset used to train a line-of-sight LOS detector, and requesting measurement activation to be performed by the UE to obtain one or more additional signal samples for the minority class; and transmitting an activation message to the UE, the activation message causing the UE to perform one or more measurements during a configured observation window.

[0010] Some embodiments are defined in the dependent claims.

[0011] Embodiments that are not within the scope of the claims are to be interpreted as examples useful for understanding this disclosure.

[0012] One or more examples of the implementation are illustrated in more detail in the accompanying drawings and the description below. Other features will be apparent from the specification, drawings, and claims. Attached Figure Description

[0013] Some embodiments will be described below with reference to the accompanying drawings, in which...

[0014] Figure 1A Examples of wireless communication systems to which embodiments can be applied are shown;

[0015] Figure 1B This illustrates the principle of arrival time estimation;

[0016] Figure 1C and Figure 1D Some examples are shown;

[0017] Figure 2 and Figure 3 A flowchart according to some embodiments is shown;

[0018] Figure 4 and Figure 5 Signal diagrams according to some embodiments are shown;

[0019] Figure 6A , Figure 6B , Figure 6C , Figure 7 , Figure 8A , Figure 8B , Figure 9 and Figure 10 Some embodiments are shown; and

[0020] Figure 11 and Figure 12 An apparatus according to some embodiments is shown. Detailed Implementation

[0021] The following embodiments are examples. Although the specification may refer to "an," "one," or "some" embodiments in multiple places, this does not necessarily mean that such references refer to (multiple) the same embodiments or that the feature applies only to a single embodiment. Individual features of different embodiments may also be combined to provide other embodiments. Furthermore, the words "comprising" and "including" should be understood not to limit the described embodiments to consisting only of those features already mentioned, and such embodiments may also include features / structures not specifically mentioned.

[0022] In the following description, radio access architectures based on Advanced Long Term Evolution (LTE-A) or New Radio (NR, 5G) will be used as examples of access architectures to which embodiments can be applied, to illustrate different exemplary embodiments without limiting the embodiments to such architectures. Those skilled in the art will recognize that, by appropriately adapting parameters and processes, the embodiments can also be applied to other types of communication networks with suitable modules. Some examples of other options for applicable systems are Universal Mobile Telecommunications System (UMTS) radio access network (UTRAN or E-UTRAN), Long Term Evolution (LTE), wireless local area network (WLAN or WiFi), Global Microwave Access Interoperability (WiMAX), Bluetooth®, Personal Communication Services (PCS), ZigBee®, Wideband Code Division Multiple Access (WCDMA), systems using Ultra Wideband (UWB) technology, sensor networks, Mobile Ad Hoc Networks (MANET), and Internet Protocol Multimedia Subsystem (IMS), or any combination thereof.

[0023] Figure 1A A simplified example of a system architecture is depicted, showing only some components and functional entities, whose implementations may differ from those shown. Figure 1A The connections shown are logical connections; the actual physical connections may differ. It will be clear to those skilled in the art that the system typically includes, in addition to... Figure 1A Other functions and structures besides those shown.

[0024] However, the embodiments are not limited to the system given as an example, but those skilled in the art can apply this solution to other communication systems with the necessary characteristics.

[0025] Figure 1A The example illustrates a portion of an exemplary radio access network. Figure 1A Terminal devices or user equipment 100 and 102 are shown, configured to wirelessly connect with an access node (such as an (e / g)NodeB) 104 providing the cell on one or more communication channels within the cell. An (e / g)NodeB refers to an eNodeB or gNodeB as defined in the 3GPP specifications. The physical link from the user equipment to the (e / g)NodeB is called an uplink or reverse link, while the physical link from the (e / g)NodeB to the user equipment is called a downlink or forward link. It should be understood that an (e / g)NodeB, or its functionality, can be implemented using any entity such as a node, host, server, or access point (AP) suitable for such purposes.

[0026] A communication system typically includes more than one (e / g)NodeB, in which case the (e / g)NodeBs may also be configured to communicate with each other via wired or wireless links designed for this purpose. These links can be used for signaling purposes, but can also be used to route data from one (e / g)NodeB to another. An (e / g)NodeB is a computing device configured to control the radio resources of the communication system to which it is coupled. A NodeB may also be referred to as a base station, access point, access node, or any other type of interface device including relay stations capable of operating in a wireless environment. An (e / g)NodeB includes or is coupled to a transceiver. From the transceiver of the (e / g)NodeB, a connection is provided to an antenna element, establishing a bidirectional radio link to the user equipment. The antenna element may include multiple antennas or antenna elements. The (e / g)NodeB is further connected to the core network 110 (CN or Next Generation Core NGC). Depending on the system, the counterpart on the CN side can be a Serving Gateway (S-GW, which routes and forwards user data packets), a Packet Data Network Gateway (P-GW, which provides connectivity between the User Equipment (UE) and external packet data networks), or a Mobility Management Entity (MME), etc.

[0027] A user device (also known as a UE, user equipment, user terminal, terminal equipment, etc.) represents a type of device to which resources on the air interface are allocated and assigned, and therefore any features described herein using a user device can be implemented using a corresponding device, such as a relay node. An example of such a relay node is a Layer 3 relay (self-backhaul relay) toward a base station.

[0028] User equipment (UAE) typically refers to portable computing devices, including wireless mobile communication devices that operate with or without a Subscriber Identity Module (SIM), including but not limited to: mobile stations (mobile phones), smartphones, personal digital assistants (PDAs), handsets, devices using wireless modems (such as alarms or measuring devices), portable computers and / or touchscreen computers, tablets, game consoles, laptops, and multimedia devices. It should be understood that UAE can also be a virtually exclusive uplink device, an example of which is a camera or camcorder that loads images or video clips onto a network. UAE can also be a device capable of operating in Internet of Things (IoT) networks, such as Industrial IoT (IIoT) networks, where objects are provided with the ability to transmit data over the network without human-to-human or human-to-computer interaction. UAE can also utilize the cloud. In some applications, UAE may include small portable devices with radio components (such as watches, headphones, or glasses), and computation may be performed in the cloud. UAE (or, in some embodiments, a Layer 3 relay node) is configured to perform one or more of the functions of a UAE. User equipment may also be referred to as subscriber unit, mobile station, remote terminal, access terminal, user terminal, or user equipment (UE), with only a few names or devices mentioned. User equipment in this document may also refer to vehicle implementations, such as vehicle UEs. Such UEs may be included in and / or communicatively coupled to a vehicle such that they can be understood as part of one or more vehicles.

[0029] The various techniques described in this article can also be applied to cyber-physical systems (CPS) (systems that control collaborative computing elements of physical entities). CPS can realize and utilize a large number of interconnected ICT devices (sensors, actuators, processors, microcontrollers, etc.) embedded in different locations within physical objects. Mobile cyber-physical systems (where the physical systems under discussion have inherent mobility) are a subcategory of cyber-physical systems. Examples of mobile physical systems include mobile robots and electronic devices transported by humans or animals.

[0030] Furthermore, although the device is depicted as a single entity, different units, processors, and / or memory units can be implemented (not necessarily in the same way). Figure 1A (As shown in the image).

[0031] 5G supports the use of multiple-input multiple-output (MIMO) antennas, and significantly more base stations or nodes than LTE (the so-called small cell concept), including macro sites that collaborate with smaller base stations and employ multiple radio technologies, depending on service requirements, use cases, and / or available spectrum. 5G mobile communications support a wide range of use cases and related applications, including video streaming, augmented reality, different data sharing methods, and various forms of machine-type applications (such as (massive) machine-type communications (mMTC)), including vehicle safety, different sensors, and real-time control. 5G is expected to have multiple radio interfaces: sub-6 GHz, centimeter wave (cmWave), and millimeter wave (mmWave), and will be able to integrate with existing legacy radio access technologies such as LTE. Integration with LTE can be achieved, at least in the early stages, as a system where macro coverage is provided by LTE and 5G radio interface access originates from small cells via aggregation to LTE. In other words, 5G is planned to simultaneously support inter-RAT interoperability (such as LTE-5G) and inter-RI interoperability (inter-radio interface interoperability, such as sub-6 GHz - cmWave, and above 6 GHz - mmWave). One of the concepts believed to be used in 5G networks is network slicing, in which multiple independent and dedicated virtual subnets (network instances) can be created within the same infrastructure to run services with different requirements for latency, reliability, throughput, and mobility.

[0032] The current architecture in LTE networks is entirely distributed across radios and typically centralized within the core network. Low-latency applications and services in 5G require content to be closer to the radios, leading to localized bursts and multiple access edge computing (MEC). 5G enables analytics and knowledge generation to occur at the data source. This approach leverages resources that may not be continuously connected to the network, such as laptops, smartphones, tablets, and sensors. MEC provides a distributed computing environment for hosting applications and services. It also has the ability to store and process content near cellular subscribers to accelerate response times. Edge computing encompasses a wide range of technologies, such as wireless sensor networks, mobile data acquisition, mobile signature analytics, collaborative distributed peer-to-peer self-organizing networks and processing (which can also be categorized as local cloud / fog computing and grid / mesh computing), dew computing, mobile edge computing, cloudlets, distributed data storage and retrieval, autonomous self-healing networks, remote cloud services, augmented and virtual reality, data caching, the Internet of Things (IoT) (massive connectivity and / or latency critical), and critical communications (autonomous vehicles, traffic safety, real-time analytics, time-critical control, healthcare applications).

[0033] The communication system can also communicate with other networks, such as the public switched telephone network or the Internet, or utilize the services provided by them. The communication network may also be able to support the use of cloud services; for example, at least a portion of the core network operation can function as a cloud service (this is in...). Figure 1A The communication system may also include a central control entity that provides facilities for different operators' networks to collaborate, for example, in spectrum sharing. (This is described in "cloud" 114).

[0034] Edge cloud can be introduced into the radio access network (RAN) by leveraging network function virtualization (NVF) and software-defined networking (SDN). Using edge cloud can mean that access node operations are performed at least partially in servers, hosts, or nodes, coupled to a remote radio head or base station, including the radio portion. Node operations can also be distributed across multiple servers, nodes, or hosts. The application of the cloudRAN architecture enables real-time RAN functions to be executed on the RAN side (in the distributed unit DU 104), while non-real-time functions can be executed in a centralized manner (in the centralized unit CU108).

[0035] It should also be understood that the functional allocation between core network operations and base station operations may differ from, or even not exist in, LTE. Some other technological advancements that may be used include big data and all-IP, which could potentially change how networks are built and managed. 5G (or New Radio) networks are designed to support multiple hierarchical structures, where MEC servers can be placed between the core and base stations or nodeBs (gNBs). It should be understood that MEC can also be applied to 4G networks.

[0036] 5G can also leverage satellite communications to enhance or supplement the coverage of 5G services, for example, by providing backhaul. Possible use cases include providing service continuity for machine-to-machine (M2M) or Internet of Things (IoT) devices or for passengers in vehicles, or ensuring the availability of critical communications and future rail, maritime, and / or air communications. Satellite communications can utilize geostationary Earth orbit (GEO) satellite systems, as well as low Earth orbit (LEO) satellite systems, particularly mega-constellations (systems deploying hundreds of (nano) satellites). Satellites 106 in a mega-constellation can cover several satellite-enabled network entities that create terrestrial cells. Terrestrial cells can be created via ground relay nodes 104 or gNBs located on the ground or in satellites.

[0037] It will be apparent to those skilled in the art that the described system is merely an example of a portion of a radio access system, and in practice, the system may include multiple (e / g) NodeBs, user equipment may access multiple radio cells, and the system may also include other devices such as physical layer relay nodes or other network elements. At least one of the (e / g) NodeBs may be a home (e / g) NodeB. Furthermore, various types of radio cells and multiple radio cells may be provided within the geographical area of ​​the radio communication system. Radio cells may be macrocells (or umbrella cells), which are large cells typically tens of kilometers in diameter, or smaller cells such as micro, femtocells, or picocells. Figure 1A An (e / g) NodeB can provide any type of these cells. Cellular radio systems can be implemented as multi-layer networks comprising several types of cells. Typically, in a multi-layer network, one access node provides one or more cells, and therefore multiple (e / g) NodeBs are required to provide such a network structure.

[0038] To meet the need for improved deployment and performance of communication systems, the concept of "plug and play" (e / g) NodeBs was introduced. Typically, in addition to home (e / g) NodeBs (H(e / g) NodeBs), networks capable of using "plug and play" (e / g) NodeBs also include home NodeB gateways or HNB-GWs ( Figure 1A (Not shown in the image). HNB gateways (HNB-GWs), typically installed within a carrier's network, can aggregate services from a large number of HNBs back to the core network. The network discussed in this article can refer to, for example, a cellular network, such as 5G.

[0039] like Figure 1A As indicated by the arrows, UE 100, 102 (and / or any other UE in the described system) can support device-to-device (D2D) communication. D2D communication may sometimes be referred to as sidechain communication.

[0040] The UE can measure and report a metric called Time of Arrival (TOA). This metric can be used, for example, for location and congestion detection in Radio Resource Management (RRM). The UE can also report a metric called Reference Signal Time Difference (RSTD), which can be calculated as the difference between TOA measurements. TOA can be understood as the shortest time a signal (e.g., a radio signal) takes to propagate across the distance between the transmitter and receiver. If the TOA measurement is correct, the distance can be obtained as d = TOA × c, where c = the speed of light. To calculate TOA, the receiver can estimate the Power Delay Distribution (PDP) of the radio propagation channel and select the delay at which the PDP exhibits a power peak as the TOA, such as... Figure 1BAs shown. For example, TOA can be given to the network, and the network can then calculate the distance d.

[0041] However, the strongest component (i.e., the power peak) may not always correspond to the LOS path. This is in the example. Figure 1C and Figure 1D As shown, the LOS path signal 186 between UE 100 and network node 104 may be blocked or attenuated by obstacles 184 (e.g., trees). Therefore, the power 196 of signal 184 may be reduced and lower than the power 199 of non-LOS (NLOS) signals 188, 189, such as those bounced from building 182. Consequently, NLOS signals 188, 189 may be selected as TOA, and this could mean that when NLOS signals 188, 189 propagate from network node 104 to building 182 and from building 182 to the UE (instead of the signal propagating directly between network node 104 and UE 100), the distance between UE 100 and network node 104 may be miscalculated (i.e., the determined distance may be longer than the actual distance).

[0042] Furthermore, the radio environment can be dynamic, particularly in the cmWave and mmWave bands, and the movement of the UE and / or radio obstacles within the environment can lead to transitions from a LOS state to a NLOS state. Therefore, in practice, the UE may rarely experience pure LOS or pure NLOS propagation conditions, and determining which of these conditions dominates the received signal and for how long can become a computationally intensive task. Utilizing a machine learning (ML)-based LOS detector can be beneficial. Typically, ML methods derived from supervised learning categories benefit from a balanced training set, i.e., sufficiently diverse measurements are used for training to obtain robust designs and highly accurate inference results. Imbalanced data can refer to a situation where the number of observations is not equal for all classes in the training dataset; that is, the training dataset may have multiple classes, where one or more of the aforementioned classes have fewer available observations than at least one other class. Available observations here can refer to measurements that exist and have reliability exceeding a given threshold. Sometimes, these available observations can be referred to as reliable or relevant observations, measurements, or signal samples. For example, LOS, NLOS, and Decaying LOS (ALOS) classes can appear in the LOS training dataset used to train the LOS detector. Therefore, providing a solution that can reduce or eliminate imbalance in the LOS training dataset could be beneficial. This could enable the LOS detector to work with improved accuracy in detecting LOS, NLOS, and / or ALOS cases, and thus more accurately select TOA and calculate distances.

[0043] Figure 2 A flowchart according to one embodiment is shown. (Reference) Figure 2 A method for a UE in a wireless communication network is provided, the method comprising: detecting an imbalance in a LOS training dataset used to train a LOS detector (box 202); determining a minority class of the LOS training dataset associated with the detected imbalance (box 204); transmitting a request message to a network element of the wireless communication system, the request message indicating the determined minority class and requesting measurement activation to be performed by the UE to obtain one or more additional signal samples for the minority class (box 206); and receiving an activation message from the network element, the activation message causing the UE to perform one or more measurements during a configured observation window (box 208).

[0044] Figure 3 A flowchart according to one embodiment is shown. (Reference) Figure 3 A method for a network element in a wireless communication network is provided, the method comprising: receiving a request message from a UE of a wireless communication system, the request message indicating a minority class associated with an imbalance in a LOS training dataset used to train a LOS detector, and requesting measurement activation to be performed by the UE to obtain one or more additional signal samples for the minority class (box 302); and transmitting an activation message to the UE, the activation message causing the UE to perform one or more measurements during a configured observation window (box 304).

[0045] For example, Figure 2 and Figure 3 The method described can be applied to Figure 1A Systems (e.g., wireless communication networks) and Figure 1C and Figure 1D Example. Regarding Figure 2 and Figure 3 The UEs discussed may be, for example, UE 100 or UE 102, or some other similar network devices, such as one or more circuit systems included in UE 100, 102. Regarding Figure 2 and Figure 3The network element discussed can refer to network node 104, or, for example, CN 110 / CU 108, or some other network element configured to perform the described method steps. For example, a network element can refer to one or more network entities (e.g., physically separate network entities). For example, one or more network entities can refer to network node 104 and / or location management function (LMF). One or more measurements can refer to multiple radio signal measurements performed by the UE. Additional signal samples can be acquired (e.g., collected) by the UE by performing one or more measurements. Signal samples can be indicated directly or through additional processing and / or include parameters and / or indicators such as signal power, Received Signal Strength Indicator (RSSI), Signal-to-Interference-Ratio (SINR), Signal-to-Noise Ratio (SNR), Doppler shift, Channel Impulse Response, and / or Power Delay Power. Therefore, in other words, the UE can determine the measurement results based on the performed measurements. Signal samples can belong to a class in the LOS training dataset. Therefore, for example, a minority class may have fewer relevant signal samples, which may lead to imbalance in the LOS training dataset. This imbalance can be corrected or at least mitigated by acquiring additional signal samples for the minority class. The acquired additional signal samples can be inserted into the corresponding class. Thus, for example, if a minority class signal sample is acquired, it can be inserted into the minority class.

[0046] The embodiments described above can reconfigure the UE's LOS detector with network assistance to compensate for potential performance limitations. To do this, the UE can detect the cause of the limited performance. As mentioned above, limited performance may be caused by, for example, an imbalanced training dataset. For example, a class may have fewer measurements or observations than some other class in the training dataset, where the difference between observations or measurements within a class exceeds a threshold. Once the UE detects class imbalance, signaling can be exchanged between the UE and the network to request and grant resources respectively, enabling the UE to collect measurements, thereby at least reducing (i.e., minimizing or eliminating) the imbalance, and thus enabling the reconfiguration of the LOS detector with higher accuracy. This can mean that if reconfiguration is performed using a training dataset with less imbalance, the ML-based LOS detector can operate more efficiently and / or more accurately. In the following text, we refer to signal samples, which can sometimes be understood as observations, measurements, measurement samples, or measurement results. Signal samples can be obtained, for example, by the UE performing multiple measurements.

[0047] Furthermore, measurement activation as used herein can refer to activating one or more measurements to obtain one or more additional signal samples. As described below, measurement activation can include an indication from the network that one or more measurements should be activated. Such an indication can be implemented by transmitting an activation message from the network to the UE. Additionally, in some examples, measurement activation also includes and / or an indication of the measurement configuration to be used when performing one or more measurements. In other examples, the measurement configuration can be provided to the UE before the UE requests measurement activation. According to some example embodiments, the UE can collect and / or store a set of channel measurements corresponding to X configured LOS classes. For example, X=3, labeled: LOS, NLOS, ALOS. Thus, the training dataset can have LOS, NLOS, and ALOS classes. However, this is just an example, and different classification systems can be used. Initial values ​​in the classes can be obtained via UE measurements, and / or they can be pre-stored in the UE, for example, in a laboratory environment.

[0048] UE 100 can further extend to the network (e.g., Figure 4 Network element 402) notifies its ML capabilities (see, for example) Figure 4 (See Box 404). For example, an ML capability message can instruct the UE to utilize an ML-based LOS detector.

[0049] In some examples, network element 402 can send configuration messages to UE 100 (see example...). Figure 4 (See box 406). The configuration message can include observation window configuration. That is, network element 402 can specify the duration Tc of the window, and the start time T_start relative to the scheduled message (which will be discussed in detail later).

[0050] In some examples, periodically or triggered by network element 402, UE 100 can estimate or detect the presence of class imbalance (see, for example) Figure 4 (Box 407). Class imbalance can be detected when, after selecting relevant signal samples, the ratio between the sizes of signal samples from different classes is less than, for example, a predetermined threshold (e.g., 0.5). Relevant signal samples may include, for example:

[0051] • Estimate signal samples whose signal-to-noise ratio (SNR) level is higher than a set threshold (see SNR threshold below).

[0052] • Signal samples with moderate or low Doppler shift. A threshold can also be used here.

[0053] Therefore, UE 100 can typically ignore potentially irrelevant signal samples, such as inaccurate or noisy ones. Thus, relevant signal samples can be considered when determining whether an imbalance exists in the training dataset.

[0054] Once a class imbalance is detected (e.g., in response to the detection of a class imbalance and / or after the detection of an imbalance), the UE 100 can send a reconfiguration request to the network to obtain new channel signal samples belonging to a minority class (e.g., the class with the fewest available relevant signal samples). This can be done, for example, in... Figure 4 See box 408. UE 100 can also report the severity of class imbalance to network element 402, such as the imbalance ratio, an imbalance indicator (e.g., medium or high), and / or another suitable indication of the relative size of signal samples from different classes. Here, we note that UE 100 can determine the severity of the imbalance. Therefore, for example, it can be determined how many additional signal samples are needed for the minority class to at least reduce the imbalance to an acceptable level. For example, the acceptable level can be network-configured or pre-configured, or determined by the UE (e.g., by the LOS detector). Therefore, for example, in some cases, the severity indication can simply indicate how many additional signal samples are requested to at least reduce the imbalance to an acceptable level. In one example, the level here could refer to the ratio between the number of signal samples in the minority class and the total number of signal samples.

[0055] Network element 402 can evaluate the request, and based on the location prediction and channel characteristics of UE 100 at such a predicted location, network element 402 can determine whether the minority class is likely to occur within the observation window (e.g., during the observation window). If network element 402 determines that minority class signal samples can be acquired within the observation window, network 402 can send a trigger to UE 100 (see example...). Figure 4 The observation window (box 412) is used to begin collecting new measurements in the observation window. The observation window may have been previously indicated to the UE 100, or the network element 402 may provide a new configuration to the UE 100 if it determines that the radio conditions of the UE 100 have changed. In some examples, it may be determined that the change meets the conditions (e.g., exceeding a threshold) before the observation window is reconfigured for the UE 100.

[0056] UE 100 can collect signal samples by performing (multiple) measurements within a configured observation window, labeling them with corresponding class tags (i.e., a minority class is the target, but due to the fact that radio conditions may differ from predictions, signal samples belonging to certain other classes may be acquired separately or alternatively), timestamping them, and caching them internally. Furthermore, UE 100 can add additional tags to the signal samples, as described in more detail below.

[0057] Therefore, network element 402 can estimate the radio conditions of UE 100 at a future location and determine whether UE 100 can acquire minority class signal samples under these estimated radio conditions. For example, if LOS is a minority class, and the network determines that UE 100 is experiencing NLOS radio conditions during the observation window based on UE 100's location prediction and radio condition estimation, the network may not necessarily trigger UE 100 to perform measurements because the signal samples may not reduce or eliminate the imbalance in the training dataset. However, if LOS radio conditions are predicted, triggering can be performed. In some examples, the configuration window duration and / or start time can also be adjusted. Thus, for example, if the initial observation window will result in incorrect class signal samples, the window can be reconfigured if reconfiguring the window may result in minority class signal samples.

[0058] Then, let's first refer to Figure 4 Let's take a closer look at some embodiments. Network element 402 may refer to performing, for example... Figure 3 The network element of the steps. In one embodiment, UE 100 is configured to acquire one or more additional signal samples for the minority class (box 416); and after acquiring one or more additional signal samples, to retrain the LOS detector with the LOS training set (box 418). Thus, after receiving an activation message from network element 402, UE 100 can perform multiple measurements to acquire multiple signal samples. Therefore, imbalance can be reduced at least, and the LOS detector can be retrained with a more balanced training dataset.

[0059] In one embodiment, UE 100 is also configured to receive a configuration message including observation window configuration information from network element 402. That is, network element 402 may transmit the configuration message to UE 100. The observation window configuration information may include, for example, Tc and T_start as defined above. As mentioned above, T_start may indicate the start time of the observation window relative to the index of the subframe carrying the predefined message or system frame number (SFN) frame. For example, Tc may be expressed in an absolute time format.

[0060] In one embodiment, the pre-defined message is the request message of box 408, and the subframe is an uplink subframe. Therefore, this essentially represents the start time of the observation window relative to the transmission of the request message of box 408.

[0061] In one embodiment, the predefined message is the activation message of box 412, and the subframe is a downlink subframe. Therefore, this essentially represents the start time of the observation window relative to the transmission of the activation message of box 412.

[0062] In some embodiments, prior to receiving the configuration message in block 406, the UE may transmit an ML capability message to network element 402 in block 404. This message may include a list of ML-based capabilities that can be retrained in real-time on the UE side. For example, the UE may use the aforementioned message to indicate that it has an ML-based LOS detector. Other capabilities that may be indicated include an ML-based channel estimator and an ML-based decoder. This information may be part of a general UE capability exchange process or may be exchanged on demand.

[0063] Network element 402 can determine that UE 100 has an ML-based LOS detector based on the capability message in block 404. Therefore, for example, it can determine to configure an observation window in block 406. This configuration message in block 406 is sometimes referred to as an ML configuration message, in which the observation window can be defined.

[0064] In box 407, UE 100 can perform imbalance detection to detect the presence of a class imbalance, and, if an imbalance is detected, report the detected imbalance to network element 402 by transmitting a request message in box 408. Imbalance detection (e.g., as in box 407) is related to... Figure 10 Detailed discussion. The request message can indicate a minority of categories, namely, the categories causing the imbalance and those that may require additional signal samples. Furthermore, the request message can indicate the severity of the imbalance (e.g., the imbalance ratio or a relative indicator (e.g., high, medium, low)). Additionally, the request message (sometimes called a reconfiguration request) can include data about the UE's speed, orientation, and / or future location. This information can sometimes be referred to as the UE mobility level. The UE mobility level can be reported to network element 402.

[0065] In block 410, network element 402 can assess when and / or whether the UE is likely to experience channel conditions corresponding to the reported minority class. Therefore, network element 402 can estimate the future location of the UE and further estimate the radio conditions at that future location. For example, network element 402 can use information about the radio environment (e.g., buildings, streets, etc.) and the locations of network element 402 and UE 100 to generate a geographic map (e.g., 2D / 3D), such as a LOS map. This map can be used to determine the radio conditions at certain future locations of the UE. Those skilled in the art understand that various techniques and methods known in the art are used to estimate the future location of the UE and the radio conditions at those future locations. Therefore, these will not be discussed in detail herein. For example, a two-step approach may be employed:

[0066] a. First, network element 402 bases its analysis on the reported mobility level (sometimes referred to as mobility data, such as...). Figure 6B The future location of UE 100 is predicted using data such as (box 618), historical data, time of day, road type, road layout, road direction, and / or building layout, to name just a few examples.

[0067] b. Second, network element 420 checks whether the channel conditions at the predicted location might correspond to the reported minority class, and if so, network element 402 sends a trigger message to UE 100 to begin measurement collection (box 412: activation message). That is, measurement can be activated.

[0068] UE 100 can receive trigger signaling (i.e., activation messages), which can be, for example, a Media Service Control (MAC) Control Element (CE) or Physical Layer (PHY) Downlink Control Information (DCI) from Network Element 402 (or other types of downlink control carried via the Physical Downlink Control Channel (PDCCH), and begin recording channel signal samples during a configured observation window (box 416). The UE can label these samples with a few classes and other locally generated tags, such as estimated SNR and rotating labels (RL). Figure 4As shown, in some embodiments, network element 402 can transmit a reference signal (RS) (box 414). The transmitted RS can be used in box 416. That is, UE 100 can perform measurements on (multiple) radio channels transmitting the RS and acquire signal samples. For example, the signal samples can be labeled with class labels, SNR labels, Doppler shift labels, and / or RL, to name just a few examples. In other words, UE 100 can perform (multiple) measurements during an observation window during which network element 402 can transmit RS, acquire additional signal samples, and label the additional signal samples accordingly. As described above, in box 418, UE 100 can retrain the LOS detector with an updated training dataset.

[0069] Figure 5 An embodiment is shown in which the activation message includes observation window configuration information. Therefore, instead of transmitting two messages (i.e., such as...) Figure 4 As required by the example (configuration message and activation message), network element 402 can transmit an activation message that includes the same or similar information as the configuration message. However, in some cases, network element 402 can transmit both the configuration message and the activation message, and still include the observation window configuration information in the activation message. This may be because, for example, the observation window configuration information can be updated after the configuration message is transmitted. In some examples, the activation message that includes configuration information is referred to as the configuration message.

[0070] Therefore, in box 504, UE 100 can transmit capability messages, as shown in box 404. Instead of waiting for configuration message 404 or some other trigger to begin imbalance detection, UE 100 can initiate imbalance detection, as shown in box 507. For example, imbalance detection can be periodic and performed as shown in box 407.

[0071] If an imbalance is detected, UE 100 can transmit a request message to network element 402 (box 508). This can be similar to that in box 408.

[0072] In box 510, network element 402 can perform UE location and radio condition estimation, as shown in box 410. This can be based on the UE mobility level reported by the UE in a request message, or on the UE mobility level determined by network element 402, for example, based on network information. For example, the network information can indicate the UE's previous serving cell. This information can be used to predict where the UE will move and at what speed.

[0073] In box 512, if network element 402 is thus determined, an activation message can be transmitted to UE 100 to initiate the execution of (multiple) measurements to balance an imbalanced training dataset. The activation message can configure UE 100 to utilize an indicated observation window to acquire additional signal samples for the minority class, for example, similar to what is discussed with respect to boxes 416 and 418. For example, network element 402 can be configured with... Figure 5 RS transmission in the middle, similar to Figure 4 middle.

[0074] Figure 6A , Figure 6B and Figure 6C Block diagrams of some embodiments are shown. References Figure 6A As described above, the observation window configuration information 600 may include the duration (i.e., Tc) 602 and the start time (i.e., T_start) 604 of the observation window. The start time may be an index relative to the subframe carrying the predetermined message or relative to the SFN frame. For example, Tc may be expressed as an absolute time. As described above, the predetermined message may be, for example, a request message or an activation message. In one embodiment, the observation window configuration information 600 further indicates which message is the predetermined message (e.g., a request message or an activation message). In this way, UE 100 and network element 402 can similarly determine the observation window, and therefore, for example, RS transmissions performed by network element 402 can be measured by UE 100 at the correct time.

[0075] refer to Figure 6B Request message 610 (e.g., boxes 408, 508) may include indications regarding minority class 612. Furthermore, request message 610 may include mobility data 618 (sometimes referred to as mobility level or mobility state) and / or a severity indicator 614 regarding UE 100. The severity indicator may indicate the severity of the imbalance in minority class 612. Severity indicator 614 may include, for example, an imbalance ratio 616.

[0076] Now for reference Figure 6C The diagram illustrates activation message 620 (e.g., transmitted in boxes 412 or 512). According to one embodiment, activation message 620 indicates an ACK or a NACK (box 622), wherein in the case of ACK, the activation message causes UE 100 to perform one or more measurements during a configured observation window, and wherein in the case of NACK, the activation message causes UE 100 to block the execution of one or more measurements. Therefore, essentially, if NACK is indicated, the activation message transmitted in boxes 412, 512 does not necessarily cause the triggering of measurements in all embodiments. An example of this is shown in… Figure 8B The following is shown to illustrate an embodiment.

[0077] refer to Figure 8B Network element 402 can obtain estimates of the location and radio conditions of UE 100 (box 812). That is, as referenced above. Figure 4 and Figure 5 The radio conditions at the estimated future location can be estimated. In block 814, network element 402 determines whether the estimated radio conditions are suitable during the observation window. If not suitable, the process can proceed to block 816. If suitable, the process can proceed to block 818.

[0078] In block 816, network element 402 can transmit an activation message including NACK or prevent the transmission of an activation message. In both cases, UE 100 can choose not to initiate a measurement.

[0079] In box 818, network element 402 can transmit an activation message including an ACK. Therefore, UE 100 can initiate measurements during the configured observation window.

[0080] In one embodiment, in block 814, even if radio conditions are suitable, the process can continue to block 816 if no training signal (i.e., RS) is available. That is, if network element 402 does not have resources for transmitting RS, the activation message may not be sent or may include NACK.

[0081] Figure 8A An embodiment is shown. Similarly, as Figure 8B As shown, in block 802, the future location of the UE and the radio conditions of that future location can be estimated by network element 402. Based on block 802, network element 402 determines the time for transmitting the activation message. In block 806, network element 402 can transmit the activation message at the aforementioned time. For example, the aforementioned time can refer to a specific moment. Therefore, the network element can adjust the time of transmitting the activation message to change the start time of the observation window. For example, this may be beneficial when the transmission time of the activation message affects when the configuration window begins (e.g., T_start is an index relative to the downlink subframe carrying the activation message). Therefore, without changing the configuration information of the observation window, network element 402 can adjust the time for performing (multiple) measurements, for example, to increase the probability of acquiring (multiple) signal samples for the minority class.

[0082] Figure 7 An embodiment is shown. (See reference) Figure 7Network element 402 can transmit a reference symbol configuration for retraining the LOS detector (block 702). UE 100 can receive this configuration and initiate the reception of a training signal according to the reference symbol configuration to use the reference symbol configuration when retraining the LOS detector. In block 706, network element 402 can transmit RS (which may be referred to as a training signal) according to the configuration. According to one embodiment, Figure 7 One or more steps are included Figure 4 In box 414.

[0083] In one embodiment, the reference symbol configuration is transmitted to UE 100 in at least one separate message.

[0084] In one embodiment, the reference symbol configuration is transmitted to UE 100 in a configuration message (e.g., block 406) or in an activation message (block 512).

[0085] Figure 9 An embodiment is shown. (See reference) Figure 9 Network element 402 can transmit an imbalance threshold (box 902) to UE 100. UE 100 can receive the imbalance threshold.

[0086] In box 904, UE 100 can determine whether the detected imbalance (e.g., box 202) meets (e.g., exceeds) an imbalance threshold.

[0087] If the imbalance exceeds an imbalance threshold (e.g., the imbalance threshold can be configured by network element 402), UE 100 may transmit a request message to network element 402 (e.g., box 206) (box 906). Otherwise (i.e., if the imbalance does not exceed the threshold), UE 100 may block the transmission of the request message.

[0088] The imbalance threshold can be, for example, an imbalance threshold ratio or a minimum imbalance ratio at which the UE is expected to send a request message. In some examples, this ratio is 0.5, meaning that a request message can be transmitted if a class has less than half the number of relevant signal samples compared to the class with the most relevant signal samples. However, this ratio can be different from 0.5 (e.g., 0.6 or 0.7 or other values). In one embodiment, the network determines the threshold ratio. As mentioned above, if a request message is transmitted, the message can include the severity of the class imbalance. For example, if it exceeds 0.5, the actual ratio can be indicated as being between 0 and 0.5.

[0089] In one embodiment, the imbalance threshold is transmitted to UE 100 as a separate message.

[0090] In one embodiment, the imbalance threshold is included in the configuration message (box 406).

[0091] Then, let's refer to Figure 10 Imbalance detection (e.g., boxes 202, 407, 507) will be discussed in more detail. Figure 10 Some embodiments are shown. References Figure 10 Imbalance detection can include maintaining signal samples with relevant labels such as class, SNR, and RL. For example, channel impulse response and / or power delay power (PDP) can be determined from the signal samples. These values ​​can be used as input to a LOS detector to train the LOS detector. Therefore, values ​​derived from the signal samples can be used as input to the LOS detector, for example. Based on the labeled signal samples stored in UE 100, the UE can perform imbalance detection as shown in box 1030. The maintained (i.e., stored) signal samples can be pre-stored and pre-labeled, and / or they can be acquired in box 1010 by performing multiple measurements and labeled in box 1020.

[0092] In box 1010, UE 100 can collect or acquire signal samples. The acquired signal samples can be used as input to a LOS detector. For example, raw signal samples can be input to the LOS detector. As described above, the signal samples may include information from which signal power and / or SNR 1012 can be derived. As described above, other parameters, such as PDP or channel impulse response, can be determined based on the signal samples (e.g., per set of signal samples). Furthermore, rotation measurement 1014 can be performed to obtain RL associated with the signal samples. For example, SNR (or SINR) and / or RL can be used to filter irrelevant signal samples to obtain relevant signal samples. For example, relevant signal samples can be used as input to the LOS detector.

[0093] In box 1020, signal samples can be labeled with class, SNR, and / or RL (box 1022), to name just a few examples. For instance, classes such as LOS, NLOS, and ALOS can be used.

[0094] Therefore, signal samples can have class, correlated SNR, and RL. For example, RL can be {high, medium, low} and can be derived from the rotation measurement of the UE performed by the sensor unit (box 1014) (see example). Figure 11 (Block 1150). The sensor unit may include a rotation sensor and / or an inertial measurement unit (IMU). For example, sensor unit 1150 can measure 3D rotations belonging to three different ranges. Sensor unit 1150 may have a refresh rate of tens of Hz and detect motion in any / all 3D directions, for example, at high resolution (e.g., mm range offset).

[0095] In one embodiment, the UE removes signal samples with high or medium RL or gives less weight to signal samples with high or medium RL. Thus, for example, if the RL exceeds a threshold, the UE may discard the associated signal sample. The threshold may be configured by the network or pre-configured.

[0096] If at least one of the following conditions is met, the UE 100 may detect class imbalance:

[0097] • The size of any class in the training dataset is less than a% of the size of the majority class (box 1032). a% may be, for example, 60.

[0098] • After untrusted signal samples are discarded or removed, the size of any class is less than a% of the size of the majority class (box 1034).

[0099] o Untrusted measurements may include signal samples with SNR exceeding a threshold (e.g., SNR < SNR threshold), signal samples with RL exceeding an RL threshold, and / or signal samples with Doppler shift exceeding a threshold.

[0100] • The size of any class is less than b% of the size of the majority class, and t has elapsed since the latest signal sample was collected t (box 1036).

[0101] o For example, b > a and t may be equal to the number of subframes z, where z may be a positive integer (e.g., 1, 2, 3, 4, etc.). Thus, even if the class imbalance is not as high as in the previous example using a%, the signal samples may be updated to improve their diversity.

[0102] According to one embodiment, if the detected imbalance exceeds a first threshold, or if the imbalance exceeds a second threshold and a predetermined time has elapsed since the last signal sample was obtained, the UE 100 transmits a request message. Thus, for example, if the detected imbalance ratio is less than the first threshold ratio, a request message may be transmitted. For example, if the detected imbalance ratio is less than the second threshold ratio, and ∆t has elapsed since the last signal sample was obtained, a request message may be transmitted. For example, the first threshold ratio may be less than the second threshold ratio. If the above conditions are not met, a request message may not be transmitted.

[0103] SNR and RL thresholds can be pre-configured or configured by the network for the UE 100. For example, SNR may need to be higher than a given threshold and / or RL may need to be lower than a given threshold for a signal sample to be considered relevant / credible. Similarly, Doppler shift may need to be lower than a given threshold for a signal sample to be considered relevant / credible. If a signal sample is considered relevant, it can be considered when determining class imbalance. If a signal sample is considered irrelevant / unreliable, it can be ignored when determining class imbalance.

[0104] The proposed scheme can provide benefits such as improving the accuracy of UE LOS reports to the network. This improved accuracy can be experienced, for example, when the UE helps the network determine its location. In UE-based positioning, the proposed method can improve location estimation by increasing the probability of correctly selecting the LOS TOA of a detectable cell. Furthermore, the proposed Radio Resource Control (RRC) signaling can be used for UEs to report changes in LOS conditions or to directly report estimated LOS or NLOS labels. RRC signaling can be used for certain adjustments, such as triggering new UE measurements (e.g., measurements in block 416). For example, RRC signaling can be used to transmit configuration and reconfiguration messages. Such configuration / reconfiguration messages can also be sent along with other messages, for example, as multicast / broadcast messages when the configuration is common to multiple UEs. Thus, for example, network element 402 can simultaneously configure multiple UEs to utilize the same observation window.

[0105] Figure 11 and Figure 12 Devices 1100 and 1200 are provided, each including a control circuitry (CTRL) system 1110 and 1210 (such as at least one processor) and at least one memory 1130 and 1230, the at least one memory 1130 and 1230 including computer program code (software) 1132 and 1232, wherein the at least one memory and the computer program code (software) 1132 and 1232 are configured to, together with the at least one processor, cause the corresponding device 1100 and 1200 to perform. Figures 1A to 10 Any of the embodiments or operations of the embodiments described herein.

[0106] refer to Figure 11 and Figure 12The memories 1130 and 1230 can be implemented using any suitable data storage technology, such as semiconductor-based memory devices, flash memory, magnetic memory devices and systems, optical memory devices and systems, fixed memory, and removable memory. The memories 1130 and 1230 may include databases 1134 and 1234 for storing data. For example, a training dataset can be stored in and updated within the memory 1130.

[0107] Devices 1100 and 1200 may further include radio interfaces (TRXs) 1120 and 1220, which include hardware and / or software for implementing communication connections according to one or more communication protocols. For example, the TRX may provide the device with communication capabilities to access a radio access network. The TRX may include standard, well-known components such as amplifiers, filters, frequency converters, (de)modulators, encoder / decoder circuitry, and one or more antennas. The TRX may be used to perform one or more measurements to obtain one or more additional signal samples.

[0108] Devices 1100 and 1200 may include user interfaces 1140 and 1240, including, for example, at least one keypad, microphone, touch display, display, speaker, etc. User interfaces 1140 and 1240 can be used for user control of corresponding devices of devices 1100 and 1200.

[0109] In one embodiment, device 1100 may perform the above, for example, regarding... Figure 2 The UE described in the method is either included in the UE. For example, device 1100 may be UE 100 or UE 102 or may be included in them.

[0110] In one embodiment, device 1200 may be capable of performing the above, for example, regarding... Figure 3 The network elements of the described method are either included therein. For example, device 1200 may be network element 402 or network node 104, or may be included therein.

[0111] According to one embodiment, reference Figure 11 The control circuit system 1110 includes components configured to perform at least the following: Figure 2 The detection circuitry system 1112 described in box 202 is configured to perform at least the operation of the detection circuitry system 1112. Figure 2 The circuit system 1114, which describes the operation of the block 204, is configured to perform at least the following operations: Figure 2 The transmission circuit system 1116 described in block 206; and configured to perform at least the operation of the transmission circuit system 1116 with respect to the operation of the transmission circuit system 1116; Figure 2 The operation of the receiving circuit system 1118 is described in box 208.

[0112] According to one embodiment, device 1100 includes sensor unit 1150, which has been discussed in more detail above.

[0113] The apparatus 1100 may also include a LOS detector 1160. The LOS detector 1160 may, for example, be based on machine learning (ML). For example, the LOS detector 1160 may be a reference... Figure 2 and Figure 3 The same or similar LOS detectors are discussed.

[0114] According to one embodiment, reference Figure 12 The control circuit system 1210 includes components configured to perform at least the following: Figure 3 The receiving circuit system 1212 described in block 302; and configured to perform at least the operation of the receiving circuit system 1212 with respect to the operation of the receiving circuit system 12 ... with respect to the operation of the receiving circuit system 1212 with respect to the operation of the receiving circuit system 1212 with respect to the operation of the receiving circuit system 12 Figure 3 The operation of the transmission circuit system 1214 is described in box 304.

[0115] In one embodiment, at least some of the functions of device 1200 can be shared between two physically separate devices to form an operational entity. Therefore, device 1200 can be viewed as an operational entity comprising one or more physically separate devices for performing at least some of the described processes. Thus, device 1200 utilizing such a shared architecture can include a remote control unit (RCU), such as a host computer or server computer, operatively coupled (e.g., via a wireless or wired network) to one of multiple remote radio head ends (RRHs) located in a base station or network node 104. In one embodiment, at least some of the described processes can be performed by the RCU. In one embodiment, the execution of at least some of the described processes can be shared between the RRH and the RCU. For example, CU / DU splitting can utilize such a shared architecture.

[0116] In one embodiment, the RCU can generate a virtual network through which it can communicate with the RRH. Typically, virtual networking can involve the process of combining hardware and software network resources and network functions into a single software-based management entity (virtual network). Network virtualization can involve platform virtualization, which is often combined with resource virtualization. Network virtualization can be categorized into external virtual networks, which combine many networks or parts of networks into a server computer or host computer (i.e., the RCU). External network virtualization aims to optimize network sharing. Another category is internal virtual networks, which provide network-like functionality to software containers on a single system.

[0117] In one embodiment, a virtual network can provide flexible allocation of operations between the RRH and RCU. In practice, any digital signal processing task can be performed in either the RRH or the RCU, and the boundary for transferring responsibility between the RRH and RCU can be chosen according to the implementation.

[0118] According to one aspect, a system is provided that includes a plurality of devices 1100 and one or more devices 1200. Thus, device 1200 can be configured with multiple UEs having observation windows, wherein the observation windows can be shared among the UEs or UE-specific.

[0119] As used in this application, the term "circuit system" can refer to: (a) a hardware circuit implementation, such as an implementation in an analog and / or digital circuit system; and (b) a combination of circuitry and software (and / or firmware), such as (if applicable): (i) a combination of (multiple) processors, or (ii) a portion of (multiple) processors / software, software, and (multiple) memories comprising (multiple) digital signal processors, which work together to cause a device to perform various functions; and (c) a circuit, such as (multiple) microprocessors or a portion of (multiple) microprocessors, which operates using software or firmware, even if the software or firmware is not physically present. This definition of "circuit system" applies to all uses of the term in this application. As another example, as used in this application, the term "circuit system" will also cover an implementation of only one processor (or multiple processors) or a portion of a processor and its accompanying software and / or firmware. For example, if applicable to a particular element, the term "circuit system" will also cover a baseband integrated circuit or application processor integrated circuit for a mobile phone, or a similar integrated circuit in a server, cellular network device, or another network device.

[0120] In one embodiment, combined Figures 1A to 10 At least some of the described processes can be performed by means including corresponding components for performing at least some of the above processes. Some example components for performing the processes may include at least one of the following: a detector, a processor (including dual-core and multi-core processors), a digital signal processor, a controller, a receiver, a transmitter, an encoder, a decoder, a memory, RAM, ROM, software, firmware, a display, a user interface, a display circuit system, a user interface circuit system, user interface software, display software, a circuit, an antenna, an antenna circuit system, and a circuit system. In one embodiment, at least one processor, memory, and computer program code form a processing component, or include one or more portions of computer program code for performing according to Figures 1A to 10 Any one of the embodiments in the embodiments or one or more operations of the operation thereof.

[0121] According to yet another embodiment, the apparatus for performing any embodiment includes a circuit system comprising at least one processor and at least one memory including computer program code. When activated, the circuit system causes the apparatus to perform according to... Figures 1A to 10 At least some of the functions or operations of any of the embodiments in the embodiments.

[0122] The techniques and methods described herein can be implemented by various means. For example, these techniques can be implemented using hardware (one or more devices), firmware (one or more devices), software (one or more modules), or a combination thereof. For hardware implementation, the embodiments may be implemented within one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital data processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, other electronic units designed to perform the functions described herein, or combinations thereof. For firmware or software, the implementation may be executed by a module (e.g., process, function, etc.) of at least one chipset performing the functions described herein. Software code may be stored in memory cells and executed by a processor. Memory cells may be implemented internally or externally to the processor. In the latter case, as is known in the art, memory cells may be communicatively coupled to the processor by various means. Furthermore, the components of the systems described herein may be rearranged and / or supplemented by additional components to implement the various aspects described therewith, and they are not limited to the precise configurations illustrated in the given figures, as will be understood by those skilled in the art.

[0123] The described embodiments can also be implemented as a computer process defined by a computer program or parts thereof. Figures 1A to 10 Embodiments of the described methods can be implemented by executing at least a portion of a computer program including corresponding instructions. The computer program may be in source code form, object code form, or some intermediate form, and may be stored on a medium, which may be any entity or device capable of carrying the program. For example, the computer program may be stored on a computer or processor-readable computer program distribution medium. The computer program medium may be, for example, but not limited to, recording media, computer memory, read-only memory, electrical carrier signals, telecommunication signals, and software distribution packages. For example, the computer program medium may be a non-transitory medium. The coding of the software used to perform the illustrated and described embodiments is entirely within the scope of those skilled in the art. In one embodiment, a computer-readable medium includes the computer program described above.

[0124] Although the invention has been described above with reference to the accompanying drawings and examples, it is apparent that the invention is not limited thereto, but can be modified in various ways within the scope of the appended claims. Therefore, the words and expressions used herein should be interpreted broadly, and they are intended to illustrate rather than limit the embodiments. It will be apparent to those skilled in the art that the concepts of the invention can be implemented in various ways as technology advances. Furthermore, it will be apparent to those skilled in the art that the described embodiments can (but are not required to) be combined with other embodiments in various ways.

Claims

1. A device for communication, comprising: At least one processor; as well as At least one memory including computer program code, the at least one memory and the computer program code being configured to cause the apparatus together with the at least one processor: Detect imbalances in the LOS training dataset used to train the line-of-sight LOS detector; Determine the minority class of the LOS training dataset associated with the detected imbalance; A request message is transmitted to a network element of a wireless communication system, the request message indicating the identified minority class and requesting measurement activation to perform one or more measurements to obtain one or more additional signal samples for the minority class; as well as The device receives an activation message from the network element, the activation message causing the device to perform one or more measurements during a configured observation window, wherein the activation message includes observation window configuration information.

2. The apparatus of claim 1, wherein the at least one memory and the computer program code are further configured to cause the apparatus together with the at least one processor: Acquire the one or more additional signal samples for the minority class; and After acquiring the one or more additional signal samples, the LOS detector is retrained using the LOS training dataset.

3. The apparatus of claim 1, wherein the at least one memory and the computer program code are configured to cause the apparatus together with the at least one processor: Receive a configuration message including observation window configuration information from the network element.

4. The apparatus of claim 1, wherein the observation window configuration information indicates the duration of the observation window and the start time of the observation window relative to the index of the subframe carrying the predetermined message or relative to the system frame number.

5. The apparatus of claim 4, wherein the predetermined message is the request message and the subframe is an uplink subframe, or the predetermined message is the activation message and the subframe is a downlink subframe.

6. The apparatus of claim 1, wherein the request message further includes a severity indicator indicating the severity of the imbalance.

7. The apparatus of claim 1, wherein the at least one memory and the computer program code are configured to cause the apparatus together with the at least one processor: Receive the imbalance threshold from the network element; Determine whether the imbalance meets the imbalance threshold; and Based on the determination that the imbalance meets the imbalance threshold, the request message is transmitted to the network element; otherwise, the transmission of the request message is blocked.

8. The apparatus of claim 1, wherein the activation message indicates an ACK or a NACK, wherein an ACK indicates that the apparatus performs the one or more measurements during the configured observation window, and wherein the NACK indicates that the apparatus prevents the execution of the one or more measurements.

9. The apparatus of claim 1, wherein the at least one memory and the computer program code are configured to cause the apparatus together with the at least one processor: Receive from the network element a reference symbol configuration to be used for retraining the LOS detector; and The LOS detector is retrained using the reference symbol configuration by initiating the reception of training signals according to the reference symbol configuration.

10. A device for communication, comprising: At least one processor; as well as At least one memory including computer program code, the at least one memory and the computer program code being configured to cause the apparatus together with the at least one processor: The user equipment (UE) of the wireless communication system receives a request message indicating a minority class associated with an imbalance in the LOS training dataset used to train the line-of-sight LOS detector, and requests measurement activation to be performed by the UE to obtain one or more additional signal samples for the minority class. as well as An activation message is transmitted to the UE, the activation message causing the UE to perform one or more measurements during a configured observation window, wherein the activation message includes observation window configuration information.

11. The apparatus of claim 10, wherein the at least one memory and the computer program code are configured to cause the apparatus together with the at least one processor: Obtain a location estimate for at least one future location of the UE; Estimate the radio conditions at the at least one future location; The timing for transmitting the activation message is determined based on the estimated radio conditions at the at least one future location; as well as The activation message is transmitted at the specified time.

12. A method for a user equipment (UE) in a wireless communication system, the method comprising: Detect imbalances in the LOS training dataset used to train the line-of-sight LOS detector; Determine the minority class of the LOS training dataset associated with the detected imbalance; A request message is transmitted to the network element of the wireless communication system, the request message indicating the determined minority class and requesting measurement activation to be performed by the UE to obtain one or more additional signal samples for the minority class; as well as The UE receives an activation message from the network element, the activation message causing the UE to perform one or more measurements during a configured observation window, wherein the activation message includes observation window configuration information.

Images