Channel prediction in wireless communication systems
AI/ML models capturing physical laws and adapting to environments address the challenges of massive channel spaces in wireless communication, improving prediction accuracy and reliability.
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- HUAWEI TECH CO LTD
- Filing Date
- 2024-11-04
- Publication Date
- 2026-05-07
AI Technical Summary
The challenge of performing channel prediction in future wireless communication systems with massive channel spaces due to increased bandwidths and base station antenna counts is exacerbated by rapid channel changes, requiring large AI/ML models that are difficult to process, store, and maintain, with issues of noise, data loss, and environmental adaptability.
Implement AI/ML models that capture fundamental physical laws of channel dynamics, tolerate noise and data loss, and adapt to specific environments, integrating with communication systems for reliable channel prediction.
Enhances channel prediction accuracy and reliability by leveraging sparse representations and environmental adaptability, optimizing system performance and resource utilization.
Smart Images

Figure CN2024129643_07052026_PF_FP_ABST
Abstract
Description
CHANNEL PREDICTION IN WIRELESS COMMUNICATION SYSTEMSTECHNICAL FIELD
[0001] The application relates generally to wireless communications, and more specifically to channel prediction.BACKGROUND
[0002] Future generation wireless communication systems are expected to support increased performance such as high throughput. In some advanced technologies, bandwidths can jump from 20 megahertz (MHz) to 200 MHz, and base station antenna counts can rise from 64 to 1024 or even 4096. This creates a massive channel space that needs to be processed and measured. Performing channel prediction in such a large channel space can be difficult due to reasons such as a massive amount of data that needs to be processed, wireless communication channels changing faster than current systems, and a demand for real-time and low latency channel measurement and prediction. Consequently, there is a need for the development of new technologies that can address these challenges and enable full potential of the future generation wireless communication systems.SUMMARY
[0003] One or more implementations of the present application provide communication methods and communication apparatuses. The techniques described in the application can improve the performance of channel prediction in wireless communication systems.
[0004] According to a first aspect, a method is provided. The method includes transmitting, by a first device, a first message to a second device. The first message indicates a first channel prediction model for the second device to use. The first channel prediction model is selected based on at least one parameter associated with the second device. The method further includes transmitting, by the first device, a second message to the second device. The second message indicates a second channel prediction model. The second channel prediction model is selected based on a change to the at least one parameter associated with the second device.
[0005] With reference to the first aspect, in some implementations, the first channel prediction model and the second channel prediction model are selected from a plurality of channel prediction models.
[0006] With reference to the first aspect, in some implementations, the at least one parameter includes at least one of: an environmental condition; a resource availability; a power condition; or a prediction score.
[0007] With reference to the first aspect, in some implementations, transmitting the first message includes transmitting at least one model. The at least one model includes at least one of a first model or a second model.
[0008] With reference to the first aspect, in some implementations, each of the first channel prediction model and the second channel prediction model includes the first model and the second model.
[0009] With reference to the first aspect, in some implementations, the method further includes transmitting configuration information. The configuration information includes a format for a device state, a format for a representation of channel state information, and at least one model parameter associated with a channel prediction model.
[0010] With reference to the first aspect, in some implementations, the at least one model parameter includes at least one of a number of layers of the channel prediction model, dimensions associated with the layers of the channel prediction model, the number of weights in each layer, the total number of weights, or the model architecture.
[0011] With reference to the first aspect, in some implementations, transmitting the first message includes: receiving a capability message from the second device, where the capability message indicates a capability of the second device to execute the first channel prediction model; transmitting a resource configuration to the second device, where the resource configuration indicates a downlink resource for transmission of the first channel prediction model; and transmitting the first message to the second device in the downlink resource.
[0012] With reference to the first aspect, in some implementations, the first message includes an index associated with the first channel prediction model. The index is identified using a predefined mapping table.
[0013] With reference to the first aspect, in some implementations, the method further includes: receiving the device state and the representation of the channel state information from the second device; obtaining, using the first model, an embedded device state based on the device state; obtaining, using the second model, at least one channel state prediction for a time window based on the embedded device state and the representation of the channel state information; and obtaining, using a third model, a prediction quality evaluation based on the device state and the at least one channel state prediction.
[0014] With reference to the first aspect, in some implementations, the method further includes transmitting an update request to the second device based on the prediction quality evaluation. The update request includes a request to update at least one of: the device state; the representation of the channel state information; or the first channel prediction model.
[0015] With reference to the first aspect, in some implementations, the method further includes receiving an update request from the second device. The update request includes a request to update at least one of: the device state; or the representation of the channel state information.
[0016] According to a second aspect, a method is provided. The method includes receiving, at a first device, a first message from a second device. The first message indicates a first channel prediction model for the first device to use. The first channel prediction model is selected based on at least one parameter associated with the first device. The method further includes receiving, at the first device, a second message from the second device. The second message indicates a second channel prediction model. The second channel prediction model is selected based on a change to the at least one parameter associated with the first device.
[0017] With reference to the second aspect, in some implementations, the first channel prediction model and the second channel prediction model are selected from a plurality of channel prediction models.
[0018] With reference to the second aspect, in some implementations, the at least one parameter includes at least one of: an environmental condition; a resource availability; a power condition; or a prediction score.
[0019] With reference to the second aspect, in some implementations, receiving the first message includes receiving at least one model. The at least one model includes at least one of a first model and a second model.
[0020] With reference to the second aspect, in some implementations, the method further includes receiving configuration information. The configuration information includes a format for a device state, a format for a representation of channel state information, and at least one model parameter associated with a channel prediction model.
[0021] With reference to the second aspect, in some implementations, the at least one model parameter includes at least one of a number of layers of the channel prediction model, dimensions associated with the layers of the channel prediction model, or number of weights corresponding to at least one layer of the channel prediction model.
[0022] With reference to the second aspect, in some implementations, the method further includes: transmitting a capability message to the second device, where the capability message indicates a capability of the first device to execute the first channel prediction model; and receiving a resource configuration from the second device, where the resource configuration indicates a downlink resource for transmission of the first channel prediction model; and receiving the first message from the second device in the downlink resource.
[0023] With reference to the second aspect, in some implementations, the first message includes an index associated with the first channel prediction model. The index is identified using a predefined mapping table.
[0024] With reference to the second aspect, in some implementations, the method further includes: transmitting the device state and the representation of the channel state information to the second device; obtaining, using the first model, an embedded device state based on the device state; obtaining, using the second model, at least one channel state prediction for a time window based on the embedded device state and the representation of the channel state information; storing the at least one channel state prediction; and obtaining one or more prediction scores based on a difference between the at least one channel state prediction and a measured channel state information.
[0025] With reference to the second aspect, in some implementations, the method further includes transmitting an update request to the second device based on the one or more prediction scores. The update request includes a request to update at least one of: the device state; or the representation of the channel state information.
[0026] According to a third aspect, an apparatus is provided. The apparatus is configured to perform the method according to the first aspect or one or more implementations of the first aspect, or the second aspect or one or more implementations of the second aspect.
[0027] According to a fourth aspect, an apparatus is provided. The apparatus includes a transmitting unit configured to: transmit a first message to a second device, where the first message indicates a first channel prediction model for the second device to use, and where the first channel prediction model is selected based on at least one parameter associated with the second device; and transmit a second message to the second device, where the second message indicates a second channel prediction model, and where the second channel prediction model is selected based on a change to the at least one parameter associated with the second device.
[0028] According to a fifth aspect, an apparatus is provided. The apparatus includes: a receiving unit configured to: receive a first message from a second device, where the first message indicates a first channel prediction model for the apparatus to use, and where the first channel prediction model is selected based on at least one parameter associated with the apparatus; and receive a second message from the second device, where the second message indicates a second channel prediction model, and where the second channel prediction model is selected based on a change to the at least one parameter associated with the apparatus.
[0029] According to a sixth aspect, an apparatus is provided. The apparatus includes: one or more processors; and an interface circuit configured to: transmit a first message to a second device, where the first message indicates a first channel prediction model for the second device to use, and where the first channel prediction model is selected based on at least one parameter associated with the second device; and transmit a second message to the second device, where the second message indicates a second channel prediction model, and where the second channel prediction model is selected based on a change to the at least one parameter associated with the second device.
[0030] According to a seventh aspect, an apparatus is provided. The apparatus includes: one or more processors; and an interface circuit configured to: receive a first message from a second device, where the first message indicates a first channel prediction model for the apparatus to use, and where the first channel prediction model is selected based on at least one parameter associated with the apparatus; and receive a second message from the second device, where the second message indicates a second channel prediction model, and where the second channel prediction model is selected based on a change to the at least one parameter associated with the apparatus.
[0031] With reference to the sixth aspect or the seventh aspect, in some implementations, the interface circuit includes one or more transceivers.
[0032] According to an eighth aspect, an apparatus is provided. The apparatus includes one or more processors and one or more memories. The one or more memories store instructions which, when executed by the one or more processors, cause the apparatus to perform the method according to the first aspect or one or more implementations of the first aspect, or the second aspect or one or more implementations of the second aspect.
[0033] According to a ninth aspect, a communication system is provided. The communication system includes a first apparatus configured to perform the method according to the first aspect or one or more implementations of the first aspect. The communication system further includes a second apparatus configured to perform the method according to the second aspect or one or more implementations of the second aspect.
[0034] According to a tenth aspect, a non-transitory computer-readable storage medium is provided. The non-transitory computer-readable storage has instructions stored thereon which, when executed by an apparatus, cause the apparatus to perform the method according to the first aspect or one or more implementations of the first aspect, or the second aspect or one or more implementations of the second aspect.
[0035] According to an eleventh aspect, a computer program product storing instructions is provided. The computer program product stores instructions which, when executed, cause an apparatus to perform the method according to the first aspect or one or more implementations of the first aspect, or the second aspect or one or more implementations of the second aspect.BRIEF DESCRIPTION OF THE DRAWINGS
[0036] FIG. 1 illustrates a schematic illustration of an example communication system.
[0037] FIG. 2 illustrates another example communication system.
[0038] FIG. 3 illustrates an example of an apparatus wirelessly communicating with another apparatus in a communication system.
[0039] FIG. 4 illustrates an example apparatus.
[0040] FIG. 5 illustrates another example apparatus.
[0041] FIG. 6 illustrates an example channel prediction process.
[0042] FIG. 7 illustrates example signaling in a configuration response step of FIG. 6.
[0043] FIG. 8 illustrates example downlink (DL) and uplink (UL) transmissions in a DL and UL transmission step of FIG. 6.
[0044] FIG. 9A illustrates example prediction operations performed in a parallel prediction and evaluation step of FIG. 6.
[0045] FIG. 9B illustrates example evaluation operations performed by a base station (BS) in the parallel prediction and evaluation step of FIG. 6.
[0046] FIG. 9C illustrates example evaluation operations performed by a user equipment (UE) in the parallel prediction and evaluation step of FIG. 6.
[0047] FIG. 10 illustrates an example communication method.DETAILED DESCRIPTION
[0048] In current and future generations of communication systems, Artificial Intelligence (AI) and Machine Learning (ML) may be utilized to enhance wireless communication quality. Particularly, the use of AI or ML for channel prediction is gaining prominence. This approach can be referred to as the “one-side model. ” The term “one-side model” stems from the fact that model training only needs to be performed on one side, either by the User Equipment (UE) provider or the base station equipment provider. The trained model may then be broadcasted and transmitted, enabling devices on one side to gain the ability to predict channel changes, thereby improving communication quality or reducing signaling overhead.
[0049] For the purposes of this disclosure the term "model" is used to refer to any AI / ML techniques such as, but not limited to, deep neural networks or other machine learning methods like SVD / EVD decomposition or Graph-based approaches, that may be used for channel prediction. In general, they are collectively referred to as "models. " These models are data-driven, meaning they learn statistical characteristics from historical data to predict future channel states.
[0050] 6G presents a significant challenge that differs from 5.5G. In 5.5G, channel sizes are relatively small, with bandwidths ranging from 5 to 20 MHz and base stations with up to 64 antenna ports, while the terminal side typically has 4 to 8 antennas. This results in a relatively small overall channel size. However, 6G introduces advanced technologies like T-MIMO (Tera-bps MIMO) , leading to a substantial increase in overall throughput. Bandwidths can jump from 20 MHz to 200 MHz, and base station antenna counts can rise from 64 to 1024 or even 4096. This creates a massive channel space that needs to be processed and measured. Performing channel prediction in such a large channel space becomes extremely difficult due to the following reasons.
[0051] Using deep neural networks to learn such a large space presents the first problem: the input size of the neural network would be the sum of the dimensions of the entire 6G channel space. Multiplying 1024 antennas by a bandwidth of 200 MHz, and then by the number of antennas on the user equipment, results in a massive amount of data, far exceeding the scale of text, images, or even videos currently processed by large-scale deep neural networks.
[0052] This means that to process and predict such a channel space using AI / ML models, a model may be required that is significantly larger than those currently used for large language models (LLMs) . Moreover, wireless communication channels are constantly changing, especially in 6G, where the changes will be even faster than in 5G and 5.5G. This is because 6G wireless communication will operate at higher carrier frequencies, making it more sensitive to environmental changes.
[0053] This also implies a higher demand for channel measurement and prediction, with the lowest possible latency and real-time capabilities. On one hand, the channel space is vast, requiring very large AI / ML models; on the other hand, the response latency of the model must be extremely fast. Therefore, directly using AI / ML models for prediction in the original channel space is very difficult.
[0054] Furthermore, the cost of data sample collection, cleaning, and maintenance needs to be considered. Due to the complexity and massive scale of 6G channels, collecting and maintaining sufficient data samples itself requires significant resources and time. Data cleaning is equally complex, requiring data accuracy and consistency. Storing such large-scale data also demands substantial storage capacity and efficient storage management systems. These factors further increase the difficulty and cost of utilizing AI / ML for channel prediction.
[0055] Another issue is the randomness of the stochastic channel, making it impossible to determine the reliability of channel prediction. In other words, when predicting a channel state, determining if the prediction is accurate or reliable is a concerning aspect.
[0056] If the prediction is inaccurate, a new measurement needs to be initiated immediately. If it remains accurate, such as in cases with minimal environmental changes and user movement, the current prediction model can be used to predict the channel state as long as possible. This can significantly reduce signaling transmission and overhead.
[0057] Therefore, there is a need to determine the reliability of the channel state predicted by the AI / ML model. Accurately assessing reliability helps to decide when to use the prediction model and when to perform actual measurements, optimizing system performance and resource utilization.
[0058] This implies that regardless of the trained model's quality, when used for inference in real-world scenarios, it cannot be relied upon in isolation. It needs to be closely coupled with the existing communication system. The communication system needs to assist the model in better completing prediction tasks. This requires a series of control messages and signaling to fully exploit the advantages offered by AI and machine learning prediction models.
[0059] When dealing with ultra-large-scale channels, particularly those with a large number of antennas and ultra-wideband channels in 6G, it may be assumed sparsity despite their massive size. Sparsity signifies that the formation of channels is based on the propagation characteristics of electromagnetic waves. During propagation in the surrounding environment, electromagnetic waves experience reflection, refraction, and scattering on surfaces with different angles and materials, forming what are known as "rays. " These rays, through varying angles, attenuation, and delay, reach the receiver from the transmitter, constituting the channel.
[0060] While seemingly complex with various fading phenomena and frequency selectivity, channels are inherently sparse due to their dependence on physical principles. This is because the fundamental physical conditions determining channel characteristics are limited.
[0061] In mobile communications, the terminal user's position changes dynamically, described by the Doppler effect, which becomes more intricate in three-dimensional environments. Influenced by 3D angle and speed, the 3D Doppler effect, along with refraction and reflection in multipath propagation, further complicates matters. However, regardless of complexity, channel variations over time or space ultimately stem from user movement.
[0062] Underlying physical laws govern channels, accompanied by small-scale random factors like environmental changes, thermal noise fluctuations, or nearby interference variations. These numerous small-scale variations create a relatively small-scale random phenomenon. Consequently, channel characteristics can be divided into two parts: large-scale and small-scale variations. Large-scale variations are determined by physical laws, tending to be more deterministic, while small-scale variations exhibit statistical probability characteristics, tending to be more stochastic. Together, they constitute the essence of wireless channels.
[0063] However, it's worth noting that capturing the essential dynamics and underlying physical laws in complex systems and expressing them clearly remains a daunting task. AI / ML can be used and leveraged to address this challenge.
[0064] Instead of directly applying AI / ML to the channels, the approach focuses on capturing the most fundamental, sparse physical laws and their expressions behind the channels. Once the models successfully capture these laws and expressions, they can then proceed with channel prediction. This allows for more effective handling of dynamic changes in complex channels, improving prediction accuracy and reliability.
[0065] Through this method, AI / ML not only helps us understand channel complexities but also enhances channel prediction performance once key physical laws are grasped. This integration of physical laws and AI technology provides a more accurate and efficient solution for channel prediction in 6G.
[0066] A significant challenge in wireless communication is the imperfection of obtained data. Noise, packet loss, and other losses are common. Moreover, determining which inputs or input combinations lead to channel changes in such a complex system, especially in channel systems, often varies across scenarios.
[0067] Therefore, a sparse representation that tolerates noise and data loss may be needed. Some parameters might be missing or non-existent, some might be noisy, and others might have significant deviations, but not all parameters will exhibit these issues. This is the first challenge.
[0068] Secondly, the same parameter combination might not lead to identical channel changes at different times and locations. For example, certain parameters might impact the channel in one scenario but not have the same effect in another. Therefore, this sparse representation must be closely related to the specific environment of the base station and terminal.
[0069] These challenges highlight the difficulty of using AI / ML models. Models need one or more of the following capabilities:
[0070] 1. Tolerance to noise and data loss: Models must function effectively with imperfect data, handling noise, packet loss, and deviations.
[0071] 2. Environmental adaptability: Models need dynamic adjustments based on the specific environment of the base station and terminal to accurately capture channel change patterns.
[0072] In some implementations, achieving this may require AI / ML models with high flexibility and adaptability while extracting valid channel characteristics from imperfect data.
[0073] The details of one or more implementations of the subject matter of this present disclosure are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages of the subject matter will become apparent from the description, the drawings, and the claims.
[0074] FIG. 1 is a schematic illustration of an example communication system according to an implementation of the present disclosure, there is shown a communication system 100 that includes a radio access network (RAN) 120, one or more communication electronic devices (EDs) 110a, 110b, 110c, 110d, 110e, 110f, 110g, 110h, 110i, 110j (collectively referred to as 110) , a core network 130, a Public Switched Telephone Network (PSTN) 140, the Internet 150, and other networks 160 . The RAN 120 may include, but is not limited to, a future generation RAN, or a legacy RAN such as, but not limited to, 5th generation (5G) , 4th generation (4G) , 3rd generation (3G) or 2nd generation (2G) radio access network. The RAN 120 may be, for example, an Evolved Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access Network (E-UTRAN) , a NextGen RAN (NG RAN) , or some other type of RAN. Examples of RAN 120 based on the evolution of telecommunications standards include, but is not limited to, GSM (Global System for Mobile Communications) and CDMA (Code Division Multiple Access) for 2G, UMTS (Universal Mobile Telecommunications System) based on WCDMA (Wideband Code Division Multiple Access) and CDMA2000 for 3G, LTE (Long-Term Evolution) and WiMAX (Worldwide Interoperability for Microwave Access) for 4G, and NR (New Radio) for 5G. In some implementations, the RAN 120 may use any radio access technology (RAT) in the wireless interface between the one or more EDs 110 and the RAN 120. In some implementations, the term “radio access” may refer to the future generation air interface standards which may include both terrestrial networks (TNs) and non-terrestrial networks (NTNs) . These networks will be described in greater detail below in conjunction with various implementations. The one or more communication EDs 110 (also referred to as “user equipment” ) are configured to connect (e.g., communicatively couple) with each other or to one or more network nodes 170a, 170b (collectively referred to as 170) in the RAN 120. The core network (CN) 130 is a part of the communication system 100 and comprises network nodes (e.g., 170a, 170b) which provide support for the network features and telecommunication services. In some implementations, the CN 130 may be dependent on the RAT used in the communication system 100. In other implementations, the CN 130 may be access-agnostic, i.e., the CN 130 may be independent of the RAT used in the communication system 100. There are different types of CN 130, for different 3GPP system generations. For example, the CN 130 is the Evolved Packet Core (EPC) in 4G, also known as the Evolved Packet System (EPS) . In another example, the CN 130 is the 5G Core (5GC) which was developed as part of the 5G System (5GS) . The CN 130 also enables integration of different 3GPP and non-3GPP access types. In some implementations and referring to FIG. 1, the CN 130 also provides the interface towards external networks that may include the PSTN 140, the Internet 150, and other networks 160 in the communication system 100.
[0075] In general, the communication system 100 facilitates interaction between multiple wireless or wired elements. The communication system 100 may transmit different types of content, such as voice, data, video, and / or text, through different transmission methods such as, but not limited to, broadcast, multicast, groupcast, and unicast. Additionally, the communication system 100 operates by allocating and / or sharing resources, such as carrier spectrum bandwidth, among its constituent elements.
[0076] The communication system 100 may provide a wide range of communication services and applications including, but not limited to, Enhanced Mobile Broadband (eMBB) services, Ultra-Reliable Low-Latency Communication (URLLC) services, Massive Machine Type Communication (mMTC) services, Integrated Sensing And Communication (ISAC) , immersive communication, Ultra-massive Machine-Type Communication (uMTC) , hyper reliable and low-latency communication, ubiquitous connectivity, integrated AI and communication, and other services that can be provided by a future generation communication system. The communication system 100 may provide other services and applications such as, but not limited to, earth monitoring, remote sensing, passive sensing and positioning, navigation and tracking, autonomous delivery and mobility and the like.
[0077] The communication system 100 may include a terrestrial communication system (or network) and / or a non-terrestrial communication system (or network) . The communication system 100 may provide a high degree of availability and robustness through a joint operation of the terrestrial communication system and the non-terrestrial communication system. For example, integrating a non-terrestrial communication system (or components thereof) into a terrestrial communication system can result in a heterogeneous network comprising multiple layers. The heterogeneous network may achieve better overall performance through efficient multi-link joint operation, more flexible functionality sharing, and faster physical layer link switching between terrestrial networks and non-terrestrial networks. The terrestrial communication system and the non-terrestrial communication system could be considered as sub-systems of the communication system 100.
[0078] FIG. 2 illustrates another example communication system 100 according to an implementation of the present disclosure. The communication system 100 includes EDs 110a, 110b, 110c, 110d (collectively referred to as ED 110) , RANs 120a, 120b, one or more CNs 130, a PSTN 140, the Internet 150, and other networks 160. Additionally, the communication system 100 may also include a non-terrestrial network (NTN) 120c. The RANs 120a and120b may include network nodes 170a and 170b respectively. Examples of network nodes 170a, 170b include base stations, which can be generally referred to as terrestrial network (TN) devices or terrestrial transmit and receive points (T-TRPs) 170a and 170b (collectively referred to as 170) . In this context, the terms "TRP" and "base station" are used interchangeably unless otherwise specified. For simplicity, this disclosure primarily refers to network nodes as base stations; however, unless explicitly stated otherwise, references to TRP are considered non-limiting and interchangeable. The T-TRPs 170a, 170b may be base stations mounted on a building or tower. In one implementation, the NTN 120c includes a RAN node such as a base station 172, which may be generally referred to as an NTN device, a non-terrestrial node, a non-terrestrial network device, a non-terrestrial base station, or a non-terrestrial transmit and receive point (NT-TRP) 172.
[0079] In some implementations, the NT-TRP 172 is not attached to the ground, for example, as in the case of an airborne base station. An airborne base station may be implemented using communication equipment supported or carried by a flying device. For example, a flying device may include, but is not limited to, an airborne platform (such as a blimp or an airship) , balloon, drone (such as quadcopter) , and other types of aerial vehicles. In some implementations, an airborne base station may be supported or carried by an unmanned aerial system (UAS) or an unmanned aerial vehicle (UAV) , such as a drone. An airborne base station may be a moveable or mobile base station that can be flexibly deployed in different locations to meet network demand. A satellite base station is another example of a non-terrestrial base station. A satellite base station may be implemented using communication equipment supported or carried by a satellite. A satellite base station may also be referred to as an orbiting base station. High altitude platforms are yet another example of non-terrestrial base stations, including international mobile telecommunication base stations.
[0080] As referred to herein, and unless specified otherwise, a “TRP” may also refer to a T-TRP or an NT-TRP, a “T-TRP” may also refer to a “TN TRP” , and an “NT-TRP” may also refer to an “NTN TRP” . The NTN 120c may be considered a RAN, sharing operational aspects with RANs 120a, 120b. The NTN 120c may include at least one NTN device and at least one corresponding terrestrial network device. The at least one NTN device may function as a transport layer device and the at least one corresponding terrestrial network device may function as a RAN node, communicating with the ED 110 via the NTN device. Additionally, there may be an NTN gateway on the ground (referred to as a terrestrial network device) that also functions as a transport layer device facilitating communication with both the NTN device and the RAN node. The RAN node may communicate with the ED 110 via the NTN device and the NTN gateway. In some implementations, the NTN gateway and the RAN node may be located within the same device.
[0081] A base station 170 (also referred to as a TRP as stated above) is a network element within a radio access network responsible for radio transmission and reception in one or more cells to or from the ED (such as a user equipment) . In different implementations, the base station 170 may also be known as a base transceiver station (BTS) , a radio base station, a network node, a network device, a device on the network side, a transmit / receive node, a Node B, an evolved NodeB (eNodeB or eNB) , a Home eNodeB, a next Generation NodeB (gNB) , a transmission point (TP) , a site controller, an access point (AP) , a wireless router, a relay station, a terrestrial node, a terrestrial network device, a terrestrial base station, a non-terrestrial node, a non-terrestrial network device, a non-terrestrial base station, and a positioning node, among other possibilities. The base station 170 may be a macro base station (BS) , a pico BS, a relay node, a donor node, or combinations thereof. When the base station 170 performs (or is configured to perform) a method described herein, it may be interpreted as the base station itself, one or more modules (or units) in the base station, a circuit or chip, or a combination thereof, performing the method. For example, the circuit or chip may include a modem chip, also referred to as a baseband chip, a system on chip (SoC) including a modem core, a system in package (SIP) chip, and the like, and may be responsible for one or more communication functions within the base station.
[0082] The EDs 110a-110d and TRPs 170a-170b, 172 are examples of communication equipment configured to implement some or all of the operations and / or implementations described herein. The T-TRP 170a forms part of the RAN 120a, which may include other TRPs, and / or other devices. Also, the TRP 170b forms part of the RAN 120b, which may include other TRPs, and / or devices. Each TRP 170a, 170b may transmit and / or receive wireless signals within a particular geographic region or area, sometimes referred to as a “cell” or a “coverage area” . The TRPs 170a-170b may be responsible for allocating and / or configuring resources and transmission and / or reception in a set of cell (s) . A cell is a radio network object that can be uniquely identified by a cell identification that is broadcasted over a geographical region or area from base stations associated with the cell. A cell can work in either FDD or TDD mode. A cell may be further divided into cell sectors, and a base station 170a-170b may, for example, employ one or more transceivers to provide services to one or more sectors. Some implementations may include pico or femto cells if supported by the radio access technology. In some implementations, one or more transceivers could be used for each cell, such as with Multiple-Input Multiple-Output (MIMO) technology. The number of RANs 120a-120b shown is merely an example. Any number of RANs may be contemplated when designing the communication system 100.
[0083] A base station may be a single element, as shown in the figures, or multiple elements distributed throughout the corresponding RAN, or otherwise configured. In some implementations, a plurality of RAN nodes coordinates to assist the ED 110 in implementing radio access, and different RAN nodes separately implement and handle different functions of the base station. For example, the RAN node may be a central unit (CU) , a distributed unit (DU) , a CU-control plane (CP) , a CU-user plane (UP) , or a radio unit (RU) etc. The CU and the DU may be separately deployed, or included within the same element (i.e., a baseband unit (BBU) ) . The RU may be included in a radio frequency device or a radio frequency unit (i.e., a remote radio unit (RRU) , an active antenna unit (AAU) , or a remote radio head (RRH) ) . In different systems, the CU (or the CU-CP and the CU-UP) , the DU, or the RU may be known by different names, but their functions are understood by a person skilled in the art. For example, in an open radio access network (ORAN) system, a CU may be referred to as an open CU (O-CU) , a DU may be referred to as an open DU (O-DU) , and a CU-CP may be referred to as an open CU-CP (O-CU-CP) . The CU-UP may also be referred to as an open CU-UP (O-CU-UP) , and the RU may also be referred to as an open RU (O-RU) . Any one of the CU (or the CU-CP, the CU-UP) , the DU, and the RU may be implemented using a software module, a hardware module, or a combination of a software module and a hardware module.
[0084] Furthermore, communication between different devices / apparatuses in various implementations of this disclosure may refer to direct communication (that is, without the need of forwarding by another device / apparatus) or may refer to communication (s) between different devices / apparatuses via another device / apparatus (that is, requiring forwarding by another device / apparatus) . Alternatively, such communication (s) may involve one functional unit inside a device / apparatus using another functional unit within the device / apparatus to communicate with another device / apparatus. In other words, phrases such as "sending (or transmitting) information to. . . (an ED or a base station) " in this disclosure may be understood as a destination endpoint of the information being an ED or a base station, including, sending / transmitting information directly or indirectly to an ED or a base station. Similarly, phrases like "receiving information from. . . (an ED or a base station) " may be understood as a source endpoint of the information being an ED or a base station, including directly or indirectly receiving information from an ED or a base station. Between the source endpoint that sends the information and the destination endpoint, necessary processing such as, but not limited to, format conversion, digital-to-analog conversion, amplification, and filtering may be performed on the information. However, the destination endpoint may understand valid information from the source endpoint. A similar understanding applies to other descriptions in this disclosure without reiterating details already described. In the present disclosure, the terms "send" and "transmit" may be used interchangeably in different implementations of this disclosure.
[0085] The ED 110 is used to connect people, objects, machines, and other entities. The ED 110 may be widely used in various scenarios including, but not limited to, cellular communications, device-to-device (D2D) , vehicle to everything (V2X) , peer-to-peer (P2P) , machine-to-machine (M2M) , MTC, internet of things (IoT) , virtual reality (VR) , augmented reality (AR) , mixed reality (MR) , metaverse, digital twin, industrial control, self-driving, remote medical, smart grid, smart furniture, smart office, smart wearable, smart transportation, smart city, drones, robots, remote sensing, passive sensing, positioning, navigation and tracking, and autonomous delivery and mobility.
[0086] Each ED 110 represents any suitable end user device for wireless operation and may include such devices (or may be referred to as, but not limited to) a user equipment (UE) or a user device or a terminal device, a wireless transmit / receive unit (WTRU) , a mobile station, a fixed or mobile subscriber unit, a cellular telephone, a station (STA) , an MTC device, a personal digital assistant (PDA) , a smartphone, a laptop, a computer, a tablet, a wireless sensor, a consumer electronics device, a smart book, a vehicle, a car, a truck, a bus, a train, or an IoT device, wearable devices (such as a watch, a pair of glasses, head mounted equipment, etc. ) , an industrial device, or an apparatus (such as a module, modem, or chip) in the foregoing devices, among other possibilities. Future generation EDs 110 may be referred to by other terms. When an ED 110 performs (or is configured to perform) a method described herein, it may be interpreted as the ED itself, one or more modules (or units) in the ED, a circuit or chip, or a combination thereof, performing the method. For example, the circuit or chip may include a modem chip, also referred to as a baseband chip, a system on chip (SoC) including a modem core, or a system in package (SIP) chip, and the like, and may be responsible for one or more communication functions in the ED.
[0087] Each ED 110 connected to TRPs 170a-170b, and / or TRPs 172 can be dynamically or semi-statically turned-on (i.e., established, activated, or enabled) , turned-off (i.e., released, deactivated, or disabled) and / or configured in response to one of more of: connection availability and connection necessity.
[0088] Any ED 110 may be alternatively or additionally configured to interface, access, or communicate with any of the TRPs 170a, 170b and 172, the Internet 150, the CN 130, the PSTN 140, the other networks 160, or any combination thereof. In some examples, the ED 110a may communicate an uplink (UL) and / or downlink (DL) transmission over a terrestrial air interface 190a with station-TRP 170a. In some examples, the EDs 110a, 110b, 110c, and 110d may also communicate directly with one another via one or more sidelink (SL) air interfaces 190b. In some examples, the EDs 110a, 110d may communicate using an UL and / or DL transmission over a non-terrestrial air interface 190c with NT-TRP 172.
[0089] An air interface (such as, for example, 190a, 190b, 190c) generally includes a number of components and associated parameters that collectively specify how a transmission is to be sent and / or received over a wireless communications link between two or more communicating devices such as EDs and base station (s) . For example, an air interface may include one or more components defining the waveform (s) , frame structure (s) , multiple access scheme (s) , protocol (s) , coding scheme (s) and / or modulation scheme (s) for conveying information (such as, data) over a wireless communications link. The air interfaces 190a and 190b may use similar communication technology, that may include any suitable radio access technology.
[0090] The non-terrestrial air interface 190c can enable communication between the EDs 110a, 110d and one or more NT-TRPs 172 via a wireless link or simply a link. In some examples, the link is a dedicated connection for unicast transmission, a connection for broadcast transmission, or a connection between a group of EDs 110 and one or more NT-TRPs 172 for multicast transmission.
[0091] The TRPs 170a-170b, 172 may communicate with one another over one or more air interfaces 190e, 190f using wireless communication links (such as radio frequency (RF) , microwave, infrared (IR) , etc. ) or wired communication links. The air interfaces 190e, 190f may utilize any suitable radio access technology, and may be substantially similar to the air interfaces 190a, 190c over which the EDs 110a-110d communicate with one or more of the TRP 170a-170b, 172 or they may be substantially different. For example, the communication system 100 may implement one or more channel access methods, such as Time Division Multiple Access (TDMA) , Frequency Division Multiple Access (FDMA) , Code Division Multiple Access (CDMA) , Single Carrier Frequency Division Multiple Access (SC-FDMA) , Low Density Signature Multicarrier Code Division Multiple Access (LDS-MC-CDMA) , Non-Orthogonal Multiple Access (NOMA) , Pattern Division Multiple Access (PDMA) , Lattice Partition Multiple Access (LPMA) , Resource Spread Multiple Access (RSMA) , and Sparse Code Multiple Access (SCMA) .
[0092] The RANs 120a and 120b are in communication with the CN 130 to provide the EDs 110a 110b, and 110c with various services such as voice, data, multimedia, and other services. The RANs 120a and 120b and / or the CN 130 may be in direct or indirect communication with one or more other RANs (not shown) , which may or may not be directly served by the CN 130, and may employ different radio access technologies from RAN 120a and / or RAN 120b. The CN 130 may also serve as a gateway access between (i) the RANs 120a and 120b and / or the EDs 110a 110b, and 110c, and (ii) other networks (such as the PSTN 140, the Internet 150, and the other networks 160) . In addition, some or all of the EDs 110a 110b, and 110c may include functionality for communicating with different wireless networks over different wireless links using different wireless technologies and / or protocols. For example, the EDs 110a 110b, and 110c communicate using different cellular communications protocols, such as, but not limited to, a Global System for Mobile Communications (GSM) protocol, a code-division multiple access (CDMA) network protocol, a Push-to-Talk (PTT) protocol, a PTT over Cellular (POC) protocol, a Universal Mobile Telecommunications System (UMTS) protocol, a 3GPP Long Term Evolution (LTE) protocol, a fifth generation (5G) protocol, a New Radio (NR) protocol, and the like. Instead of wireless communication (or in addition thereto) , the EDs 110a 110b, and 110c may communicate using wired communication channels to a service provider or switch (not shown) , and / or to the Internet 150. The PSTN 140 may include circuit switched telephone networks for providing plain old telephone service (POTS) . The Internet 150 may include a network of computers and subnets (intranets) or both, and incorporate protocols, such as internet protocol (IP) , transmission control protocol (TCP) , user datagram protocol (UDP) . EDs 110a 110b, and 110c may be multimode devices capable of operation according to multiple radio access technologies, and may incorporate one or more transceivers necessary to support such technologies and / or functions.
[0093] In addition, the communication system 100 may comprise a sensing agent (not shown) to manage the sensed data from ED 110 and / or any one of TRPs 170a, 170b, 172. In one implementation, the sensing agent may be part of any one of TRPs 170a, 170b, 172. In another implementation, the sensing agent is a separate node that can communicate with the CN 130 and / or the RAN 120 (such as any one of TRPs 170a, 170b, 172) .
[0094] FIG. 3 is a schematic illustration showing an apparatus 310 wirelessly communicating with another apparatus 320 within a communication system (e.g., the communication system 100) according to an implementation of the present disclosure. The apparatus 310 may be an electronic device (such as ED 110) . The apparatus 320 may be a network node (e.g., the network node 170) such as T-TRP 170 or an NT-TRP 172. Although only one apparatus 310, and one apparatus 320 are shown in the figure, the number of apparatus 310 and / or number of apparatus 320 can vary, potentially including one or more of each. For example, a single ED 110 may be served by a single T-TRP 170 (or a single NT-TRP 172) , or by multiple T-TRPs 170 (or multiple NT-TRPs 172) . Similarly, a single ED 110 may be served by one or more T-TRPs 170 and one or more NT-TRPs 172. Similarly, a single T-TRP 170 (or a single NT-TRP 172) may serve one or more EDs 110.
[0095] The apparatus 310 may include one or more processors 210. For clarity and to avoid overcrowding the illustration, only a single processor 210 is illustrated. The apparatus 310 may further include a transmitter 201 and a receiver 203 coupled to one or more antennas 204. For clarity, only a single antenna 204 is illustrated. One, some, or all of the antennas 204 may alternatively be panels. In some implementations, the transmitter 201 and the receiver 203 are separate from each other. In other implementations, the transmitter 201 and the receiver 203 may be integrated into a single unit, for example, as a transceiver. The transceiver is configured to modulate data or other content for transmission by the one or more antennas 204 or a network interface controller (NIC) . The transceiver may also be configured to demodulate data or other content received by the one or more antennas 204. A transceiver may include any suitable structure for generating signals for wireless or wired transmission and / or for processing signals received through wireless or wired communication. Each antenna 204 includes any suitable structure for transmitting and / or receiving wireless or wired signals. The apparatus 310 may include a memory 208. In some implementations, the apparatus 310 may include multiple memories 208. Only a single transmitter 201, receiver 203, processor 210, memory 208, and antenna 204 is illustrated for simplicity, but the apparatus 310 may include one or more other components. In some implementations of the present disclosure, the transceiver (or transmitter 201 and / or receiver 203) may be viewed as an interface circuit.
[0096] The memory 208 is configured to store instructions used to perform operations described herein. The memory 208 may also be configured to store data that is used, generated, or collected by the apparatus 310. For example, the memory 208 can store software instructions or modules configured to implement some or all of the functionalities and / or operations described herein and that which are executed by the one or more processors 210.
[0097] The apparatus 310 may further include one or more input / output devices (not shown) or interfaces. The input / output devices or interfaces facilitate interaction with a user or other devices in the network. Each input / output device or interface includes suitable components for facilitating transmission of information to a user and reception of information from a user, and for various network interface communications. Such components may include, but are not limited to, a speaker, microphone, keypad, keyboard, display, touch screen, and the like.
[0098] The processor 210 may be configured to perform (or control the apparatus 310 to perform) operations (or methods) described herein as being performed by the apparatus 310. For example, the processor 210 performs or controls the apparatus 310 to perform the operations of: a) receiving one or more transport blocks (TBs) , b) using a resource for decoding at least one of the received TBs, c) releasing the resource for decoding another of the received TBs, and / or d) receiving configuration information configuring a resource. Specifically, the operations may include tasks related to: preparing a transmission for UL transmission to the apparatus 320, processing DL transmissions received from the apparatus 320, and handling SL transmission to and from another apparatus 310. Processing operations related to preparing a transmission for UL transmission may include operations such as, but not limited to, encoding, modulating, transmit beamforming, and generating symbols for transmission. Processing operations related to processing DL transmissions may include operations such as, but not limited to, receive beamforming, demodulating and decoding received symbols. Processing operations related to processing SL transmissions may include operations such as, but not limited to, transmit / receive beamforming, modulating / demodulating and encoding / decoding symbols. Depending upon the implementation, a DL transmission may be received by the receiver 203, possibly using receive beamforming, and the processor 210 may extract signaling from the DL transmission (such as by detecting and / or decoding the signaling) . An example of signaling may be a reference signal transmitted by the apparatus 320. In some implementations, the processor 210 implements the transmit beamforming and / or the receive beamforming based on the indication of beam direction, such as beam angle information (BAI) , received from the apparatus 320. In some implementations, the processor 210 may be configured to perform operations relating to network access (such as initial access) and / or downlink synchronization, which includes operations for detecting a synchronization sequence, decoding and obtaining the system information, and the like. In some implementations, the processor 210 may perform channel estimation, such as using a reference signal received from the apparatus 320.
[0099] Although not illustrated, in some implementations, the processor 210 may either be a part of the transmitter 201 or a part of the receiver 203 or a part of both the transmitter 201 and the receiver 203. Although not illustrated, in some implementations, the memory 208 may be a part of the processor 210.
[0100] The processor 210, along with the processing components of the transmitter 201 and the receiver 203 may each be implemented by one or more processors that may the same or different. These processors are configured to execute instructions stored in a memory (such as in the memory 208) .
[0101] The apparatus 320 includes one or more processors 260 (only one processor 260 is illustrated) . The apparatus 320 may further include one or more transmitters 252 and one or more receivers 254 coupled to one or more antennas 256. Only a single antenna 256 is illustrated to avoid clutter in the illustration. One, some, or all of the antennas 256 may alternatively be panels. In some implementations, the transmitter 252 and the receiver 254 are separate from each other. In other implementations, the transmitter 252 and the receiver 254 may be integrated into a single unit such as, for example, as a transceiver. The apparatus 320 may further include a memory 258. In some implementations, the apparatus 320 may include multiple memories 258. The apparatus 320 may further include a scheduler 253. Only a single transmitter 252, receiver 254, processor 260, memory 258, antenna 256 and scheduler 253 are illustrated for simplicity, however the apparatus 320 may include one or more other components. In the present disclosure, in some implementations, the transceiver (or transmitter 252 and / or receiver 254) may be viewed as an interface circuit.
[0102] In some implementations, various components of the apparatus 320 may be distributed. For example, some of the modules of the apparatus 320 may be located remotely from the equipment housing the antennas 256 for the apparatus 320 (and therefore can also be viewed as one or more nodes) . These modules, which can be considered as one or more nodes, may be coupled to the equipment that houses the antennas 256 over a communication link (not shown) , sometimes referred to as front haul, such as the Common Public Radio Interface (CPRI) . Therefore, in some implementations, the term apparatus 320 may also refer to network-side nodes that perform processing operations such as, but not limited to, determining the location of the apparatus 310, resource allocation (scheduling) , message generation, and encoding / decoding, and that which are not necessarily part of the equipment that houses the antennas 256 of the apparatus 320. The nodes may also be coupled to other apparatuses 320. In some implementations, the apparatus 320 may actually be a plurality of nodes that are operating together to serve the apparatus 310, such as through the use of coordinated multipoint transmissions, or through the use of an ORAN system as described above in the disclosure.
[0103] The processor 260 is configured to perform operations including those related to: preparing a transmission for DL transmission to the apparatus 310, processing a UL transmission received from the apparatus 310, preparing a transmission for backhaul transmission to another apparatus 320, and processing a transmission received over backhaul from another apparatus 320. Processing operations related to preparing a transmission for DL or backhaul transmission may include operations such as, but not limited to, encoding, modulating, precoding (such as MIMO precoding) , transmit beamforming, and generating symbols for transmission. Processing operations related to processing received transmissions in the UL or over backhaul may include operations such as, but not limited to, receive beamforming, demodulating received symbols, and decoding received symbols. The processor 260 may also be configured to perform operations relating to network access (such as initial access) and / or DL synchronization, such as generating the content of synchronization signal blocks (SSBs) , generating the system information, and the like. In some implementations, the processor 260 is further configured to generate an indication of beam direction, such as BAI, which may be scheduled for transmission by the scheduler 253 which will be described below. In some implementations, the processor 260 implements the transmit beamforming and / or receive beamforming based on beam direction information (such as BAI) received from another apparatus 320. The processor 260 is configured to perform other network side processing operations described herein, such as, but not limited to, determining the location of the apparatus 310, determining where to deploy another apparatus 320, and the like. In some implementations, the processor 260 may generate signaling data, to configure one or more parameters of the apparatus 310 and / or one or more parameters of another apparatus 320. Any signaling data generated by the processor 260 is sent by the transmitter 252. In some implementations, the apparatus 320 implements physical layer processing. In some implementations, the apparatus 320 may perform higher layer functions such as those at the Medium Access Control (MAC) or Radio Link Control (RLC) layers in addition to physical layer processing. In the apparatus 320, the scheduler 253 may be coupled to the processor 260 or integrated within the processor 260. In some implementations, the scheduler 253 may be integrated within the apparatus 320 or may be operated separately from the apparatus 320. The scheduler 253 may schedule UL, DL, SL, and / or backhaul transmissions, including issuing scheduling grants and / or configuring scheduling-free (such as “configured grant” ) resources.
[0104] The apparatus 320 may further include a memory 258 that is configured to store instructions for performing the operations described herein. The memory 258 may also store data that is used, generated, or collected by the apparatus 320. For example, the memory 258 can store software instructions or modules configured to implement some or all of the functionalities and / or implementations described herein and that which are executed by the processor 260.
[0105] Although not illustrated, the processor 260 may be implemented as part of the transmitter 252 and / or a part of the receiver 254. Although not illustrated, in some implementations, the processor 260 may implement the scheduler 253 and the memory 258 may be implemented as part of the processor 260.
[0106] The processor 260, the scheduler 253, the processing components of the transmitter 252, and the processing components of the receiver 254 may each be implemented by the same or different processors that are configured to execute instructions stored in a memory, such as in the memory 258.
[0107] The apparatus 320 and / or the apparatus 310 may include other components, not shown or described herein for the sake of clarity.
[0108] Note that the term “signaling” , as used herein, may alternatively be referred to as control signaling, control message, control information, or message for simplicity. Signaling between a base station (such as the TRP 170a. 170b, 172) and a UE or sensing device (such as ED 110) , or signaling between a different UE or sensing device (such as between ED 110a and ED 110b) may be carried in physical layer signaling (also called as dynamic signaling) , which is transmitted in a physical layer control channel. For DL, the physical layer signaling may be known as downlink control information (DCI) which is transmitted in a physical downlink control channel (PDCCH) . For UL, the physical layer signaling may be known as uplink control information (UCI) which is transmitted in a physical uplink control channel (PUCCH) . For SL, signaling between different UEs or sensing devices (such as between ED 110a and ED 110b) may be known as SL control information (SCI) which is transmitted in a physical sidelink control channel (PSCCH) . Signaling may be carried in a higher layer (such as higher than physical layer) signaling, which is transmitted in a physical layer data channel, such as in a physical downlink shared channel (PDSCH) for downlink signaling, in a physical uplink shared channel (PUSCH) for uplink signaling, and in a physical sidelink shared channel (PSSCH) for SL signaling. Higher layer signaling may also be called static signaling, or semi-static signaling. The higher layer signaling may include radio resource control (RRC) protocol signaling or media access control -control element (MAC-CE) signaling. Signaling may be included in a combination of physical layer signaling and higher layer signaling.
[0109] It should be noted that in the present disclosure, “information” , when different from “message” , may be carried within a single message, or may be carried in multiple separate messages.
[0110] FIG. 4 illustrates an example apparatus 410 according to an implementation of the present disclosure. The apparatus 410 may be a communication device or an apparatus implemented in a communication device such as the ED 110 or the TRPs 170a, 170b, 172. For example, the apparatus 410 implemented in an ED may be an integrated circuit, which in some instances may be referred to as a chip, a modem, a modem chip, a baseband chip, or a baseband processor. In some implementations, one or more integrated circuits can be packaged into a system-on-chip, a system-in-package, or a multi-chip module. The apparatus 410 can include one or more integrated circuits and other discrete components. In some implementations, the apparatus 410 may be a module within the ED 110, or within the apparatus 310. In some implementations, the apparatus 410 may be a module within one of the TRPs 170a, 170b, 172, or the apparatus 320.
[0111] In an example, the apparatus 410 may include one or more processors / processor cores 411, and an interface circuit 412. The apparatus 410 may further include a memory 413. The one or more processors 411 are configured to process signals and execute one or more communication protocols. The memory 413 is configured to store at least a part of the corresponding computer program instructions and / or data. In an example, the one or more processors 411 execute the computer program instructions stored in the memory 413 to implement related operations (for example, inputting, outputting, receiving, and transmitting) in the method implementations disclosed herein. In some implementations, the memory 413 being configured to store the corresponding computer program instructions and / or data may mean that the memory 413 is configured to store all of the corresponding computer program instructions and / or data for execution by the one or more processors 411. In some implementations, the memory 413 being configured to store the corresponding computer program instructions and / or data may mean that the memory 413 is configured to store a part of the corresponding computer program instructions and / or data. For example, the part of the corresponding computer program instructions and / or data may include computer program instructions and / or data that need to be currently executed by the one or more processors 411. Thus, the memory 413 may store different parts of computer program instructions and / or data for a plurality of times for the one or more processors 411 to perform related operations in the method implementations disclosed herein. As a communication interface, the interface circuit 412 is configured to implement communication with another component. For example, the interface circuit 412 may communicate a signal with another apparatus or system, such as a radio frequency processing apparatus or another processor. The signal may include or carry information intended as a payload, such as user data, control information, etc. The signal may also include or carry information useful to a receiver, but not necessarily as a payload, such as a pilot signal or reference signal. Communicating the signal may include transmitting the signal to another component or device. Communicating the signal may additionally or alternatively include receiving the signal from another component or device. Transmitting the signal may include outputting the signal to a component or device that is directly or indirectly coupled to the interface circuit 412. Receiving the signal may include inputting or obtaining the signal from a component or device that is directly or indirectly coupled to the interface circuit 412. In some implementations, to reduce a load of the one or more processors, a baseband signal processing circuit 414 may be also disposed to implement processing of at least a part of baseband signals, including signal demodulation, modulation, encoding, decoding, or the like.
[0112] The apparatus 410 may be the processor 210 (or 260) within the apparatus 310 (or 320) , in some scenarios, or may be included within the processor 210 (or 260) within the apparatus 310 (or 320) in some scenarios. The apparatus 410 may be a baseband chip or may include a baseband chip. In some implementations, the apparatus 410 may be independently packaged into a chip. In some implementations, the apparatus 310 (or 320) includes different types of chips. The apparatus 410 may be packaged into a processor chip (for example, an SoC chip or an SIP chip) with the different types of chips. In some implementations, the apparatus 410 may be packaged into a chip with some or all of circuits of a radio frequency processing system that may further be included in the apparatus 310 (or 320) .
[0113] FIG. 5 illustrates an example apparatus 510 according to an implementation of the present disclosure. The apparatus 510 may include corresponding modules or units configured to implement methods and / or implementations described herein. In some implementations, the apparatus 510 includes a processing unit 512 and a communication unit 513. In some implementations, the apparatus 510 may further include a storage unit 511 configured to store apparatus program code (or instructions) and / or data. In some implementations, the apparatus 510 may further include an AI unit or an ML unit 514 configured to perform tasks related to processing, training, and executing machine learning models. The ML unit 514 can be coupled to one or more of the storage unit 511, the processing unit 512, and the communication unit 513.
[0114] The apparatus 510 may be an ED side apparatus, for example, an ED or a module in an ED, or a circuit or a chip responsible for a communication function in an ED. In some implementations, apparatus 510 may be the apparatus 310. The processing unit 512 may be the processor 210. The communication unit 513 may comprise a receiving unit and / or a transmitting unit. The receiving unit and / or the transmitting unit may be the transmitter 201 and / or the receiver 203 respectively. The storage unit 511 may be the memory 208.
[0115] The apparatus 510 may be a base station side apparatus, for example, a base station or a module in a base station, or a circuit or a chip responsible for a communication function in a base station. In some implementations, apparatus 510 may be apparatus 320. The processing unit 512 may be the processor 260 (the scheduler 253 may also be included) . The communication unit 513 may include a receiving unit and / or a transmitting unit. The receiving unit and / or the transmitting unit may be the transmitter 252 and / or the receiver 254 respectively. The storage unit 511 may be the memory 258.
[0116] In some implementations, when the apparatus 510 is an ED 110 or a module in an ED 110, a function of the apparatus 510 may be implemented by one or more processors. Specifically, the processor may include a modem chip, or a system on chip (SoC) chip or an SIP chip that includes a modem core. A function of the communication unit 513 may be implemented by a transceiver circuit.
[0117] In some implementations, when the apparatus 510 is a circuit or a chip that is responsible for a communication function in an ED 110, such as a modem chip, a system on chip (SoC) chip or an SIP chip that includes a modem core, a function of the processing unit 512 may be implemented by a circuit system within the chip which includes one or more processors. A function of the communication unit 513 may be implemented by an interface circuit or a data transceiver circuit on the chip.
[0118] It may be understood that the units in the apparatus 510 may be logical or functional. Each function may correspond to one functional unit, or two or more functions may be integrated into a single functional unit. In actual implementation, all or some of the units may be integrated into a single physical entity, or may be distributed across different physical entities. In addition, the functional units may be implemented in the form of hardware, software, or a combination of hardware and software. Whether a function is implemented in the form of hardware or software depends on particular applications and design constraint conditions of the technical solutions. A person skilled in the art may use different methods to implement the described functions for specific applications, but it should not be considered that the implementation goes beyond the scope of this disclosure.
[0119] In an example, a functional unit in any one of the apparatuses may be configured as one or more integrated circuits for implementing the methods disclosed herein, for example, as one or more application-specific integrated circuits (application-specific integrated circuits, ASICs) , one or more central processing units (CPUs) , one or more microprocessors or microprocessor units (MPUs) , one or more microcontrollers or microcontroller units (MCUs) , one or more digital signal processors (DSPs) , one or more field programmable gate arrays (FPGAs) , or a combination of these.
[0120] In an example, the storage unit 511 may include a random access memory, a flash memory, a read-only memory, a programmable read-only memory, an electrically erasable programmable memory, and / or a register.
[0121] A processor may be referred to as a processor system, an application processor, a baseband processor, a processor circuit, or a processor core. The processor may include one or a combination of one or more central processing units (CPUs) , one or more digital signal processors (DSPs) , one or more microprocessors (microprocessor units, MPUs) , one or more microcontrollers (microcontroller units, MCUs) , one or more graphics processing units (GPUs) , one or more field programmable gate arrays (FPGAs) , one or more artificial intelligence processors (AI processors) , or one or more neural network processing units (NPUs) .
[0122] Memory or a storage unit may include one or more of the following storage media: a random access memory (RAM) , a static random access memory (static RAM, SRAM) , a dynamic random access memory (dynamic RAM, DRAM) , a phase-change memory (PCM) , a resistive random access memory (resistive RAM, ReRAM) , a magnetoresistive random access memory (magnetoresistive RAM, MRAM) , a ferroelectric random access memory (ferroelectric RAM, FRAM) , a cache, a register, a read-only memory (ROM) , a flash memory (flash memory) , an erasable programmable read-only memory (erasable programmable ROM, EPROM) , a hard disk, and the like. In an example, computer program instructions used to execute implementations may be stored in a non-volatile memory, for example, at least a part of a memory or storage unit (for example, one or more of a ROM, a flash memory, an EPROM, or a hard disk) . When a terminal runs, a part or all of corresponding computer program instructions may be loaded to a memory that has a higher transmission speed with the processor, for example, at least a part of a memory or a storage unit (for example, one or more of a RAM, an SRAM, a DRAM, a PCM, a RERAM, an MRAM, a FRAM, a cache, or a register) , so that the processor executes the computer program instructions to perform the steps in the method implementations disclosed herein.
[0123] In some implementations, a transformer is used in the described techniques. Transformers can refer to a novel deep neural network architecture that inherits and extends the advantages of Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) , including Long Short-Term Memory networks (LSTM) . At their core is the Attention mechanism, allowing them to find deeper and longer-range patterns in data.
[0124] The Attention mechanism can combine and expand upon the strengths of CNNs, RNNs, and LSTMs. The Attention mechanism can enable finding hidden, more complex, and persistent patterns within data sequences. Through multi-layered Attention, the model identifies higher-order pattern relationships within data or signal sequences. When the input is signal data instead of text or images, and these signals have a physical basis, theoretically, sufficiently deep and numerous Attention mechanisms can capture the dynamic changes in the signals, including first-order, second-order, and even higher-dimensional relationships.
[0125] A transformer architecture can include two main components: an Encoder and a Decoder.
[0126] The Encoder functions similarly to a fill-in-the-blank exercise. The input data sequence consists of a set of sub-fields of information, whose ordering and positions imply logic interrelationships. The Encoder uses the Attention mechanism to learn long-range relational patterns across and within these sub-fields. Through continuous reinforcement of the deep neural network, the Encoder's Attention mechanism finds and learns these relational patterns. During inference, if noise or missing data affects a sub-field, the Encoder utilizes information from other sub-fields to compensate for these losses or noisy parts, forming an output sequence.
[0127] The Decoder primarily handles prediction or generation. In the encoder-decoder architecture, the decoder takes the output of the encoder as input, and generates the next symbol step-by-step based on this sequence. Usually, this process runs for either a maximum number of symbols, or until a termination symbol is produced. The Decoder employs both a self-attention mechanism, where it uses past decoded symbols to predict future symbols, and a cross-attention mechanism, where it uses the encoder’s output as side-information to make these predictions.
[0128] In some instances, the Decoder in transformers is becoming increasingly crucial as a key component in large language models. Consequently, it has spawned numerous variations for generating more information. The initial transformer architecture's Decoder merely received the completed sequence from the Encoder. However, modern Decoders can accept additional inputs to aid prediction, including context information and external knowledge bases, enhancing prediction accuracy and comprehensiveness.
[0129] In multiple machine learning problems, a common approach is to have two networks. One of the networks can be denoted as a “primary” network. The primary network is responsible for performing the task associated with the problem, for example, prediction, generation or selecting actions. The second network can be denoted as an “auxiliary” network, is responsible for evaluating the performance of the primary network.
[0130] For example, in Generative Adversarial Networks (GANs) , the primary network, called the generator network, is tasked with generating images, while the auxiliary network, called the discriminator network, is tasked with classifying whether an image is generated by the generator network or a real image. In Reinforcement Learning, the primary network is called the “actor” , while the auxiliary network is called the “critic” . The actor is tasked with performing actions, while the critic is tasked with predicting the expected accumulated reward of that action. In both these applications, the auxiliary network is used for stabilizing the training process, and is usually not active in inference time.
[0131] Recently, this concept has been extended to language processing in order to improve the reasoning capabilities of large language models. In this case, the primary network is the large language model, and it is tasked with generating a solution to a reasoning problem (e.g., a math question) . The auxiliary network is called a reward model, and it is tasked with evaluating the likelihood of the answer being correct, or, in case of step-by-step answers, the likelihood of a step being correct. In inference time, the large language model generates multiple answers, and the reward model evaluates each answer. The answer with the best evaluation is then chosen.
[0132] In future wireless communication systems, effectively sounding and representing the channel space can be challenging when bandwidth and base station antenna counts are extremely large.
[0133] For example, in 6G, the combination of ultra-wide bandwidth and a large number of base station antennas results in a channel in a massive tensor space. However, the physical laws governing this tensor space are sparse. In other words, although the channel space appears highly complex, it is constrained by fundamental physical conditions.
[0134] To sound and represent this sparse channel space, data-driven ML methods can be employed to identify commonalities among these channel data samples. These commonalities can be mathematically represented as a basis (matrix) . Example steps of this approach are as follows.
[0135] 1. High-Dimensional Tensor Vectorization: The high-dimensional channel tensor space is transformed into a one-dimensional vector. Regardless of the original space's dimensionality (2D, 3D, 4D, or higher) , it can be flattened into a one-dimensional vector. Since the channel space is single-modality, it can be directly converted into a unified vector representation.
[0136] 2. Data Matrix Construction: All vector samples (e.g., column-wise vectors) can be combined (or simply juxtaposed) to form a large data matrix. Each sample vector can be very long, and these vectors together constitute a massive data matrix.
[0137] 3. Feature Vector Extraction: Techniques like matrix decomposition can be applied to extract feature vectors from this large data matrix. For instance, SVD matrix decomposition yields a unitary matrix where each column or row vector is strictly orthonormal. In other words, the inner product of any two different column vectors is zero, and only the inner product of a vector with itself is one.
[0138] This unitary matrix U, representing the commonality between different data samples, is considered the basis of the channel space.
[0139] Each original channel space data can be linearly projected onto an equivalent subspace through the unitary matrix U (basis) , showcasing the advantage of sparsity. Some methods can be used for ultra-large data matrix decomposition, such as dimensionality reduction using random matrix sampling and dividing large matrices into smaller matrices. Regardless of the algorithm, the underlying principle remains the same: finding the basis U representing commonality from the data to perform necessary algorithmic processing in a manageable equivalent subspace.
[0140] The unitary matrix U allows for projecting the vast 6G channel space onto a lower-dimensional equivalent subspace. In some implementations, this method is mathematically proven as one of the most optimal compression methods. Specific approaches include Random Matrix Sampling and Matrix Blocking. Random Matrix Sampling can refer to dimensionality reduction of the large matrix by randomly sampling to extract key features. Matrix Blocking can refer to dividing the large matrix into several smaller matrices for individual processing, then combining the results.
[0141] However, compressing and projecting the channel space onto an equivalent subspace obscures the physical characteristics of the original channel. Physical phenomena in the original channel space, such as electromagnetic wave propagation and environmental influences, become less intuitive after projection. This makes qualitative analysis or formulaic interpretation in the equivalent subspace challenging.
[0142] If prediction is performed in the equivalent subspace (due to lower complexity) , identifying hidden higher-order relationships, or modes, can become crucial. This can be a highly challenging problem. Identifying these modes in the equivalent subspace enables further channel compression and processing. To achieve this, non-linear methods might be necessary to approximate the optimal solution. Once these modes are found, one can back-project to the original space using the inverse of the basis U, understanding the characteristics within the original channel.
[0143] Therefore, linear projection through the basis (unitary matrix U) can effectively compress the 6G channel space into a lower-dimensional equivalent subspace. However, interpreting the physical characteristics of the original channel within the equivalent subspace remains a significant challenge.
[0144] In some example channel prediction techniques, a BS (e.g., the network node 170 of FIGS. 1-2) has access to a model that takes as input a “UE State” and predicts the UE State in the next time instant. The UE State includes diverse data such as location, velocity and channel information. The BS shares the model with UEs (e.g., the ED 110 of FIGS. 1-2) , who train the model in a federated learning mechanism, and report the model weights back to the BS, who aggregates them by averaging the models. Specifically, the BS shares a subspace projection mechanism with the UEs, other specifications about the UE State, and a Model. The UEs receive, parse and store these. The UEs constantly measure the channel state and compute their UE State, which includes data specified by the BS, in particular such as their positions, velocities, and compressed channel state (projected into the specified subspace) . The UEs utilize the received Model to predict the UE State in parallel, and compare the prediction to the measured UE State. They train the model in real time utilizing the predicted and measured UE States. After finishing a training epoch, the UEs report the model weights back to the BS. In parallel, the BS receives UE State reports from the UE, and utilize its own Model to make predictions on the future UE States. This reduces the communication overhead from UE to BS, as it requires sparser reports on the UE State, since the BS is able to utilize the predictions. These techniques can face one or more of the following challenges. The subspace projection utilized by some channel prediction techniques may be unable to fully capture the redundancy in the channel information. It can be challenging to fully exploit long-term-memory knowledge of UE States. The UE State reporting strategy (e.g., periodic, selective transmission) may lead to excess information transmission or to error propagation. In case of significant environmental changes, catastrophic forgetting (e.g., a machine learning model forgets previously learned information upon learning new information) may occur, and thus the model may need to be essentially re-trained.
[0145] In some implementations, aspects of the present disclosure relate to the deployment of three models focused on channel prediction in wireless communication networks. The three models include an Encoder Model or embedder, a Decoder Model, and an Evaluator Model. The Encoder Model or embedder takes as input a “UE State, ” and outputs an “Embedded UE State, ” which captures the most important parameters of the UE State and processes these parameters to a suitable domain for use in channel prediction. The Decoder Model takes as input the Embedded UE State, as well as a sequence of “Subspace Representations of Channel States” (SRCSs) , and outputs a prediction of the SRCS at the next time step. The Decoder Model is an auto-regressive model, that is, it can recursively generate predictions for multiple time steps by appending its prediction to the input and executing the prediction again. Finally, the Evaluator Model, or evaluator, takes as input the UE State or the Embedded UE State, the sequence of predicted SRCSs, and an environment representation, and outputs a Prediction Score for each SRCS. In some implementations, the BS and the UE can have separate Evaluator Models and can be tasked with predicting the same underlying metric (e.g., the mean squared error between predicted SRCS and actual SRCS) . The Encoder and the Decoder models together constitute the Predictor Model. The Prediction Scores are used by a device (e.g., a UE or BS) to update its Predictor Model. Further details about the Subspace Representation of Channel State and each model are provided as follows.
[0146] Aspects of the present disclosure relate to processing the vast MIMO channel space by mapping it to an equivalent manifold instead of directly handling the original space. The following steps are involved.
[0147] 1. Projection into the manifold: Firstly, a manifold projection of the channel space can be first identified through various methods. The purpose of this mechanism is to compress the data in an equivalent manifold. For instance, a linear projection can be learned through principal component analysis (PCA) , or non-linear manifolds can be learned through auto-encoder models, or other suitable methods. In other words, the manifold projection can refer to a function that maps from the original channel state domain to another space, e.g., one of lower dimension, which can be denoted a latent space.
[0148] 2. Projection into original dimension: A projection from the manifold back to the original channel space dimension is also found through various methods. Similar to the manifold projection, the projection into original dimension can be a linear projection learned through PCA, a non-linear projection learned through auto-encoder models, or other suitable methods. In other words, the projection into original dimension can refer to a function that maps from the latent space domain to the original channel state domain, and can be an approximate inverse to the function that projects into the manifold. In some implementations, if both functions are applied to a channel state, the output can be close to the original channel state.
[0149] 3. Equivalent subspace processing: After mapping to the equivalent manifold, the processing scale is significantly reduced, e.g., decreasing from tens of millions to a few thousand or even smaller scales. This not only improves processing efficiency but also reduces computational complexity.
[0150] Mathematically, the equivalent manifold ensures that any transformation (linear or non-linear) in the equivalent manifold can be equivalently performed in the original high-dimensional space. The reduced-dimension vector is denoted as Manifold Representation of Channel State (MRCS) .
[0151] This Manifold Representation may also be called a “Codebook” , as it relates to source coding and compression, where a high-dimensional variable is compressed into a lower-dimension variable through various mechanisms.
[0152] In some implementations, the Encoder Model transforms the original UE State space into an Embedded UE State. Specifically, the UE collects various user state parameters, such as a position of the UE, a velocity or a movement speed of the UE, UE category, RF channel parameters (e.g., received signal strength, Angle of Departure (AoD) , Angle of Arrival (AoA) ) , etc. For example, the position of the UE in a 3D space can be represented by coordinate values (X, Y, Z) , where X is the longitude, Y is the latitude, and Z is the height or altitude. The received signal strength can be a strength of signals received at the UE. The AoD can be an AoD of the signals. The AoA can be an AoA of the signals. In other words, the UE state can include the above described user state parameters. The Encoder Model (embedder) then maps the data to a sequence of symbols, which can be effectively processed by the Decoder Model. Because the Encoder Model (embedder) may use the AI / ML (esp. transformer-based) model, the Encoder Model can input the multi-modality data. In some implementations, some prompts can be integrated into the input too. These prompts can instruct the Encoder Model (embedder) and ensuing Decoder Model.
[0153] The UE State is determined not only by the user's position and environment relative to the base station within the cell but also by the user's receiver implementation, RF front-end and antenna configuration, and electromagnetic wave propagation characteristics. These factors or parameters or sub-fields, combine to form the user's current channel state. In some implementations, the UE state further includes one or more of the above factors, parameters, or sub-fields.
[0154] Complex relationships can exist between different state parameters (sub-fields) , aiding in predicting the user's channel changes in the next moment. For instance, certain state parameters might have a decisive impact on channel changes, while others have minimal effects, requiring the prediction function to consider relative weights. Furthermore, the combined relationship between specific parameters (e.g., parameter A and parameter B) might play a key role, and even multi-parameter combinations (e.g., parameters A, B, and C) might significantly influence channel changes.
[0155] Through AI / ML encoder models, persistent relational patterns between state parameters and their combinations may be captured or learned, and these learned patterns are utilized to predict future user channel states. Due to system complexity, it may not be possible to predetermine which sub-field or parameter combination decisively influences channel changes. Therefore, data-driven approaches are employed, training models on vast data samples to capture these complex relational models.
[0156] In some implementations, through AI / ML encoder models, e.g., transformer-based models, the prompts as a kind of instructions can be integrated into the Embedded UE State to explicitly control the following prediction.
[0157] During the inference stage, continuously updating user state parameters fed into the AI / ML encoder model reflects changes in their different times and locations. This dynamic updating helps the model predict channel changes more accurately.
[0158] The Encoder Model acts as a “translator” that maps the UE State to the embedding space of the Decoder Model. This processing can allow the Embedded UE State to be robust to noisy and missing UE State data, and can provide the Decoder Model with data in a format suitable for the Decoder processing.
[0159] In some implementations, the Encoder Model may take as input a control variable, for example, the output of the Evaluator Model, and adjust its Embedding according to the score. This can be used, for example, to control the complexity of the model. Unlike the prompt, these control variables may change the neuron weights on some layers of the model or select one model from a repository of a number of candidate models.
[0160] In some implementations, the Decoder Model takes as input a sequence of SRCSs and the Embedded UE State, and outputs the prediction of the next-time-step SRCS. This allows for recursive prediction, where the output SRCS is appended to the input sequence of SRCSs, and this extended sequence is used as input, to generate a new prediction.
[0161] In some implementations, the Decoder Model may take as input a control variable, for example, the output of the Evaluator Model, and adjust its prediction according to the score. This can be used, for example, to control the complexity of the model. These control variables may change the neuron weights on some layers of the model or select one model from a repository of a number of candidate models.
[0162] In some implementations, the Evaluator Model used by the BS takes as input the UE State or the Embedded UE State, the sequence of predicted SRCSs, and an environment representation, and outputs a Prediction Score for each SRCS. The environment representation includes, for example, weather conditions, amount of traffic, and other parameters that may influence channel conditions channel dynamics. In some implementations, the Evaluator Model may output controlling variables or prompts, instead of, or in addition to, the scores.
[0163] In some implementations, the Evaluator Model used by the UE can simply handle computation of the desired metric, taking as input the predicted SRCSs and the actual measured SRCSs.
[0164] FIG. 6 illustrates an example channel prediction process 600. The channel prediction process 600 (also referred to as the process) can be performed by a BS (e.g., the network node 170 of FIGS. 1-2) and a UE (e.g., the ED 110 of FIGS. 1-2) of a RAN (e.g., the RAN 120 of FIG. 1) . While the process 600 is described with reference to the BS and the UE, this description is provided for illustrative purposes only and is not intended to be limiting. In practice, the process 600 can be applied to other instances of network node and terminal devices or equivalents thereof. It is understood that steps or operations shown in the process 600 are not exhaustive and that other operations can be performed as well before, after, or between any of the illustrated operations. Further, some of the steps or operations may be omitted, performed simultaneously, or performed in a different order than that shown in FIG. 6.
[0165] In some implementations, one or more of the models (e.g., Encoder Model, Decoder Model, and Evaluator Model described above) can be in the BS initially. For example, the BS can have access to a Subspace Representation Basis, an Encoder Model, a Decoder Model, and an Evaluator Model. In some implementations, the BS can have access to multiple Predictor Models, and each Predictor Model can include a pair of Encoder Model and Decoder Model. The pair of Encoder Model and Decoder Model can be trained and / or fine-tuned jointly as described below.
[0166] The Subspace Representation Basis can be used to map the measured channel state to the MRCS, and a MRCS back to the channel state. In some implementations, this basis has a maximum dimension, but the BS may choose to use a reduced dimension for compression purposes, which may cause some performance loss. This basis can be learned through PCA, and the manifold representation can be a linear projection. Therefore, in some implementations, a MRCS of the channel state information can be referred to as a Subspace Representation of Channel State (SRCS) . Thus, the projection functions are matrix operations (e.g., linear matrix multiplications) , and can be represented by appropriate matrices. In some implementations, the proposed solution can support both original measured channel state and traditional CSI in 5G.
[0167] The BS can have access to multiple Encoder Models. Each of the multiple Encoder Models can take as input the UE position and velocity, denoted as the UE State, and outputs an Embedded UE State. In some implementations, the Encoder Model may take as input and / or output prompts. In some implementations, each Encoder Model can be fine-tuned to respective environmental parameters. The environmental parameter refers to environmental factors that impact a channel state between the BS and the UE. In some implementations, the environmental parameter can include factors that impact performance of channel prediction. For example, the environmental parameter can include one or more of an environmental condition, a resource availability, a power condition, and a prediction score, or any other suitable parameters. The environmental condition can be a weather condition or a traffic condition in an area, which is or close to locations of the BS and / or the UE. The weather condition can include different types of atmosphere status, such as “Sun, ” “Rain, ” “Fog, ” and “Snow. ” The traffic condition can refer to a density and flow of vehicles or users in the area, including, but is not limited to, “Low Traffic, ” “Medium Traffic, ” and “High Traffic. ” The resource availability can include parameters such as bandwidth, network capacity, and computational resources, and can indicate the current accessibility or status of such resources for performing channel prediction. The power condition refers to the status of the power supply or energy reserves required for operation of the UE or the BS, which can include battery levels, power supply status, or energy efficiency indicators (e.g., power saving mode) . The prediction score can refer to an evaluation metric or score of one or more current or past channel predictions. In some implementations, the environmental parameter can be obtained through sensing by the BS.
[0168] The BS can have access to multiple Decoder Models. Each Decoder Model can take as input the appropriate Embedded UE State and a sequence of SRCSs, and outputs a prediction of the next SRCS. In some implementations, each Decoder Model is trained jointly with an Encoder Model. Thus, Decoder Models are similarly fine-tuned to the environmental parameters (e.g., different types of weather and traffic conditions) . In some implementations, a pair of Encoder Model and Decoder Model (e.g., jointly trained) are, together, denoted as a Predictor Model.
[0169] The Evaluator Model can take as input the UE State, the sequence of predicted SRCSs, and outputs a Prediction Score for each SRCS.
[0170] The process 600 can begin at a configuration step 601. In this step, the BS can transmit configuration information to the UE.In some implementations, the configuration information includes a format of a UE state of the UE and a format of a representation of a channel state information between the BS and the UE. For example, the representation of the channel state information can be a SRCS.
[0171] During this step, the BS can periodically broadcast the UE State domain to all relevant UEs (e.g., through System Information Blocks (SIBs) ) . For example, the relevant UEs can include the UE and other UEs in the same RAN. In particular, the BS reports the format (e.g., a quantization format) used for the UE State, and that the UE State includes the UE Position and Velocity. The BS can also inform the UEs of the SRCS domain specifications, that is, the dimension of the subspace representation (e.g., through SIBs) . The BS can inform the UEs of model specifications, such as a number of layers and a number of parameters of the models used by the BS. The model specifications can include dimensions associated with the layers of the models used by the BS. In some implementations, the model specifications can include number of weights corresponding to at least one layer of the models used by the BS. The BS can inform the UEs that the BS is running a Predictor Model for the SRCS at default intervals, what parameters the BS requires the UE to transmit in the uplink, and that the UE may request the model to proactively evaluate the parameters (e.g., in real time) . Additionally, the BS can also report a metric target of the Evaluator Model. In some implementations, the metric is the mean squared error (MSE) between predicted SRCS and measured SRCS. The UEs can receive, parse, and store these specifications.
[0172] Table 1
[0173] Table 1 illustrates content of example configuration information. While Table 1 shows various parameters or information that can be included in the configuration information, this example is not intended to be construed in a limiting sense. In practice, the configuration information may only include one or more of the parameters as shown in Table 1. The same notion is applied to other tables in the present disclosure.
[0174] While in some implementations of the present disclosure, information transmitted between the BS and the UE is carried in RRC messages (e.g., SIBs and “RRCConnectionRequest” as described below) , the examples are for illustration purposes and are not intended to be construed in a limiting sense. In practice, any suitable signaling, including, but not limited to, higher layer signaling (e.g., RRC or MAC-CE) and physical layer signaling, can be used to carry the information transmitted between the BS and the UE.
[0175] In a configuration response step 602, the UE can transmit a capability message in response to the configuration information the UE receives from the BS. FIG. 7 illustrates example signaling in the configuration response step 602. In some implementations, when the UE wishes to connect to the BS, at 702, the UE initiates an RRC Connection Request (“RRCConnectionRequest” ) . The UE can transmit the RRC Connection Request message to the BS. This message can include standard information required for connecting to the BS, and additionally the UE informs the BS about its capabilities to provide the required information for channel prediction. In particular, the UE informs the BS whether the UE can perform the processing required to generate the SRCS in the desired dimension. If the UE cannot perform the processing, the communication may fall back to a standard communication protocol with no channel prediction on the BS. The UE may also inform the BS whether the UE can execute a channel prediction model (e.g., an Encoder Model and a Decoder Model) in real-time. For example, if the UE can execute the model, the UE can include a “PredictorModelRequest” signal in the message (e.g., the capability message or the RRC Connection Request) . Table 2 illustrates content of an example capability message.
[0176] Table 2
[0177] Once the BS receives the “RRCConnectionRequest” message, the BS can perform the standard procedure for establishing the RRC Connection. Additionally, if the UE has informed that it is capable of executing the channel prediction model, the BS allocates downlink resources for model transfer. If the UE has informed that it is capable of providing the requested UE State and SRCS, the BS allocates uplink resources accordingly. As shown in FIG. 7, at 704, the BS can transmit an “RRCConnectionSetup” message to the UE. That is, the BS can respond to the request and establish the RRC Connection with the “RRCConnectionSetup” message, which includes the information about the uplink resources where the UE will transmit the UE State and SRCS. In some implementations, the “RRCConnectionSetup” message can include information about downlink resources where the BS will perform the model transfer (e.g., as described below with respect to 804 of FIG. 8) . In some implementations, if the UE is unable to execute the channel prediction model in real time, the communication may fall back to a protocol where the UE is not required to perform channel prediction.
[0178] In a DL and UL transmission step 603, the BS can transmit the Subspace Representation Projection and a current Predictor Model to the UE, and the UE can transmit the UE State and the initial SRCS to the BS. FIG. 8 illustrates example DL and UL transmissions in the DL and UL transmission step 603. In some implementations, once the connection has been established, at 802, the BS transmits the Subspace Representation Projection to the UE through the allocated downlink resources. This transmission can occur through the Physical Downlink Shared Channel (PDSCH) (e.g., as shown in FIG. 8) , Physical Downlink Control Channel (PDCCH) , or a dedicated physical channel introduced in future standards, such as Physical Downlink Artificial Intelligence / Machine Learning Channel (PDAI / MLCH) . The UE receives the Subspace Representation Projection and stores it. The UE can then utilize this Subspace Representation Projection to compute a current SRCS (e.g., an initial SRCS) . For example, the UE can measure current channel state information, and determine the initial SRCS based on the current channel state information and the Subspace Representation Projection. The UE can also compute its current UE State, utilizing the format specified by the BS, which has been stored (e.g., in the configuration step 601) . At 804, the BS can transmit a channel prediction model to the UE. For example, at 804, the BS can transmit a message (e.g., through PDSCH) indicating a channel prediction model for the UE to use. The channel prediction model can be selected by the BS based on an environmental parameter associated with the UE. As described above, the environmental parameter can include one or more of an environmental condition, a resource availability, a power condition, and a prediction score. The channel prediction model can include at least one of an Encoder Model or a Decoder Model selected by the BS from the multiple Encoder Models and the multiple Decoder Models that the BS have access to. In some implementations, the message that the BS transmits to the UE at 804 can include an index associated with the channel prediction model. For example, the index is identified using a mapping table and is mapped to the channel prediction model based on the mapping table. The mapping table can be pre-defined or preconfigured. In some instances, the mapping table can be configured by higher layer signaling. At 806, the UE then transmits the UE State and the initial SRCS utilizing the uplink resources assigned by the BS. This transmission can occur through the Physical Uplink Shared Channel (PUSCH) or a dedicated physical channel introduced in future standards. The BS receives, parses, and stores the UE State and the SRCS through the PUSCH.
[0179] In some implementations, in the DL and UL transmission step 603, the BS can transmit a Decoder Model to the UE, instead of the Predictor Model. For example, the UE may have access to a general-purpose Encoder Model, which is compatible with the Decoder Models present in the BS. In another example, the UE can have access to a set of Encoder Models that are pair-wise compatible with the Decoder Models present in the BS. In these cases, the BS may only transfer an appropriate Decoder Model to the UE (e.g., at 804 of FIG. 8) . In some implementations, instead of transmitting the UE State to the BS (e.g., at 806 of FIG. 8) , the UE may transmit the Embedded UE State, that is, the output of the Encoder Model used by the UE. Alternatively, the UE may transmit both the UE State and the Embedded UE State to the BS.
[0180] Returning to FIG. 6, in a parallel prediction and evaluation step 604, the BS can utilize the stored UE State as input to its Encoder Model, generating an Embedded UE State. In addition, the BS can then utilize the Embedded UE State and the stored Initial SRCS as input to the Decoder Model, generating a sequence of SRCS predictions for a pre-defined window, and the BS can store the sequence.
[0181] FIG. 9A illustrates example prediction operations performed in the parallel prediction and evaluation step 604. In some implementations, the prediction operations can be performed by the BS and / or the UE. As shown in FIG. 9A, the UE State 902 can be stored in any storage structure that may include, but is not limited to, a database 901 or one or more memories. The BS can utilize the UE State 902 as input to an Encoder Model 903, generating an Embedded UE State 904. The BS utilizes the Embedded UE State 904 and one or more SRCSs 905 (e.g., stored in a database 908) as input to a Decoder Model 906, which generates a SRCS prediction 907 for a next time step. In an initial time step, the SRCSs 905 can be an Initial SRCS (e.g., the initial SRCS that the BS received from the UE in the DL and UL transmission step 603) . In the following time steps, the SRCSs 905 can be a sequence of SRCSs, which include the Initial SRCS and the SRCS prediction 907 generated by the Decoder Model 906. For example, the BS can append the SRCS prediction 907 to the Initial SRCS, and can utilize the combined sequence as input to the Decoder Model 906, generating the following prediction 907. The BS can repeat this process recursively.
[0182] FIG. 9B illustrates example evaluation operations performed by the BS in the parallel prediction and evaluation step 604. The sequence of SRCSs 905 (e.g., from database 908) and the UE State 902 (e.g., from database 901) can be used as input to an Evaluator Model 911, which outputs a sequence of Prediction Scores 912. For example, a Prediction Score in the sequence of Prediction Scores 912 can be generated for each SRCS in the sequence of SRCSs 905. The sequence of Prediction Scores 912 can be stored in a database 913. In some implementations, the Prediction Scores 912 can be an estimate of the MSE between the predicted SRCS and the measured SRCS (e.g., as shown in FIG. 9C) . The measured SRCS can also be referred to as true SRCS. The Prediction Scores 912 are utilized by the BS, in combination with system parameters 914 (e.g., stored in database 915) to decide whether to “Continue” the prediction. The system parameters 914 include, but are not limited to, current resources availability, an environmental condition, and other parameters available to the BS that may impact this decision. The environmental condition can refer to a physical environment representation, such as locations of other significant objects, weather conditions, and other environment conditions that may influence the channel prediction. In some implementations, the environmental condition can be obtained through sensing by the BS. This decision process can be performed by, for example, a decision unit 916 of the BS. An action 917 can be generated by the decision unit 916. In some implementations, the action 917 can be either to “Continue” the prediction, without additional information exchange between BS and UE, or to “Update” the input and / or the model. In some implementations, the BS may trigger a Model Update and / or an Input Update as described in detail below with reference to an update step 605. The BS can transmit a request for the Model Update and / or the Input Update through a downlink transmission.
[0183] In some implementations, if the BS has received the Embedded UE State from the UE (as described above in the DL and UL transmission step 603) , the BS may employ the Embedded UE State directly as input to the Decoder Model of the BS. Alternatively, if the BS has received both the UE State and the Embedded UE State from the UE, the BS may compute the Embedded UE State generated by the Encoder Model of the BS and compare it against the state generated by the Encoder Model of the UE (e.g., the Embedded UE State received from the UE) . In case they are different (e.g., the difference exceeding a threshold) , the BS may trigger a Model Update to aid the UE in updating the UE’s model for the current environment.
[0184] In some implementations, the UE has access to its own general-purpose Encoder and Decoder models, which are compatible with the Evaluator Model of the BS. The Evaluator Model of the BS, instead of outputting a score, can output a Model Update action. For example, the BS may identify that the Predictor Model of the UE (e.g., the UE’s general-purpose Encoder and Decoder models) is not suitable for prediction in the current weather, for example, when snowing. Thus, the BS may output an action requesting the UE to fine-tune the Predictor Model for the current weather. The BS may have access to mechanisms that can help the UE with this model update. For example, the BS may have access to a model similar to the general-purpose model employed by the UE. In another example, the BS may have access to a fine-tuned model specific for the weather (e.g., snowing) , and the BS may inform the UE about the suggested model update. Alternatively, the UE may have a pre-defined mechanism for updating the model, and the BS merely needs to inform the UE about the environment state.
[0185] In some implementations, the BS utilizes the channel predictions in the downlink communication. If the BS evaluates that the current predictions are satisfactory, the BS can utilize the channel prediction (e.g., the SRCS prediction 907 or the sequence of SRCSs 905 stored in database 908) for tasks such as scheduling and precoding. In some implementations, the channel prediction can be used to compute a precoder matrix and perform power allocation for downlink communication. For example, at each time step, the BS utilizes the SRCS prediction for the current time step to reconstruct the channel state matrix, utilizing the Subspace Representation Basis to do so. Then, the BS can perform a singular value decomposition (SVD) of the channel matrix, which returns a left-hand basis, the singular values, and a right-hand basis. The singular values are then utilized to perform power allocation, for example, utilizing water-filling techniques. The Hermitian of the right-hand basis can be utilized as the precoder matrix for downlink communication. In some implementations, the BS generates a number of data streams (e.g., encoded data) equal to a number of receive antennas, and pre-multiplies the data streams with the precoder matrix. The BS then transmits the resulting vector through the appropriate downlink resource elements.
[0186] In parallel, the UE can mirror the prediction process (e.g., as shown in FIG. 9A) . The UE can utilize a Predictor Model (e.g., the Encoder Model 903 and the Decoder Model 906) received in the DL and UL transmission step 603, and store the predictions (e.g., the SRCS prediction 905 or 907) generated by the received Predictor Model for a pre-defined window.
[0187] In some implementations, the BS may not compute its own predictions (e.g., the SRCS prediction 907) . Rather, the BS may receive the predictions generated by the UE and use the received predictions as input to the Evaluator Model of the BS (e.g., the Evaluator Model 911 of FIG. 9B) . In some implementations, the Evaluator Model 911 takes a sequence of UE States, rather than a single UE State.
[0188] For example, at each time step, the UE computes its updated UE State. The UE utilizes the current UE state as input to its Encoder Model, generating an Embedded UE State. The UE then utilizes the Embedded UE State and the current sequence of SRCSs as input to the Decoder Model, generating a prediction (or estimate) for the current SRCS. This prediction is appended to the current sequence of SRCSs, to be utilized as input in the next time step.
[0189] The UE then transmits this prediction (e.g., the UE state and the SRCS prediction) to the BS (e.g., in the PUSCH or a dedicated physical channel) . Specifically, the UE requests uplink resources through a scheduling request message. Once the UE has been granted a resource assignment, the UE transmits both the current prediction of SRCS and the current UE State. The BS stores the received SRCS and the UE State, each in a sequence.
[0190] The sequence of SRCSs and the sequence of UE States are then used by the BS as input to the Evaluator Model (e.g., the Evaluator Model 911 of FIG. 9B) , which outputs a sequence of Prediction Scores. For example, a Prediction Score can be generated for each SRCS in the sequence. In some implementations, these scores are an estimate of the MSE between the predicted SRCS and the measured true SRCS. As described with reference to FIG. 9B, the Prediction Scores can be utilized by the BS, in combination with system parameters, such as current resources availability, current environment, and other parameters available to the BS that may impact this decision, to decide whether to allow the UE to “Continue” the prediction, without additional information exchange between the BS and the UE, or to “Update” either the input or the model. In some implementations, the BS may trigger a Model Update and / or an Input Update.
[0191] FIG. 9C illustrates example evaluation operations performed by the UE in the parallel prediction and evaluation step 604. At each time step, the UE can compute actual measured SRCS 919 (which can be referred to as true SRCS sequence and can be stored in database 918) , compare the measured SRCS 919 against the SRCS (e.g., the SRCS 905 or 907) predicted by the Predictor Model, compute a metric 921 (e.g., the EvaluationMetric as shown in Table 1) informed by the BS, and store the metric 921. For example, as shown in FIG. 9C, the metric 921 can be computed by a metric determination unit 920 and can be stored in a database 922. The metrics 921 (e.g., for a plurality of time steps) are denoted as Prediction Score Targets. In some implementations, the metric is MSE. In other words, the BS can inform the UE to use the MSE as the metric, and the UE computes the MSE between the predicted SRCS (e.g., the SRCS 905 or 907) and measured SRCS (e.g., the measured SRCS 919) . Then, the UE may choose to trigger an Input Update. The Prediction Scores 921 can be utilized by the UE to decide whether to trigger the Input Update. This decision process can be performed by, for example, a decision unit 923 of the UE. An action 924 can be generated by the decision unit 923. In some implementations, the action 924 can be either to trigger or to not trigger the Input Update. For example, the UE can choose to trigger the Input Update if the MSE between the measured SRCS and the SRCS predicted by the Predictor Model is larger than a certain threshold. The UE can transmit a request for the Input Update through an uplink transmission. If the UE chooses to not trigger the Input Update, the UE may recursively generate new predictions, beyond the initial pre-defined window.
[0192] Returning to FIG. 6, the process 600 proceeds to an update step 605 when the BS and / or the UE trigger the “Update. ” In some implementations, if the BS decides to trigger a Model Update, the BS allocates downlink resources for the update, and then prepares and transmits an “UpdateRequest” message to the UE, informing the UE that the BS is requesting the Model Update and the assigned resources (e.g., in a PDSCH) .
[0193] In some implementations, if the BS decides to trigger an Input Update, the BS allocates uplink resources for the update, and then prepares and transmits an “UpdateRequest” message to the UE, informing the UE that the BS is requesting the Input Update and the assigned resources (e.g., in a PUSCH) .
[0194] In some implementations, if the BS decides to trigger both a Model Update and an Input Update, the BS allocates uplink and downlink resources for the update, and then prepares and transmits an “UpdateRequest” message to the UE, informing the UE that the BS is requesting both the Model Update and the Input Update, and the assigned resources (e.g., in the PUSCH and the PDSCH) . Table 3 illustrates content of an example “UpdateRequest” message transmitted from the BS to the UE.
[0195] Table 3
[0196] In some implementations, when the UE receives the “UpdateRequest” message from the BS, the UE prepares and transmits an “UpdateRequestAcknowledge” message to the BS. The “UpdateRequestAcknowledge” message can be transmitted through the PUCCH. This message can simply acknowledge that the UE has received the request, and that the UE is prepared to either listen to the PDSCH for the Model Update, or transmit the Input Update in the PUSCH, or both. Table 4 illustrates content of an example “UpdateRequestAcknowledge” message transmitted from the UE to the BS.
[0197] Table 4
[0198] In some implementations, if the UE decides to trigger an Input Update, the UE prepares and transmits an “UpdateRequest” message to the BS, informing the BS that the UE is requesting an Input Update. The UE can transmit this message through the PUCCH. Table 5 illustrates content of an example “UpdateRequest” message transmitted from the UE to the BS.
[0199] Table 5
[0200] In some implementations, when the BS receives the “UpdateRequest” message from the UE, the BS allocates uplink resources for the update, and then prepares and transmits an “UpdateRequestAcknowledge” message to the UE, informing the UE of the assigned resources in the PUSCH. Although the “UpdateRequest” was triggered by the UE’s request for the Input Update, the BS may also decide to update the model. Thus, the “UpdateRequestAcknowledge” message can also indicate that the BS is requesting the Model Update and indicate downlink resources (e.g., in the PDSCH) allocated by the BS for the Model Update. Table 6 illustrates content of an example “UpdateRequestAcknowledge” message transmitted from the BS to the UE.
[0201] Table 6
[0202] In some implementations, if the BS has requested a Model Update and has received an acknowledgement (e.g., in the “UpdateRequestAcknowledge” message of Table 4) from the UE, the BS selects an appropriate model for the current environment and transmits the model update to the UE through the PDSCH. The UE receives the model and stores the model. In some implementations, the appropriate model can be a new channel prediction model selected by the BS from the multiple Encoder Models and the multiple Decoder Models that the BS have access to. The new channel prediction model can include at least one of an Encoder Model or a Decoder Model. For example, the BS can select the new channel prediction model based on a current environmental parameter associated with the UE. The current environmental parameter may be different from the environmental parameter the BS relied on when selecting the channel prediction model in the DL and UL transmission step 603. In other words, the BS can select the new channel prediction model based on a change to the environmental parameter associated with the UE.
[0203] In some implementations, if an Input Update has been requested, the UE computes the updated UE State and SRCS and transmits the updated UE State and SRCS to the BS using the assigned resources (e.g., in the PUSCH) .
[0204] In some implementations, the UE has access to one or more Encoder Models (e.g., as described in DL and UL transmission step 603) . In this case, when a Model Update has been requested, the BS may request that the UE switches to a more appropriate Encoder Model (e.g., in case the UE has access to multiple Encoder Models) , or may request that the UE updates or fine-tunes the general-purpose Encoder Model for the current environment.
[0205] FIG. 10 illustrates an example communication method 1000. The communication method 1000 (also referred to as the method) can be performed by a device 1001 (e.g., a BS or the network node 170 of FIGS. 1-2) and a device 1002 (e.g., a UE or the ED 110 of FIGS. 1-2) and of a RAN (e.g., the RAN 120 of FIG. 1) . While the method 1000 is described with reference to the devices 1001 and 1002, this description is provided for illustrative purposes only and is not intended to be limiting. In practice, the method 1000 can be applied to other instances of network node and terminal devices or equivalents thereof. It is understood that steps or operations shown in the method 1000 are not exhaustive and that other operations can be performed as well before, after, or between any of the illustrated operations. Further, some of the steps or operations may be omitted, performed simultaneously, or performed in a different order than that shown in FIG. 10.
[0206] In some implementations, the method 1000 can start at 1003, where the device 1001 transmits a first message to the device 1002 (e.g., as described with reference to 804 of FIG. 8) . The first message can indicate a first channel prediction model for the device 1002 to use. The first channel prediction model can be selected based on at least one parameter (e.g., the environmental parameter as described with reference to FIG. 6) associated with the device 1002.
[0207] At 1004, the device 1001 transmits a second message to the device 1002. The second message can indicate a second channel prediction model (e.g., the new channel prediction model as described with reference to the update step 605 of FIG. 6) . The second channel prediction model can be selected based on a change to the at least one parameter associated with the device 1002.
[0208] In some implementations, the first channel prediction model and the second channel prediction model are selected from a plurality of channel prediction models (e.g. the multiple Encoder Models and the multiple Decoder Models that the BS can access) .
[0209] In some implementations, the at least one parameter includes at least one of: an environmental condition; a resource availability; a power condition; or a prediction score.
[0210] In some implementations, transmitting the first message includes transmitting at least one model. The at least one model includes at least one of a first model (e.g., an Encoder Model) or a second model (e.g., a Decoder Model) .
[0211] In some implementations, each of the first channel prediction model and the second channel prediction model includes the first model and the second model.
[0212] In some implementations, the method 1000 further includes transmitting configuration information (e.g., as shown in Table 1) . The configuration information includes a format for a device state (e.g., the UEStateFormat in Table 1) , a format for a representation of channel state information (e.g., the MRSpec in Table 1) , and at least one model parameter associated with a channel prediction model (e.g., the PredictorModelSpec in Table 1) .
[0213] In some implementations, the at least one model parameter includes at least one of a number of layers of the channel prediction model, dimensions associated with the layers of the channel prediction model, or number of weights corresponding to at least one layer of the channel prediction model.
[0214] In some implementations, the method 1000 further includes the device 1001 receiving a capability message from the device 1002 and the device 1001 transmitting a resource configuration to the device 1002. These operations can be considered as part of the operation of the device 1001 transmitting the first message to the device 1002. The capability message can indicate a capability of the device 1002 to execute the first channel prediction model. The resource configuration can indicate a downlink resource for transmission of the first channel prediction model. The first message can be transmitted to the device 1002 in the downlink resource.
[0215] In some implementations, the first message includes an index associated with the first channel prediction model. The index is identified using a predefined mapping table.
[0216] In some implementations, the method 1000 further includes: the device 1001 receiving the device state and the representation of the channel state information from the device 1002; the device 1001 obtaining, using the first model (e.g., the Encoder Model 903 of FIG. 9A) , an embedded device state (e.g., the Embedded UE State 904) based on the device state (e.g., the UE State 902) ; the device 1001 obtaining, using the second model (e.g., the Decoder Model 906) , at least one channel state prediction (e.g., the SRCS prediction 905 or 907) for a time window based on the embedded device state and the representation of the channel state information; and the device 1001 obtaining, using a third model (e.g., the Evaluator Model 911 of FIG. 9B) , a prediction quality evaluation (e.g., the Prediction Scores 912) based on the device state and the at least one channel state prediction.
[0217] In some implementations, the method 1000 further includes the device 1001 transmitting an update request (e.g., the UpdateRequest message in Table 3) to the device 1002 based on the prediction quality evaluation. The update request includes a request to update at least one of: the device state; the representation of the channel state information; or the first channel prediction model.
[0218] In some implementations, the method 1000 further includes the device 1001 receiving an update request (e.g., the UpdateRequest message in Table 4) from the device 1002. The update request includes a request to update at least one of: the device state; or the representation of the channel state information.
[0219] In some implementations, the device 1002 can transmit the device state and the representation of the channel state information to the device 1001 (e.g., at 806 of FIG. 8) . The device 1002 can obtain, using an Encoder Model of the device 1002, an embedded device state based on the device state; obtain, using a Decoder Model of the device 1002, at least one channel state prediction for a time window based on the embedded device state and the representation of the channel state information; store the at least one channel state prediction (e.g., as described with reference to FIG. 9A) ; and obtain one or more prediction scores (e.g., the metric 921 of FIG. 9C) based on a difference between the at least one channel state prediction and a measured channel state information (e.g., as described with reference to FIG. 9C) .
[0220] In some implementations, the device 1002 can transmit an update request (e.g., the UpdateRequest message in Table 4) to the device 1001 based on the one or more prediction scores. The update request includes a request to update at least one of: the device state; or the representation of the channel state information.
[0221] In the present disclosure, the terms “a” or “an” are defined to mean “at least one” , that is, these terms do not exclude a plural number of items, unless stated otherwise.
[0222] In the present disclosure, terms such as “substantially” , “generally” and “about” , which modify a value, condition or characteristic of a feature of an example embodiment, should be understood to mean that the value, condition or characteristic is defined within tolerances that are acceptable for the proper operation of the example embodiment for its intended application.
[0223] In the present disclosure, unless stated otherwise, the terms “connected” and “coupled” , and derivatives and variants thereof, refer herein to any structural or functional connection or coupling, either direct or indirect, between two or more elements. For example, the connection or coupling between the elements can be acoustical, mechanical, optical, electrical, thermal, logical, or any combinations thereof.
[0224] In the present disclosure, expressions such as “match” , “matching” and “matched” , including variants and derivatives thereof, are intended to refer herein to a condition in which two or more elements are either the same or within some predetermined tolerance of each other. That is, these terms are meant to encompass not only “exactly” or “identically” matching the two elements but also “substantially” , “approximately” or “subjectively” matching the two or more elements, as well as providing a higher or best match among a plurality of matching possibilities.
[0225] In the present disclosure, the expression “based on” is intended to mean “based at least partly on” , that is, this expression can mean “based solely on” or “based partially on” , and so should not be interpreted in a limited manner. More particularly, the expression “based on” could also be understood as meaning “depending on” , “representative of” , “indicative of” , “associated with” or similar expressions.
[0226] In the present disclosure, the terms "system" and "network" may be used interchangeably in different embodiments of this application. "At least one" means one or more, and "a plurality of" means two or more. The term "and / or" describes an association relationship of associated objects, and indicates that three relationships may exist. For example, A and / or B may indicate the following three cases: Only A exists, both A and B exist, and only B exists, where A and B may be singular or plural. The character " / " indicates an "or" relationship between associated objects. "At least one of the following items (pieces) " or a similar expression thereof indicates any combination of these items, including a single item (piece) or any combination of a plurality of items (pieces) . For example, "at least one of A, B, or C" includes: only A; only B; only C; A and B; A and C; B and C; or A, B, and C, and "at least one of A, B, and C" may also be understood as including: only A; only B; only C; A and B; A and C; B and C; or A, B, and C. In addition, unless otherwise specified, ordinal numbers such as "first" and "second" in embodiments of this application are used to distinguish between a plurality of objects, and are not used to limit a sequence, a time sequence, priorities, or importance of the plurality of objects.
[0227] A person skilled in the art should understand that embodiments of this application may be provided as a method, an apparatus (or system) , computer-readable storage medium, or a computer program product. Therefore, this application may use a form of a hardware-only embodiment, a software-only embodiment, or an embodiment with a combination of software and hardware. Moreover, this application may use a form of a computer program product that is implemented on one or more computer-usable storage media (including but not limited to a disk memory, an optical memory, and the like) that include computer-usable program code.
[0228] This application is described with reference to the flowcharts and / or block diagrams of the method, the device (system) , and the computer program product according to this application. It should be understood that computer program instructions may be used to implement each process and / or each block in the flowcharts and / or the block diagrams and a combination of a process and / or a block in the flowcharts and / or the block diagrams. The computer program instructions may be provided for a general-purpose computer, a dedicated computer, an embedded processor, or a processor of another programmable data processing device and enable a machine to execute the instructions. When executed by any computer or the processor of a programmable data processing device, the instructions cause the apparatus to implement specific functions as described in one or more procedures in the flowcharts and / or one or more blocks in the block diagrams. The computer program instructions may alternatively be stored in a computer-readable memory that can indicate a computer or another programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate an artifact that includes an instruction apparatus. The instruction apparatus implements a specific function in one or more procedures in the flowcharts and / or one or more blocks in the block diagrams.
[0229] The computer program instructions may alternatively be loaded onto a computer or another programmable data processing device, so that a series of operations and steps are performed on the computer or the other programmable data processing device, so that computer-implemented processing is generated. Therefore, the instructions executed on the computer or on another programmable device provide steps for implementing specific functions as described in one or more procedures in the flowcharts and / or one or more blocks in the block diagrams.
[0230] It is clear that a person skilled in the art can make various modifications and variations to this application without departing from the scope of this disclosure. This disclosure is intended to cover these modifications and variations of this application provided that they fall within the scope of protection defined by the following claims and their equivalent technologies.
Claims
A method comprising:transmitting, by a first device, a first message to a second device, the first message indicating a first channel prediction model for the second device to use, wherein the first channel prediction model is selected based on at least one parameter associated with the second device; andtransmitting, by the first device, a second message to the second device, wherein the second message indicates a second channel prediction model, wherein the second channel prediction model is selected based on a change to the at least one parameter associated with the second device.The method of claim 1, wherein the first channel prediction model and the second channel prediction model are selected from a plurality of channel prediction models.The method of claim 1 or claim 2, wherein the at least one parameter comprises at least one of:an environmental condition;a resource availability;a power condition; ora prediction score.The method of any one of claims 1-3, wherein transmitting the first message comprises transmitting at least one model, wherein the at least one model comprises at least one of a first model or a second model.The method of claim 4, wherein each of the first channel prediction model and the second channel prediction model comprises the first model and the second model.The method of claim 4 or claim 5, further comprising:transmitting configuration information, wherein the configuration information comprises a format for a device state, a format for a representation of channel state information, and at least one model parameter associated with a channel prediction model.The method of claim 6, wherein the at least one model parameter comprises at least one of a number of layers of the channel prediction model, dimensions associated with the layers of the channel prediction model, or number of weights corresponding to at least one layer of the channel prediction model.The method of any one of claims 1-7, wherein transmitting the first message comprises:receiving a capability message from the second device, wherein the capability message indicates a capability of the second device to execute the first channel prediction model;transmitting a resource configuration to the second device, wherein the resource configuration indicates a downlink resource for transmission of the first channel prediction model; andtransmitting the first message to the second device in the downlink resource.The method of any one of claims 1-8, wherein the first message comprises an index associated with the first channel prediction model, wherein the index is identified using a predefined mapping table.The method of claim 6, further comprising:receiving the device state and the representation of the channel state information from the second device;obtaining, using the first model, an embedded device state based on the device state;obtaining, using the second model, at least one channel state prediction for a time window based on the embedded device state and the representation of the channel state information; andobtaining, using a third model, a prediction quality evaluation based on the device state and the at least one channel state prediction.The method of claim 10, further comprising:transmitting an update request to the second device based on the prediction quality evaluation, wherein the update request comprises a request to update at least one of:the device state;the representation of the channel state information; orthe first channel prediction model.The method of claim 10 or claim 11, further comprising:receiving an update request from the second device, wherein the update request comprises a request to update at least one of:the device state; orthe representation of the channel state information.A method comprising:receiving, at a first device, a first message from a second device, the first message indicating a first channel prediction model for the first device to use, wherein the first channel prediction model is selected based on at least one parameter associated with the first device; andreceiving, at the first device, a second message from the second device, wherein the second message indicates a second channel prediction model, wherein the second channel prediction model is selected based on a change to the at least one parameter associated with the first device.The method of claim 13, wherein the first channel prediction model and the second channel prediction model are selected from a plurality of channel prediction models.The method of claim 13 or claim 14, wherein the at least one parameter comprises at least one of:an environmental condition;a resource availability;a power condition; ora prediction score.The method of any one of claims 13-15, wherein receiving the first message comprises receiving at least one model, wherein the at least one model comprises at least one of a first model and a second model.The method of claim 16, further comprising:receiving configuration information, wherein the configuration information comprises a format for a device state, a format for a representation of channel state information, and at least one model parameter associated with a channel prediction model.The method of claim 17, wherein the at least one model parameter comprises at least one of a number of layers of the channel prediction model, dimensions associated with the layers of the channel prediction model, or number of weights corresponding to at least one layer of the channel prediction model.The method of any one of claims 13-18, further comprising:transmitting a capability message to the second device, wherein the capability message indicates a capability of the first device to execute the first channel prediction model;receiving a resource configuration from the second device, wherein the resource configuration indicates a downlink resource for transmission of the first channel prediction model; andreceiving the first message from the second device in the downlink resource.The method of any one of claims 13-19, wherein the first message comprises an index associated with the first channel prediction model, wherein the index is identified using a predefined mapping table.The method of claim 17, further comprising:transmitting the device state and the representation of the channel state information to the second device;obtaining, using the first model, an embedded device state based on the device state;obtaining, using the second model, at least one channel state prediction for a time window based on the embedded device state and the representation of the channel state information;storing the at least one channel state prediction; andobtaining one or more prediction scores based on a difference between the at least one channel state prediction and a measured channel state information.The method of claim 21, further comprising:transmitting an update request to the second device based on the one or more prediction scores, wherein the update request comprises a request to update at least one of:the device state; orthe representation of the channel state information.An apparatus, configured to perform the method of any one of claims 1-12 or any one of claims 13-22.An apparatus comprising:a transmitting unit configured to:transmit a first message to a second device, the first message indicating a first channel prediction model for the second device to use, wherein the first channel prediction model is selected based on at least one parameter associated with the second device; andtransmit a second message to the second device, wherein the second message indicates a second channel prediction model, wherein the second channel prediction model is selected based on a change to the at least one parameter associated with the second device.An apparatus comprising:a receiving unit configured to:receive a first message from a second device, the first message indicating a first channel prediction model for the apparatus to use, wherein the first channel prediction model is selected based on at least one parameter associated with the apparatus; andreceive a second message from the second device, wherein the second message indicates a second channel prediction model, wherein the second channel prediction model is selected based on a change to the at least one parameter associated with the apparatus.An apparatus comprising:one or more processors; andan interface circuit configured to:transmit a first message to a second device, the first message indicating a first channel prediction model for the second device to use, wherein the first channel prediction model is selected based on at least one parameter associated with the second device; andtransmit a second message to the second device, wherein the second message indicates a second channel prediction model, wherein the second channel prediction model is selected based on a change to the at least one parameter associated with the second device.An apparatus comprising:one or more processors; andan interface circuit configured to:receive a first message from a second device, the first message indicating a first channel prediction model for the apparatus to use, wherein the first channel prediction model is selected based on at least one parameter associated with the apparatus; andreceive a second message from the second device, wherein the second message indicates a second channel prediction model, wherein the second channel prediction model is selected based on a change to the at least one parameter associated with the apparatus.The apparatus of claim 26 or claim 27, wherein the interface circuit comprises one or more transceivers.An apparatus comprising:one or more processors; andone or more memories storing instructions which, when executed by the one or more processors, cause the apparatus to perform the method of any one of claims 1-12 or any one of claims 13-22.A communication system, wherein the communication system comprises a first apparatus configured to perform the method of any one of claims 1-12 and a second apparatus configured to perform the method of any one of claims 13-22.A non-transitory computer-readable storage medium having instructions stored thereon which, when executed by an apparatus, cause the apparatus to perform the method of any one of claims 1-12 or any one of claims 13-22.A computer program product storing instructions which, when executed, cause an apparatus to perform the method of any one of claims 1-12 or any one of claims 13-22.
Citation Information
Patent Citations
Channel prediction method and device
CN115396055A
Method for determining channel prediction model and method for indicating monitoring configuration information
CN118631367A
Devices, methods and apparatuses for channel prediction
US20240129756A1