Apparatus, method and computer program

Abstract

An apparatus comprising: means for determining a first value for channel state information, the first value being based on a channel and interference between a user equipment and a base station; means for determining a code based on a first value for the channel state information, said code being generated by an encoder; means for determining a second value for channel state information by removing at least a component of interference from the first value; and means for causing one or more of the second value or the code to be provided to the base station.

Description

Technical Field

[0001] This disclosure generally relates to apparatus, methods and computer programs in communication systems, and particularly, but not exclusively, to apparatus, methods and computer programs relating to channel state information. Background Technology

[0002] A communication system can be viewed as a facility that enables communication between two or more communication devices or provides communication devices with access to a data network.

[0003] A communication system can be a wireless communication system. Examples of wireless communication systems include mobile systems, satellite communication systems, and Wi-Fi communication systems that operate based on radio access technology standards provided by 3GPP (3rd Generation Partnership Project) or ETSI (European Telecommunications Standards Institute). Wireless communication systems operating based on radio access technologies have coverage areas that can typically be divided into cells and are therefore often referred to as cellular systems.

[0004] Radio access technology standards specify what various entities in a communication system are allowed to do and how they should be implemented. For example, they define the communication protocols and / or parameters used by communication equipment to access or connect to a radio access network.

[0005] An example of the standard is the so-called 5G (fifth generation) standard provided by 3GPP. Summary of the Invention

[0006] According to a first aspect, an apparatus is provided, comprising: means for determining a first value for channel state information, the first value being based on a channel and interference between a user equipment and a base station; means for determining an encoding based on the first value for the channel state information, the encoding being generated by an encoder; means for determining a second value for the channel state information by removing at least the interference component from the first value; and means for providing one or more of the second value or the encoding to the base station.

[0007] One or more components of the interference can be the result of one or more of the following: inter-carrier interference; inter-symbol interference; co-channel interference; or cross-link interference.

[0008] The device may include components for receiving reference signals from a base station.

[0009] The transmission power of the reference signal received from the base station can depend on the signal-to-interference-plus-noise ratio (SIR) information associated with the channel between the user equipment and the base station.

[0010] The device may include components for receiving reference signals of multiple different patterns from a base station.

[0011] The device may include components for receiving from a base station one or more instructions associated with one or more patterns of different patterns used for a reference signal.

[0012] The device may include components for determining a matrix of multiple first values ​​using a first pattern among the different patterns.

[0013] The device may include components for determining one or more interference values ​​associated with one or more components of the interference using one or more other patterns among the different received signal patterns.

[0014] Two or more different patterns can be used to determine the interference values ​​associated with different components of the interference.

[0015] The device can be provided in a communication device or can be a communication device.

[0016] According to another aspect, a method is provided, the method comprising: determining a first value for channel state information, the first value being based on a channel between a user equipment and a base station and interference; determining an encoding based on the first value for the channel state information, the encoding being generated by an encoder; determining a second value for the channel state information by removing at least the interference component from the first value; and causing one or more of the second value or the encoding to be provided to the base station.

[0017] One or more components of the interference can be the result of one or more of the following: inter-carrier interference; inter-symbol interference; co-channel interference; or cross-link interference.

[0018] The method may include receiving a reference signal from a base station.

[0019] The transmission power of the received reference signal can depend on the signal-to-interference-plus-noise ratio (SIR) information associated with the channel between the user equipment and the base station.

[0020] The method may include receiving multiple reference signals with different patterns from a base station.

[0021] The method may include receiving from a base station one or more instructions associated with one or more different patterns used as a reference signal.

[0022] The method may include using a first pattern from the different patterns to determine a matrix of multiple first values.

[0023] The method may include using one or more other patterns from the different received signal patterns to determine one or more interference values ​​associated with one or more components of the interference.

[0024] Two or more different patterns can be used to determine the interference values ​​associated with different components of the interference.

[0025] This method can be performed by a device. The device can be provided in a communication device or can be a communication device itself.

[0026] According to another aspect, an apparatus is provided, the apparatus comprising at least one processor and at least one memory storing instructions, the instructions, when executed by the at least one processor, causing the apparatus to at least: determine a first value for channel state information, the first value being based on a channel between a user equipment and a base station and interference; determine an encoding based on the first value for the channel state information, the encoding being generated by an encoder; determine a second value for the channel state information by removing at least an interference component from the first value; and cause one or more of the second value or the encoding to be provided to the base station.

[0027] One or more components of the interference can be the result of one or more of the following: inter-carrier interference; inter-symbol interference; co-channel interference; or cross-link interference.

[0028] This device can be used to receive reference signals from a base station.

[0029] The transmission power of the reference signal received from the base station can depend on the signal-to-interference-plus-noise ratio (SIR) information associated with the channel between the user equipment and the base station.

[0030] This device can enable the reception of multiple reference signals with different patterns from a base station.

[0031] The device can be configured to receive from a base station one or more instructions associated with one or more patterns of different patterns used as a reference signal.

[0032] The device can be used to determine a matrix of multiple first values ​​using the first pattern among the different patterns.

[0033] The device can be made to use one or more other patterns among the different received signal patterns to determine one or more interference values ​​associated with one or more components of the interference.

[0034] Two or more different patterns can be used to determine the interference values ​​associated with different components of the interference.

[0035] The device can be provided in a communication device or can be a communication device.

[0036] According to another aspect, a method is provided, the method comprising: receiving from a user equipment one or more of the following: a) encoding, said encoding being determined by an encoder based on a first value of channel state information for a channel between the user equipment and a base station and interference; or b) a second value of the channel state information, the second value being determined by removing at least a component of interference from the first value.

[0037] The method may include training a decoder such that the decoder receives the encoding as input and provides the second value as output.

[0038] One or more components of the interference can be the result of one or more of the following: inter-carrier interference; inter-symbol interference; co-channel interference; or cross-link interference.

[0039] This method may include sending a reference signal.

[0040] The transmission power of the reference signal can depend on the signal-to-interference-plus-noise ratio (SIR) information associated with the channel between the user equipment and the base station.

[0041] This method may include sending multiple reference signals with different patterns.

[0042] The method may include sending one or more instructions associated with one or more patterns in different patterns used for a reference signal.

[0043] The first pattern among the different patterns can be used to determine a matrix of multiple first values.

[0044] One or more of the different received signal patterns can be used to determine one or more interference values ​​associated with one or more components of the interference.

[0045] Two or more different patterns can be used to determine the interference values ​​associated with different components of the interference.

[0046] This method can be performed by a device. This device can be provided in a base station.

[0047] According to another aspect, an apparatus is provided, the apparatus comprising: a component for receiving one or more of the following from a user equipment: a) encoding, said encoding being determined by an encoder based on a first value of channel state information for a channel between the user equipment and a base station and interference; or b) a second value of the channel state information, said second value being determined by removing at least a component of interference from the first value.

[0048] The apparatus may include components for training a decoder such that when the decoder receives the encoding as input, the decoder provides the second value as output.

[0049] One or more components of the interference can be the result of one or more of the following: inter-carrier interference; inter-symbol interference; co-channel interference; or cross-link interference.

[0050] The device may include components for transmitting reference signals.

[0051] The transmission power of the reference signal can depend on the signal-to-interference-plus-noise ratio (SIR) information associated with the channel between the user equipment and the base station.

[0052] The device may include components for transmitting reference signals of multiple different patterns.

[0053] The device may include components for transmitting one or more instructions associated with one or more patterns in different patterns used as reference signals.

[0054] The first pattern among the different patterns can be used to determine a matrix of multiple first values.

[0055] One or more of the different received signal patterns can be used to determine one or more interference values ​​associated with one or more components of the interference.

[0056] Two or more different patterns can be used to determine the interference values ​​associated with different components of the interference.

[0057] The device can be provided in base stations.

[0058] According to another aspect, an apparatus is provided, the apparatus comprising at least one processor and at least one memory storing instructions, which, when executed by the at least one processor, cause the apparatus to at least: receive from a user equipment one or more of the following: a) encoding, said encoding being determined by an encoder based on a first value of channel state information for a channel between the user equipment and a base station and interference; or b) a second value of the channel state information, said second value being determined by removing at least a component of interference from the first value.

[0059] The device can be trained such that when the decoder receives the encoding as input, the decoder provides the second value as output.

[0060] One or more components of the interference can be the result of one or more of the following: inter-carrier interference; inter-symbol interference; co-channel interference; or cross-link interference.

[0061] The device can be made to send a reference signal.

[0062] The transmission power of the reference signal can depend on the signal-to-interference-plus-noise ratio (SIR) information associated with the channel between the user equipment and the base station.

[0063] This device can be used to send reference signals with multiple different patterns.

[0064] The device can be configured to send one or more instructions associated with one or more patterns in different patterns used as a reference signal.

[0065] The first pattern among the different patterns can be used to determine a matrix of multiple first values.

[0066] One or more of the different received signal patterns can be used to determine one or more interference values ​​associated with one or more components of the interference.

[0067] Two or more different patterns can be used to determine the interference values ​​associated with different components of the interference.

[0068] The device can be provided in base stations.

[0069] The device may include at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the device to perform any prior method.

[0070] The device may include a circuit system that enables the device to perform any of the prior methods.

[0071] According to another aspect, a computer program including instructions is provided that, when executed by a device, causes the device to perform any of the aforementioned methods.

[0072] According to another aspect, a computer program including instructions is provided that, when executed, cause any of the aforementioned methods to be performed.

[0073] According to one aspect, a computer program including computer-executable code is provided, which causes any of the aforementioned methods to be performed.

[0074] According to one aspect, a computer-readable medium is provided, comprising program instructions stored thereon for performing at least one of the methods described above.

[0075] According to one aspect, a non-transient computer-readable medium is provided that includes program instructions, which, when executed by a device, cause the device to perform any of the aforementioned methods.

[0076] According to one aspect, a non-transient computer-readable medium is provided comprising program instructions that, when executed, cause any of the aforementioned methods to be performed.

[0077] According to one aspect, a non-volatile physical memory medium is provided, comprising program instructions stored thereon for performing at least one of the methods described above.

[0078] The preceding text has described many different aspects. It should be understood that other aspects can be provided through any combination of two or more of the aspects described above.

[0079] Various other aspects are described in the following detailed description and the appended claims. Attached Figure Description

[0080] Some examples will now be described in more detail, with reference to the accompanying drawings, which are illustrated only in the figures:

[0081] Figure 1 This shows a schematic representation of a 5G system;

[0082] Figure 2 A schematic representation of the device is shown;

[0083] Figure 3 A schematic representation of the user equipment is shown;

[0084] Figure 4 This schematically illustrates the propagation of interference effects from the user equipment side to the base station's codebook selection.

[0085] Figure 5 An example CSI-RS (Channel State Information Reference Signal) pattern for separating interference from the desired channel is shown;

[0086] Figure 6 A denoising autoencoder for purifying colored channel state information is illustrated schematically in some embodiments;

[0087] Figure 7 The signal flow of some embodiments is shown;

[0088] Figure 8 Methods of some embodiments are shown; and

[0089] Figure 9 Another method of some embodiments is shown. Detailed Implementation

[0090] In the following explanation, certain embodiments are described with reference to communication devices capable of communicating via wireless cellular systems and mobile communication systems serving such communication devices. Before detailed explanation, refer to... Figure 1 , Figure 2 and Figure 3 A brief explanation of the example embodiments, some general principles of wireless communication systems, their access systems, and communication devices is provided to help understand the technical background of the described examples.

[0091] Figure 1 This diagram illustrates a schematic representation of a communication system operating based on fifth-generation radio access technology (commonly referred to as a 5G system (5GS)). The 5GS may include a (radio) access network (AN), a 5G core network (5GC), one or more application functions (AF), and one or more data networks (DN). User equipment can access or connect to one or more DNs via the 5GS.

[0092] An AN may include one or more base stations or radio access network (RAN) nodes, such as gNodeB (gNB). A base station or RAN node may include one or more distributed units connected to a central unit.

[0093] 5GC can include various network functions such as Access and Mobility Management (AMF), Session Management (SMF), Authentication Server (AUSF), User Data Management (UDM), User Plane (UPF), Network Data Repository, Network Exposure (NEF), Service Communication Agent (SCP), Edge Application Server Discovery (EASDF), Policy Control (PCF), Network Slice Access Control (NSACF), Network Slice Specific Authentication and Authorization (NSSAAF), and / or Network Slice Selection (NSSF).

[0094] Figure 2 An example of apparatus 200 is shown. Apparatus 200 may be provided in user equipment and / or base stations. Apparatus 200 may have at least one processor and at least one memory storing instructions for one or more functions, which, when executed by at least one of the at least one processor, cause the operation or action of one or more functions to be performed. In this example, the apparatus may include at least one random access memory (RAM) 211a and / or at least one read-only memory (ROM) 211b. The apparatus may include at least one processor 212, 213 and / or an input / output interface 214. At least one processor 212, 213 may be coupled to at least one memory, which in this example is RAM 211a and ROM 211b. At least one processor 212, 213 may be configured to execute appropriate software code 215. Software code 215 may, for example, allow the apparatus to perform one or more steps of this aspect.

[0095] Figure 3 An example of a communication device 300 is shown. The communication device 300 can be any device capable of or configured to transmit and receive wireless signals (including radio signals). Other non-limiting examples of the communication device 300 include a mobile station (MS) or mobile device such as a mobile phone or referred to as a "smartphone," a computer equipped with a wireless interface card or other wireless interface device (e.g., a USB dongle), a personal digital assistant (PDA) or tablet computer, a machine-type communication (MTC) device, a cellular Internet of Things (CIoT) device, or any combination of these devices.

[0096] The communication device 300 can be configured to transmit or receive wireless signals, such as radio signals carrying communication. The communication can be one or more of the following: voice, email, text message, multimedia, data, machine data, etc.

[0097] The communication device 300 can be configured to transmit and / or receive radio signals via an air or radio interface 307 through a transceiver unit 306. The transceiver unit 306 may include a radio section and an associated antenna arrangement. The radio section may convert baseband signals to radio frequency, and / or vice versa. The antenna arrangement may be located inside or outside the mobile device and may include a single antenna or multiple antennas. The antenna arrangement may be an antenna array including multiple antenna elements.

[0098] The communication device 300 may include at least one processor 301 and / or at least one memory. The at least one memory may be at least one ROM 302a and / or at least one RAM 302b. Other possible components 303 may be provided for the execution of software and hardware-assisted tasks designed to perform, including control of access to access networks such as 5G RAN and other communication devices, and communication with the access networks. At least one processor 301 is coupled to RAM 302b and ROM 302a. The at least one processor 301 may be configured to execute instructions of software code 308. Execution of the instructions of software code 308 may, for example, allow the communication device 300 to perform one or more operations. Software code 308 may be stored in ROM 302a. It should be understood that in other embodiments, any other suitable memory may be used alternatively or additionally.

[0099] At least one processor 301, at least one ROM 302a, and / or at least one RAM 302b may be provided on a suitable circuit board, in an integrated circuit, and / or in a chipset. This feature is indicated by reference numeral 304.

[0100] The communication device 300 may optionally have a user interface, such as a keypad 305, a touch-sensitive screen or touchpad, or a combination thereof. Optionally, the communication device may have one or more of a display, a speaker, and a microphone.

[0101] In the following examples, the term UE or user equipment is used. This term includes any examples of the previously discussed communication device 300 and / or any other communication device.

[0102] Currently, radio access technologies standardized by 3GPP are commonly referred to as 5G or NR. Other radio access technologies standardized by 3GPP include Long Term Evolution (LTE) or LTE Advanced Professional Edition of the Universal Mobile Telecommunications System (UMTS).

[0103] Wireless communication systems typically include access networks, such as radio access networks that operate based on radio access technologies that include base stations or radio access network nodes.

[0104] Wireless communication systems may alternatively or additionally include other types of access networks, such as wireless local area networks (WLANs) and / or WiMAX (Global Microwave Access Interoperability) networks.

[0105] It should be understood that the example embodiments can also be used with standards for future radio access technologies such as 6G and above.

[0106] For Frequency Division Duplex (FDD) schemes, the base station (BS) uses downlink Channel State Information (CSI) to acquire channel response and precoding for beamforming in the downlink of massive MIMO systems. The BS can use accurate CSI to achieve higher signal-to-noise ratio (SNR) and channel capacity. However, in some networks such as FDD networks, the UE estimates the downlink CSI. Therefore, the estimated CSI needs to be shared with the BS, which incurs network overhead.

[0107] To reduce overhead, several vector quantization methods have been proposed to generate codebooks that achieve relatively high compression ratios (CR, i.e., the ratio of compressed size to uncompressed size). CR can be a scalar in the range (0, 1], and a lower CR implies more compression. These methods may not provide sufficient accuracy for CSI reconstruction with limited feedback overhead.

[0108] Compressed sensing (CS)-based methods can be used to compress CSI based on its spatial and temporal correlations. However, the channel may not be sparse, which is an assumption of CS-based methods. Furthermore, CS-based methods use random projections, which may lead to some reconstruction loss in the base station (BS).

[0109] Since the problem involves compression and reconstruction, autoencoder (AE) architectures from deep learning (DL) methods can be applied to CSI feedback applications. A CsiNet architecture using an encoder and decoder to accomplish the task has been proposed. CsiNet was proposed by Wen, C., Shih, W., and Jin, S. in Deep Learning for CSI Feedback in Massive MIMO, IEEE Wireless Communications Letters, Vol. 7, 2018, pp. 748-751. Further work on this architecture has been done by others in the art.

[0110] CsiNet consists of convolutional layers, dense layers, and a refined network structure. The encoder generates a compressed representation of the input CSI, and the decoder reconstructs the CSI based on the compressed information. Through joint training of the encoder and decoder, CsiNet learns the channel structure and can provide more accurate reconstructions compared to CS-based methods.

[0111] A modified version of the CsiNet architecture has been proposed and is called CsiNet+. This modifies the kernel size and reconstruction blocks.

[0112] A CSI feedback method based on a fully convolutional neural network (FullyConv) is proposed, which improves the quality of reconstructed CSI and reduces the number of trainable parameters and computational resources compared with existing methods.

[0113] CsiNet, CsiNet+, and FullyConv methods were designed for fixed CR.

[0114] To reduce feedback payload, the encoder output is the subsequently quantized code. The UE transmits this code over the air, and the gNB receives it. The dequantizer uses the received code to construct the decoder input. Finally, the decoder processes the compressed signal and reconstructs the complete CSI.

[0115] There are various training strategies.

[0116] In CSI compression using two-sided model use cases, the following AI / ML models can be trained collaboratively:

[0117] Jointly train the two-sided model at a single-sided / entity location, such as on the UE side or the network side.

[0118] The two-sided model is jointly trained on both the network side and the UE side.

[0119] Training is performed on both the network side and the UE side, with the UE-side CSI generation part and the network-side CSI reconstruction part being trained by the UE side and the network side respectively.

[0120] Joint training may mean that the generative model and the reconstructed model should be trained in the same loop used for forward and backward propagation. Joint training can be performed at a single node and / or across multiple nodes. For example, in the case of multiple nodes, this can be done through gradient exchange between nodes.

[0121] Individual training can include sequential training starting from the UE side, or sequential training starting from the NW side, or parallel training on both the UE and NW sides.

[0122] There are various possibilities for generalizing ML models. As examples, some different cases are illustrated below. One or more of the following cases can be considered to validate the generalization performance of AI / ML models across various scenarios / configurations.

[0123] Case 1: Train an AI / ML model on a training dataset from a scenario #A / configuration #A, and then perform inference / testing on a dataset from the same scenario #A / configuration #A.

[0124] Case 2: Train an AI / ML model on a training dataset from a scenario #A / configuration #A, and then perform inference / testing on datasets different from scenario #A / configuration #A (e.g., scenario #B / configuration #B, scenario #A / configuration #B).

[0125] Case 3: Train an AI / ML model on a training dataset constructed by mixing datasets from multiple scenarios / configurations including scenario #A / configuration #A and one or more datasets different from scenario #A / configuration #A (e.g., scenario #B / configuration #B, scenario #A / configuration #B). Then, the AI / ML model performs inference / testing on datasets from a single scenario / configuration (e.g., scenario #A / configuration #A, scenario #B / configuration #B, scenario #A / configuration #B) from multiple scenarios / configurations.

[0126] Case 2A: Train an AI / ML model on a training dataset from a scenario #A / configuration #A, then update the AI / ML model on a fine-tuned dataset that is different from scenario #A / configuration #A (e.g., scenario #B / configuration #B, scenario #A / configuration #B). Subsequently, test the AI / ML model on datasets that are different from scenario #A / configuration #A (e.g., constrained by scenario #B / configuration #B, scenario #A / configuration #B).

[0127] It should be noted that these cases are merely examples, and one or more other cases may be used. Not all cases are required.

[0128] For CSI enhancement evaluation, in order to verify the generalization performance of AI / ML models in various scenarios, a set of scenarios focusing on one or more of the following aspects can be considered as a starting point:

[0129] Various deployment scenarios (e.g., UMa (city macro station), UMi (city micro station) and / or InH (indoor hotspot));

[0130] For various outdoor / indoor UE distributions for UMa / UMi (e.g., 10:0, 8:2, 5:5, 2:8, 0:10); and / or

[0131] Various carrier frequencies (e.g., 2GHz, 3.5GHz).

[0132] Other aspects can be considered in one or more scenarios. For example, one or more of the following may be considered alternatively or additionally: various antenna spacings; various antenna virtualizations (TxRU mapping - TxRU is the interface between the baseband unit and the radio distribution network); various ISDs (inter-site distances); various UE speeds; and / or the like.

[0133] For CSI enhancement evaluation, in order to verify the generalization / scalability performance of AI / ML models across various configurations (e.g., which may potentially lead to different dimensions of model input / output), a set of configurations focusing on one or more of the following aspects can be used as a starting point:

[0134] Various bandwidths (e.g., 10MHz, 20MHz) and / or frequency granularity (e.g., subband size);

[0135] CSI feedback payloads of various sizes, and / or the number of candidate payloads;

[0136] Various antenna port layouts (e.g., (N1 / N2 / P)), and / or the number of antenna ports (e.g., 32 ports, 16 ports); and / or

[0137] Other configurations include various digital technologies, various ranks / layers, etc.

[0138] CSI-RS is used to estimate CSI. CSI describes the channel response in the frequency and spatial domains. The channel estimator used to retrieve the response can assume:

[0139] The received subcarriers are orthogonal, meaning there is no inter-carrier interference (ICI).

[0140] The maximum channel delay spread is shorter than the duration of the CP (cyclic prefix), i.e., there is no inter-symbol interference (ISI); and / or

[0141] CSI-RS is not interfered with by simultaneous DL or UL transmissions (no co-channel interference (CI) and no cross-link interference (CLI)).

[0142] However, in reality, ICI can occur whenever the UE moves at high speed. ISI may occur in rural areas (see the hilly terrain channel model). CI can occur when data and location services are triggered simultaneously and are not coordinated (this is the default operation). CLI can occur in flexible duplex scenarios.

[0143] The received CSI-RS may be contaminated by self-link interference or co-link interference / cross-link interference and / or additive white Gaussian noise. The channel response h estimated using a non-interference-aware receiver absorbs errors caused by various interference and noise sources. These errors in h are transformed in the colored (i.e., interference) CSI matrix sent to the CSI autoencoder. The decoder side of the autoencoder will reconstruct the colored CSI that inherently degrades codebook selection. At this point, refer to Figure 4 On UE 400, as shown by reference numeral 402, the CSI-RS is disrupted by self-interference or co-channel interference. The disrupted CSI-RS is provided to the non-interference-aware estimator 404. As shown by reference numeral 406, the incorrect channel response is estimated. As shown by reference numeral 408, the CSI matrix is ​​constructed. The output CSI matrix is ​​a colored CSI matrix, as shown by reference numeral 410. This colored matrix is ​​encoded by reference numeral 412 to provide codeword selection.

[0144] On gNB 416, the received codeword is decoded as shown in Figure 418. This results in the reconstruction of the colored CSI dataset as shown in Figure 420. The codebook selection is based on the reconstructed CSI dataset, as shown in Figure 422. This leads to an incorrect codebook selection, as shown in Figure 424.

[0145] To ensure accurate codebook selection in the base station, it is desirable for the decoder to reconstruct a clean version of CSI.

[0146] To ensure that the autoencoder architecture is unaware of the severity of noise and / or self-interference and / or co-link / cross-link interference, the decoder is expected to reconstruct a clean version of CSI.

[0147] In some embodiments, a CSI noise reduction framework can be provided. Within this framework, a decoder can reconstruct a clean CSI matrix. The decoder can be an AI / ML decoder. A clean CSI matrix can be reconstructed regardless of the severity of self-interference and / or co-channel interference.

[0148] In some embodiments, a trained CSI-RS pattern may be optionally used, which enables the UE to label colored CSI using both the encoding generated by the encoder and the clean CSI.

[0149] To achieve a clean CSI reconstruction, some implementations can use the following training process.

[0150] The gNB can obtain information about the UE. This information can be any suitable information. For example, it can include the UE's speed status. It can include CI and CLI conditions. It can include the maximum excess latency of the channel.

[0151] It should be understood that the base station decoder can be trained using training data and / or simulated data provided by different UEs.

[0152] Using information about the UE, the gNB can design CSI-RS patterns. Training-specific CSI-RS can be provided at the gNB using non-ML methods. These CSI-RS are used to extract colored CSI and clean CSI at the encoder.

[0153] In this regard, refer to Figure 5 , Figure 5 An example of a training-specific CSI-RSI pattern to be used by the UE is shown. Examples of training-specific CSI-RSI patterns include one or more, or all of the following:

[0154] The first pattern portion 500 of the CSI-RS includes interference;

[0155] The second pattern section 502 involves punching holes in the CSI-RS in the frequency domain to isolate the ICI effect. The hole size can be proportional to the maximum Doppler shift. Therefore, the CSI-RS can be silenced on some resources, allowing self-interference (leakage from active subcarriers) to be observed and ICI to be estimated. UE velocity can be used to configure the comb pattern in the frequency domain that the UE can be used for to isolate the ICI effect. This UE velocity information can be obtained by the gNB using recent or recent data / control signal exchanges by estimating the maximum Doppler shift associated with the signal exchanges.

[0156] The third pattern section 504 involves punching the CSI-RS in the time domain to isolate ISI effects. The punch size can be proportional to the duration of the channel tail beyond the CP duration. The channel tail is the superposition of all multipath components arriving at the UE with a delay greater than the CP duration. This can be based on the maximum excess delay of the channel. This information can be used to configure a silence pattern in the time domain, which the UE can use to isolate ISI effects. This information can be cell-specific and / or beam-specific; and / or

[0157] The fourth pattern section 506 involves punching holes in the CSI-RS in both the time and frequency domains to separate co-channel interference and cross-link interference. This can be achieved using information about concurrent UL / DL transmissions. This information is used to configure a silence pattern in the time-frequency domain to separate CI and CLI effects. This information can be obtained from the neighboring gNB via an explicit request procedure. This can be done via the X2 / Xn interface or any other gNB-to-gNB interface.

[0158] The gNB can indicate the selected pattern and / or pattern effect to the UE (i.e., separate ICI, ISI, CI, CLI).

[0159] gNB can instruct the UE to use the first pattern to calculate the colored CSI matrix (called colored CSI).

[0160] The gNB can instruct the UE to use one or more of the second, third, and fourth patterns. This is for estimating ICI, ISI, CI, and / or CLI.

[0161] gNB can instruct the UE to use the colored CSI matrix and the estimated ICI, ISI, CI and / or CLI to calculate the pure CSI matrix (referred to as pure CSI).

[0162] The SNR of the signal received by the UE can be determined. This information can be used to adapt the transmit power of the reference signal at the gNB. Adaptive transmit power can help estimate the clean CSI under high SNR.

[0163] The following can be provided as a denoising autoencoder that uses colored CSI input and provides clean CSI output.

[0164] The UE can apply instructions from the gNB. The UE can first determine the colored CSI, and from there determine the pure CSI.

[0165] The UE can use the colored CSI obtained by the UE as input to the encoder and generate the encoded C.

[0166] The UE prepares a training set to share with the gNB. The UE can generate tuples: {colored CSI, C, clean CSI} for all observations. In some embodiments, these samples can be used for further model tuning.

[0167] The UE transmits a subset of training data to the gNB, wherein the subset includes the following pair: {C, pure CSI}. In some embodiments, the UE may transmit all tuples.

[0168] gNB uses the received pairs to train a matching decoder, which takes the encoded C as input and outputs clean CSI.

[0169] Therefore, CSI autoencoders can be denoised. For example... Figure 6 As schematically illustrated, during training, the colored CSI 600 acquired by the UE is used as input to the encoder 602 to produce the encoded C as shown by reference numeral 604. The clean CSI, as a label of the encoded C, is sent to the decoder 606 of the gNB. In other words, the gNB receives (encoded C, clean CSI) and trains the decoder using input = C, label = clean CSI. By doing so, the CSI autoencoder has denoising capability when the decoder outputs the clean CSI 608.

[0170] This can be advantageous during the inference phase (i.e., after deployment). When the CSI-RS is interfered with in the field, the UE can use its non-interference-aware CSI estimator and calculate the colored CSI, which is then encoded in the transmission to the gNB, knowing that the gNB decoder will implicitly purify the encoding from all interference effects and output the pure CSI.

[0171] The channel estimator of the NR modem can assume that there is no ICI, ISI, CI, or CLI contamination of the CSI-RS reception. In other words, the estimator assumes that the received signal y(a, j, k) at antenna a, symbol j, and subcarrier k contains only the contribution of the transmitted signal x(j, k) at symbol j and subcarrier k multiplied by the channel response h(a, j, k) and corrupted by AWGN (additive white Gaussian noise) n(a, j, k): y(a, j, k) = h(a, j, k)x(j, k) + n(a, j, k), ∀a = 1:A, j = 1:J, k = 1:K Model (1).

[0172] The estimator can use the above Model (1) to reconstruct h and construct the CSI matrix.

[0173] However, there may be some cases where Model (1) does not apply.

[0174] For example, at high speeds, the orthogonality of the subcarriers is lost. The ICI phenomenon (the first type of self-interference) occurs.

[0175] Then, instead of Model (1), the received signal at subcarrier k contains the contribution of the TX signal at all subcarriers (i.e., k), and d ≠ k. , .

[0176] In a second example, when the UE experiences a channel with a delay longer than the CP (e.g., in a hilly area), the received signal at symbol j is contaminated by ISI (the second type of self-interference), i.e., the signal transmitted during symbol i < j contaminates the current symbol j reception: , .

[0177] In a third example, concurrent transmissions from L other cells can create CLI or CI, such that the received signal becomes the sum of the desired signal and interference signals from other gNBs and UEs. , .

[0178] In another example, the UE can experience any two or more of the conditions in Examples 1 to 3.

[0179] Besides AWGN, the received CSI-RS may be contaminated by self-interference and / or co-link / cross-link interference. In this case, the channel response h estimated using the mismatch model (1) absorbs the errors caused by various interference sources, as discussed with respect to models (2-4). An incorrect h estimate results in a colored (i.e., interfering) CSI matrix being sent to the CSI autoencoder. The decoder side of the autoencoder will reconstruct the colored CSI that causes the codebook selection to be degraded.

[0180] Training-specific CSI-RS can be used as previously discussed. In some embodiments, the gNB assists the UE in separating various interference signals based on various interference signal sources only during the training phase. Then, after each interference signal is separated, it can be estimated and removed from the received CSI-RS. By doing so, the gNB enables the UE to calculate:

[0181] Colored CSI estimation - including both useful channels and all interference; and

[0182] Pure CSI estimation - includes only useful channels.

[0183] To achieve the above, the UE uses the Training-Specific (TS) CSI-RS pattern as previously described.

[0184] refer to Figure 7 , Figure 7 Signaling flows for some embodiments are shown.

[0185] As shown in Figure 1, gNB can perform one or more of the following:

[0186] Determine the UE speed;

[0187] Determine the maximum excess delay of the channel;

[0188] Collect CI and / or CLI information from one or more neighboring gNBs; and

[0189] SNR is predicted based on the signal received from the UE.

[0190] As shown in Figure 2, the gNB can evaluate to determine one or more of ICI, ISI, CI, CLI, and SNR. This may be a situation where data received in 1 needs to be processed.

[0191] As mentioned, select patterns one through four. The pattern used depends on one or more values ​​determined in step 2. This can be as described previously.

[0192] As shown in Figure 4, the gNB configures the TS CSI-RS and transmit power to the UE. The transmit power can depend on the SNR. The gNB can additionally indicate which TS CSI-RS pattern should be used to separate each type of interference. This can be as described previously.

[0193] As shown in Figure 5, the gNB sends the TS CSI-RS.

[0194] As shown in Figure 6, the UE uses TS CSI-RS to calculate both the colored CSI and the clean CSI.

[0195] As shown in Figure 7, the UE encodes the colored CSI into encoding C.

[0196] As shown in Figure 8, the UE generates training dataset tuples (colored CSI, encoded C, pure CSI).

[0197] As shown in Figure 9, the UE sends a subset of data to the gNB, which includes the following pair: {C, Pure CSI}. Colored CSIs can also be sent to the gNB. This can be used by the gNB to determine how much interference the UE is experiencing. This can be used for future link configurations, such as adapting to the MCS (Modulation and Coding Scheme).

[0198] As shown in Figure 10, gNB uses the pair (pure CSI, C) to train the decoder, so that the decoder can implicitly learn to recover the pure CSI from the interference coding C.

[0199] Some implementations may not increase the computational complexity of the baseband modem. The channel estimator may not require modification.

[0200] Some implementations can compensate for both self-channel and co-channel interference.

[0201] Some implementations can provide accurate CSI reconstruction even when the type of interference is unknown.

[0202] In some embodiments, TS CSI-RS can be defined and configured for use in models that update AI / MLCSI feedback. In inference, standard CSI-RS can be used with ML models trained on TS-CSI-RS.

[0203] The ML model can be any suitable model, such as those previously described.

[0204] Some embodiments may include receiving configuration at the user equipment (UE) from a network entity (e.g., gNB). The configuration may include training a specific channel state indicator reference signal (TS CSI_RS) configuration, transmission power, and / or expected interference type, at least one of these.

[0205] In some embodiments, a TS-CSI-RS configuration can be applied to estimate colored CSI and clean CSI. Colored CSI may include useful channels and interference, while clean CSI includes useful channels with at least some interference removed.

[0206] In some embodiments, a training dataset tuple is generated. The training dataset tuple may include at least one of colored CSI, encoded C, or pure CSI. Encoded C is encoded by colored CSI.

[0207] In some embodiments, at least one of encoded C or pure CSI is sent to the network entity.

[0208] It should be understood that a colored or interfering CSI has been referenced. This can include one or more values. In the case of multiple values, these values ​​can include a colored CSI matrix.

[0209] It should be understood that a clean CSI has been referenced. This can include one or more values. In the case of multiple values, these values ​​can include a clean CSI matrix.

[0210] In some embodiments, the encoding may be provided by codewords.

[0211] refer to Figure 8 It illustrates methods of some embodiments.

[0212] This method can be performed by a device. The device can be in or within a communication device.

[0213] The apparatus may include suitable components, such as a circuit system for providing the method.

[0214] Alternatively or additionally, the apparatus may include at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to provide at least the following methods.

[0215] Alternatively or additionally, the device may be, for example, regarding Figure 2 or Figure 3 The apparatus under discussion.

[0216] This method can be provided by computer program code or computer executable instructions.

[0217] As shown by reference numeral A1 in the attached figure, the method may include determining a first value for channel state information, the first value being based on the channel between the user equipment and the base station, as well as interference.

[0218] As shown by reference numeral A2 in the attached figure, the method may include determining an encoding based on a first value for channel state information, the encoding being generated by an encoder;

[0219] As shown by reference numeral A3 in the attached figure, the method may include determining a second value for channel state information by removing at least interfering components from a first value; and

[0220] As shown in the attached figure, reference numeral A4, the method may include providing one or more of the second value or encoding to the base station. The first value may be provided to the base station.

[0221] It should be understood that, Figure 8 The steps can occur in any suitable order.

[0222] It should be understood that, Figure 8 The methods outlined herein can be modified to include any previously described features.

[0223] Computer program code can be downloaded and stored in one or more memories of a relevant device or equipment.

[0224] refer to Figure 9 , Figure 9 Methods of some embodiments are shown.

[0225] This method can be executed by a device. The device can be in a base station or can be a base station.

[0226] The device may include suitable components, such as a circuit system for providing the method.

[0227] Alternatively or additionally, the apparatus may include at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to provide at least the following methods.

[0228] Alternatively or additionally, the device may be, for example, regarding Figure 2 The apparatus under discussion.

[0229] This method can be provided by computer program code or computer executable instructions.

[0230] As indicated by reference numeral B1 in the figure, the method may include receiving one or more of the following from the user equipment: a) encoding, which is determined by an encoder based on a first value of channel state information for the channel between the user equipment and the base station and interference; or b) a second value of the channel state information, which is determined by removing at least the interference component from the first value.

[0231] It should be understood that, Figure 9 The methods outlined herein can be modified to include any previously described features.

[0232] Computer program code can be downloaded and stored in one or more memories of a relevant device or equipment.

[0233] Therefore, while the above references to certain example architectures for wireless networks, technologies, and standards describe certain embodiments by way of example, these embodiments can be applied to any other suitable form of communication system besides those shown and described herein. In this example, some embodiments have been described with respect to 5G networks. It should be understood that other embodiments can be provided in any other suitable network in which multicast services are provided by access nodes. Those multicast services can be provided to UEs connected to the network or in an inactive state. In an inactive state, the connection is not disconnected but rather the UE is inactive. For example, the UE's context can be preserved.

[0234] It should also be noted that although exemplary embodiments have been described above, several changes and modifications can be made to the disclosed solutions without departing from the scope of the invention.

[0235] As used herein, “at least one of the following: ” and “at least one of ” and similar wording (where a list of two or more elements is connected by “and” or “or”) means at least any one element, or at least any two or more elements, or at least all elements.

[0236] Generally, various embodiments can be implemented in hardware or special-purpose circuitry systems, software, logic, or any combination thereof. Some aspects of this disclosure can be implemented in hardware, while others can be implemented in firmware or software, which can be executed by a controller, microprocessor, or other computing device, but this disclosure is not limited thereto. Although various aspects of this disclosure may be shown and described as block diagrams, flowcharts, or using some other graphical representation, it is well understood that the blocks, apparatuses, systems, techniques, or methods described herein can be implemented as non-limiting examples in hardware, software, firmware, special-purpose circuitry or logic, general-purpose hardware or controllers or other computing devices, or combinations thereof.

[0237] As used in this application, the term "circuit system" may refer to one or more or all of the following: (a) Hardware circuit implementation only (such as implementation in analog and / or digital circuit systems only) and (b) A combination of hardware circuitry and software, such as (if applicable): (i) A combination of (multiple) analog and / or digital hardware circuits and software / firmware, and (ii) Any part of a hardware processor (including (multiple) digital signal processors), software, and (multiple) memories, which work together to enable a device such as a mobile phone or server to perform various functions. (c) A hardware circuit (multiple) and / or a processor (such as a microprocessor or a portion thereof) that requires software (e.g., firmware) to operate, but which may not exist when the software is not required to operate.

[0238] The definition of "circuit system" applies to all uses of the term in this application (including in any claim). As another example, as used in this application, the term "circuit system" also covers only hardware circuitry or a processor (or processors) or a portion of hardware circuitry or a processor and its (or their) accompanying software and / or firmware implementation. The term "circuit system" also covers (e.g., and if applicable to a particular claim element) integrated circuits, such as baseband integrated circuits or processor integrated circuits for mobile devices, or similar integrated circuits in servers, cellular network devices, or other computing or networking devices.

[0239] Embodiments of this disclosure can be implemented by computer software executable by a data processor of a mobile device (such as in a processor entity), or by hardware, or by a combination of software and hardware. Computer software or programs (also referred to as program products), including software routines, applets, and / or macros, can be stored in any device-readable data storage medium, and they include program instructions for performing specific tasks. A computer program product may include one or more computer-executable components configured to execute embodiments during program runtime. The one or more computer-executable components may be at least one piece of software code or a portion thereof.

[0240] Furthermore, it should be noted that any box in the logical flow diagram can represent a program step, or an interconnected logic circuit, a box and function, or a combination of program steps and logic circuits, boxes and functions. Software can be stored on physical media such as memory chips or memory blocks implemented within a processor, magnetic media such as hard disks or floppy disks, and optical media such as DVDs and their data variants, CDs. Physical media are non-transient media.

[0241] The term “non-transient” used here refers to a limitation on the medium itself (i.e., tangible, not a signal), rather than a limitation on the persistence of data storage (e.g., RAM versus ROM).

[0242] The memory can be of any type suitable for the local technical environment and can be implemented using any suitable data storage technology, such as semiconductor-based memory devices, magnetic memory devices and systems, optical memory devices and systems, fixed memory, and removable memory. As a non-limiting example, the data processor can be of any type suitable for the local technical environment and can include one or more of general-purpose computers, special-purpose computers, microprocessors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), FPGAs, gate-level circuits, and processors based on multi-core processor architectures.

[0243] The embodiments of this disclosure can be practiced in various components such as integrated circuit modules. The design of integrated circuits is largely a highly automated process. Sophisticated and powerful software tools can be used to transform logic-level designs into semiconductor circuit designs ready for etching and formation on a semiconductor substrate.

[0244] The foregoing description has provided a comprehensive and informative description of exemplary embodiments of the present disclosure by way of non-limiting examples. However, various modifications and adaptations may become apparent to those skilled in the art when read in conjunction with the accompanying drawings and appended claims, given the foregoing description. Indeed, other embodiments exist that combine one or more embodiments with any of the other embodiments discussed above. The scope of protection sought for some embodiments of the present disclosure is given by the claims. Embodiments and features (if any) described in this specification that do not fall within the scope of the claims are to be interpreted as examples useful for understanding the various embodiments of the present disclosure. It should be noted that different claims with different scopes may be pursued in related applications such as divisional or continuation applications.

Claims

1. An apparatus comprising: A component for determining a first value for channel state information, the first value being based on the channel between the user equipment and the base station, as well as interference; A component for determining encoding based on the first value for the channel state information, the encoding being generated by an encoder; A component for determining a second value for channel state information by removing at least the interference component from the first value; as well as A component for providing one or more of the second value or the encoding to the base station.

2. The apparatus of claim 1, wherein one or more components of the interference are the result of one or more of the following: inter-carrier interference; inter-symbol interference; co-channel interference; or cross-link interference.

3. The apparatus of claim 2, wherein the transmission power of the reference signal received from the base station depends on the signal-to-interference-plus-noise ratio (SINR) information associated with the channel between the user equipment and the base station.

4. The apparatus according to any of the preceding claims, comprising: A component for receiving reference signals of multiple different patterns from the base station.

5. The apparatus according to claim 4, comprising: A component for receiving from the base station one or more instructions associated with one or more patterns of the different patterns used for a reference signal.

6. The apparatus according to claim 4 or 5, comprising: A component for determining a matrix of multiple first values ​​using the first pattern among the different patterns.

7. The apparatus according to claim 4, 5 or 6, comprising: Components for determining one or more interference values ​​associated with one or more components of the interference using one or more other patterns among the different received signal patterns.

8. The apparatus of claim 7, wherein two or more different patterns are used to determine interference values ​​associated with different components of the interference.

9. A method comprising: A first value for channel state information is determined, the first value being based on the channel between the user equipment and the base station, as well as interference; The encoding is determined based on the first value for the channel state information, and the encoding is generated by an encoder; A second value for the channel state information is determined by removing at least the interference components from the first value; and This causes one or more of the second value or the encoding to be provided to the base station.

10. The method of claim 9, wherein one or more components of said interference are the result of one or more of the following: inter-carrier interference; inter-symbol interference; co-channel interference; or cross-link interference.

11. The method according to claim 9 or 10, comprising: Multiple reference signals with different patterns are received from the base station.

12. The method of claim 11, wherein one or more instructions associated with the one or more different patterns for a reference signal are received from the base station.

13. The method according to claim 11 or 12, comprising: A matrix of multiple first values ​​is determined using the first pattern among the different patterns.

14. The method according to claim 11, 12 or 13, comprising: One or more other patterns from the different received signal patterns are used to determine one or more interference values ​​associated with one or more components of the interference.

15. The method of claim 14, wherein two or more different patterns are used to determine interference values ​​associated with different components of the interference.

16. A method comprising: Receive one or more of the following from the user equipment: a) Encoding, the encoding being determined by an encoder based on a first value of channel state information for the channel between the user equipment and the base station, and interference; or b) A second value for the channel state information, the second value being determined by removing at least the interference component from the first value.

17. The method of claim 16, comprising: The decoder is trained such that it receives the encoding as input and provides the second value as output.

18. An apparatus comprising: Components for receiving one or more of the following from user equipment: a) Encoding, the encoding being determined by an encoder based on a first value of channel state information for the channel between the user equipment and the base station, and interference; or b) A second value for the channel state information, the second value being determined by removing at least the interference component from the first value.

19. The apparatus of claim 18, comprising: The component used to train the decoder so that when the decoder receives the encoding as input, the decoder provides the second value as output.

20. A computer program comprising computer-executable instructions that, when executed, cause the method according to any one of claims 9 to 17 to be performed.

Images