Digital pre-distortion under receiver quantization constraints

Abstract

Embodiments of the application relate to digital pre-distortion under receiver quantization constraints. Systems, methods, apparatuses, and computer program products for digital pre-distortion under receiver quantization constraints. A method can include receiving, from a network element, information associated with an analog-to-digital converter response for each carrier, bandwidth, or antenna chain. The method can also include determining at least one digital pre-distortion model based on the information received from the network element. The method can further include transmitting the determined at least one digital pre-distortion model to the network element.

Description

Technical Field

[0001] Some example embodiments may generally relate to mobile or wireless telecommunications systems, such as Long Term Evolution (LTE) or 5th Generation (5G) New Radio (NR) access technologies, or post-5G, or 6th Generation (6G) access technologies, or other communication systems. For example, some example embodiments may relate to digital predistortion under receiver quantization constraints. Background Technology

[0002] Examples of mobile or wireless telecommunications systems may include Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access Network (UTRAN), Evolved UTRAN (E-UTRAN) for Long Term Evolution (LTE), LTE-Advanced (LTE-A), MulteFire, LTE-A Pro, 5G or New Radio (NR) access technologies, and / or 6G radio access technologies. 5G and 6G radio systems refer to next-generation (NG) radio systems and network architectures. While 5G and 6G network technologies are primarily based on New Radio (NR) technologies, 5G / 6G (or NG) networks can also be built on E-UTRAN radios. NR is estimated to provide bit rates of 10 Gbit / s to 20 Gbit / s or higher and can at least support enhanced mobile broadband (eMBB), ultra-reliable low-latency communications (URLLC), and massive machine-type communications (mMTC). NR promises to provide extreme broadband and ultra-robust, low-latency connectivity and massive networking to support the Internet of Things (IoT). Summary of the Invention

[0003] Some example embodiments may relate to a method. The method may include: receiving from a network element information associated with the analog-to-digital converter response for each carrier, bandwidth, or antenna chain. The method may also include: determining at least one digital predistortion model based on the information received from the network element. The method may further include: transmitting the determined at least one digital predistortion model to the network element.

[0004] Other example embodiments may relate to an apparatus. The apparatus may include: at least one processor; and at least one memory including computer program code that, when executed by the at least one processor, causes the apparatus to at least: receive from network elements information associated with the analog-to-digital converter response of each carrier, bandwidth, or antenna chain. The apparatus may also be caused to: determine at least one digital predistortion model based on the information received from the network elements. The apparatus may further be caused to: transmit the determined at least one digital predistortion model to the network elements.

[0005] Other example embodiments may relate to an apparatus. The apparatus may include: components for receiving information from network elements associated with the analog-to-digital converter response of each carrier, bandwidth, or antenna chain. The apparatus may also include: components for determining at least one digital predistortion model based on the information received from the network elements. The apparatus may further include: components for transmitting the determined at least one digital predistortion model to the network elements.

[0006] According to other example embodiments, a non-transitory computer-readable medium may be encoded with instructions that, when executed in hardware, can perform a method. The method may include: receiving from a network element information associated with the analog-to-digital converter response of each carrier, bandwidth, or antenna chain. The method may also include: determining at least one digital predistortion model based on the information received from the network element. The method may further include: transmitting the determined at least one digital predistortion model to the network element.

[0007] Other example embodiments may relate to a computer program product that performs a method. The method may include: receiving from a network element information associated with the analog-to-digital converter response of each carrier, bandwidth, or antenna chain. The method may also include: determining at least one digital predistortion model based on the information received from the network element. The method may further include: transmitting the determined at least one digital predistortion model to the network element.

[0008] Other example embodiments may relate to an apparatus that may include: a circuit system for receiving information associated with the analog-to-digital converter response of each carrier, bandwidth, or antenna chain from network elements. The apparatus may also include: a circuit system for determining at least one digital predistortion model based on the information received from network elements. The apparatus may further include: a circuit system for transmitting the determined at least one digital predistortion model to network elements.

[0009] Another example embodiment may involve a method. This method may include: sending information to a user equipment associated with the analog-to-digital converter (ADC) response for each carrier, bandwidth, or antenna chain. The method may also include: receiving from the user equipment at least one digital predistortion model to be used based on the sent information. The method may further include: receiving from the user equipment a predistorted signal generated from the at least one digital predistortion model. Furthermore, the method may include: generating an ADC response based on the predistorted signal.

[0010] Other example embodiments may relate to an apparatus. The apparatus may include: at least one processor; and at least one memory including computer program code that, when executed by the at least one processor, causes the apparatus to at least: transmit to a user equipment information associated with the analog-to-digital converter response for each carrier, bandwidth, or antenna chain. The apparatus may also be configured to: receive from the user equipment at least one digital predistortion model to be used based on the transmitted information. The apparatus may also be configured to: receive from the user equipment a predistorted signal generated from the at least one digital predistortion model. Furthermore, the apparatus may be configured to: generate an analog-to-digital converter response based on the predistorted signal.

[0011] Other example embodiments may relate to an apparatus. The apparatus may include: components for transmitting to a user equipment information associated with the analog-to-digital converter response of each carrier, bandwidth, or antenna chain. The apparatus may also include: components for receiving from the user equipment at least one digital predistortion model to be used based on the transmitted information. The apparatus may further include: components for receiving from the user equipment a predistorted signal generated from the at least one digital predistortion model. Additionally, the apparatus may include: components for generating an analog-to-digital converter response based on the predistorted signal.

[0012] According to other example embodiments, a non-transitory computer-readable medium may be encoded with instructions that, when executed in hardware, can perform a method. The method may include: sending to a user equipment information associated with the analog-to-digital converter response for each carrier, bandwidth, or antenna chain. The method may also include: receiving from the user equipment at least one digital predistortion model to be used based on the sent information. The method may further include: receiving from the user equipment a predistorted signal generated from the at least one digital predistortion model. Furthermore, the method may include: generating an analog-to-digital converter response based on the predistorted signal.

[0013] Other example embodiments may relate to a computer program product that performs a method. The method may include: sending to a user equipment information associated with the analog-to-digital converter response for each carrier, bandwidth, or antenna chain. The method may also include: receiving from the user equipment at least one digital predistortion model to be used based on the sent information. The method may further include: receiving from the user equipment a predistorted signal generated from the at least one digital predistortion model. Furthermore, the method may include: generating an analog-to-digital converter response based on the predistorted signal.

[0014] Other example embodiments may relate to an apparatus that may include: a circuit system configured to transmit to a user equipment information associated with the analog-to-digital converter response of each carrier, bandwidth, or antenna chain. The apparatus may also include: a circuit system configured to receive from the user equipment at least one digital predistortion model to be used based on the transmitted information. The apparatus may further include: a circuit system configured to receive from the user equipment a predistorted signal generated from at least one digital predistortion model. Furthermore, the apparatus may include: a circuit system configured to generate an analog-to-digital converter response based on the predistorted signal. Attached Figure Description

[0015] To correctly understand the exemplary embodiments, reference should be made to the accompanying drawings, in which: Figure 1 An example signal flow diagram according to certain example embodiments is shown.

[0016] Figure 2 An example system model based on certain example embodiments is shown.

[0017] Figure 3 Example generation and configuration of an artificial intelligence and machine learning digital predistortion (AIML DPD) model are shown according to certain example embodiments.

[0018] Figure 4 An example digital twin of a network element front end is shown according to certain example embodiments.

[0019] Figure 5 An example flowchart of a method according to certain example embodiments is shown.

[0020] Figure 6 An example flowchart of another method according to some example embodiments is shown.

[0021] Figure 7 A set of apparatuses according to certain example embodiments is shown. Detailed Implementation

[0022] As will be readily understood, as generally described and illustrated in the accompanying drawings, the components of certain example embodiments can be arranged and designed in a wide variety of different configurations. The following is a detailed description of some example embodiments of systems, methods, apparatuses, and computer program products for digital predistortion under receiver quantization constraints. For example, some example embodiments may relate to artificial intelligence and machine learning (AIML) digital predistortion under receiver quantization constraints.

[0023] The features, structures, or characteristics of the exemplary embodiments described throughout this specification can be combined in any suitable manner in one or more exemplary embodiments. For example, throughout this specification, the phrases "certain embodiments," "exemplary embodiments," "some embodiments," or other similar language refer to the fact that a particular feature, structure, or characteristic described in connection with an embodiment can be included in at least one embodiment. Therefore, the phrases "in some embodiments," "exemplary embodiments," "in some embodiments," "in other embodiments," or other similar language appearing throughout this specification do not necessarily refer to the same set of embodiments, and the described features, structures, or characteristics can be combined in any suitable manner in one or more exemplary embodiments. Furthermore, the terms "base station," "cell," "node," "gNB," "network," or other similar language throughout this specification are used interchangeably.

[0024] As used herein, “at least one of the following: ” and “at least one of ” and similar wording, where the list of two or more elements is connected by “and” or “or”, indicates at least any one of the elements, or at least any two or more of the elements, or at least all of the elements.

[0025] In the field of wireless communication systems, digital predistortion (DPD) techniques can be used to mitigate nonlinear effects introduced by power amplifiers (PAs). DPD techniques in transmitters take into account the quantization constraints of the receiver analog-to-digital converters (ADCs) in 6G and advanced multiple-input multiple-output (MIMO) systems.

[0026] In 6G, massive MIMO antenna arrays can be used to achieve higher data rates, improved spectral efficiency, and enhanced coverage. These systems can operate at higher frequency ranges and wider bandwidths, which typically requires the deployment of multiple receiver chains equipped with ADCs to convert analog signals into digital form for baseband processing.

[0027] On the transmitter side, DPD can be used to mitigate the nonlinear effects introduced by PA, which may be necessary to boost signal power to the desired transmission level. PA nonlinearity can lead to undesirable spectral regeneration, resulting in interference in adjacent channels (e.g., adjacent channel leakage ratio (ACLR)) and degraded signal quality (e.g., error vector magnitude (EVM)). Therefore, stringent requirements on ACLR and EVM may be necessary to ensure compliance and interoperability in the network.

[0028] To mitigate the nonlinear effects of the PA, the DPD algorithm can be used to linearize the PA output. This is achieved by minimizing the difference between the transmitted signal and the desired linear reference, regardless of the quantization effects of the receiver's ADC. This implies that the transmitter may be performing complex signal processing to produce a high-fidelity signal that the receiver's low-resolution ADC cannot fully utilize, resulting in inefficiency and suboptimal system performance.

[0029] MIMO systems with increasingly large antenna arrays, operating at high frequencies and / or large bandwidths, risk becoming cost-inhibited due to the power consumption and hardware cost of the ADC. Specifically, the larger the number of quantization bits, the higher the hardware cost. To reduce hardware costs, low-resolution (e.g., a few bits and single bits) ADCs can be used instead of commercially available solutions providing 8 to 12 quantization bits. However, a drawback of low-resolution ADCs is the inherent quantization distortion they introduce into the received signal, which naturally translates to performance loss. Furthermore, transmitters performing DPDs may often disregard the receiver's ADC resolution, and therefore DPDs may perform unnecessarily complex operations to provide a clean transmit signal. Due to the resolution loss of the ADC, the transmitted signal may ultimately be completely unusable by the receiver.

[0030] In view of the above-mentioned drawbacks, certain example embodiments can utilize AIML technology in communication systems and can provide a way to enhance DPD algorithms. For example, AIML technology can incorporate models that can adaptively learn and compensate for various system impairments, including those introduced by both the transmitter's PA and the receiver's ADC. AIML-based DPD can also provide a more flexible and efficient solution by modeling complex nonlinearities and adapting to changing system conditions.

[0031] The DPD block in the transmitter can be configured to minimize the error between the PA output signal and the original transmitted signal. However, due to the ADC resolution loss during reception, the receiver may lose access to all the information contained in the received signal (and therefore in the transmitted signal) and can only access the quantized version of the received signal. Therefore, in some example embodiments, the DPD can be adjusted to account for receiver quantization loss. According to some example embodiments, instead of attempting to pre-compensate only the PA response, the user equipment (UE) can derive an AIML DPD configured to pre-compensate both the PA response and the ADC response for each receive antenna chain.

[0032] Figure 1 An example signal flow diagram according to certain example embodiments is shown. Figure 1As shown, a DPD model can be provided, wherein the DPD model can be obtained via information exchange between a receiver (e.g., a gNB in ​​the uplink (UL)) and a transmitter (e.g., a UE in the downlink (DL), which allows the UE to derive and use AIML quantization to sense DPD blocks).

[0033] like Figure 1 As shown, at 110, UE 110 sends a capability report to gNB 105. According to some example embodiments, the capability report may include information about UE 100's ability to derive and use AIML DPD under ADC quantization. According to some example embodiments, UE 100 may send the capability report in response to a capability request from gNB 105. At 115, gNB 105 sends ADC characteristic information to UE 110. For example, the UE may obtain information from gNB 105 about the ADC response for each carrier, bandwidth, and / or antenna chain. In some example embodiments, gNB 105 may transmit a response function or ADC function that approximates all ADC responses from all antennas of the gNB for a given ADC resolution range. In other example embodiments, the information received from gNB 105 may also include a set of sample pairs corresponding to inputs and corresponding ADC outputs. In some example embodiments, UE 100 may use a set of sample pairs to generate a function using AIML or interpolation techniques. In other example embodiments, the information received from gNB105 may include a set of signal samples and their associated signal-to-noise ratios (SNRs), which UE 100 may use to generate an ADC model.

[0034] According to some example embodiments, UE 100 and gNB 105 can agree on a data collection process, wherein UE 100 is configured for tagged UL transmissions, or gNB 105 generates samples offline using locally generated signals. For example, the agreement can be completed before triggering the data collection process, or as part of a Radio Resource Control (RRC) configuration. According to other example embodiments, gNB 105 can assist UE 100 in selecting a loss function for training the AIML DPD.

[0035] At 120, UE 100 uses information received from gNB 105 to define nominal gNB receiver behavior that can be used to train the UE DPD, where the nominal gNB can correspond to perceived gNB behavior (e.g., an approximate model of gNB behavior as seen by the UE). Additionally, at 120, the UE trains the AIML DPD using loss functions that provide various outcomes. For example, the loss function can minimize the error between transmitted bits and reconstructed bits (e.g., a binary cross-entropy function). The loss function can also minimize the error between transmitted symbol d and reconstructed symbol d. The loss function can further minimize the ADC output relative to the original transmitted OFDM signal after quantization. s The error between them.

[0036] At point 120, UE 100 further derives, generates, or determines AIML DPD models and trains multiple AIML DPD models covering different ADC resolution ranges. For example, according to some example embodiments, AIML models may include models for 1-bit resolution, models for 2-bit to 4-bit resolution, etc. In some example embodiments, the UE may also associate an identifier (ID) with each trained model, and at point 125, notify gNB 105 of the model ID, optionally including complete features of the model (e.g., model ID X supports ADC resolution b, carrier frequency...). (Bandwidth B, etc.). UE 100 and gNB 105 can synchronize with each other to determine which DPD model ID should be used. According to some example embodiments, if gNB 105 does not agree with the indicated DPD model ID, gNB 105 can send a request to change the DPD model ID. According to other example embodiments, UE 100 can also request a change of DPD model ID similar to that of gNB 105. Alternatively, in other example embodiments, gNB 105 or UE 100 may not request an agreement, but instead force UE 100 or gNB 105 to use the DPD model ID.

[0037] At point 130, if the PA (in the UE) and ADC (in the gNB) operations change, UE 100 and gNB 105 can also request a change in the model ID.

[0038] Figure 2An example system model according to certain example embodiments is shown. In some example embodiments, the UE AIMLDPD can operate under the quantization constraints of the receiver (e.g., in a gNB) ADC. As an example, a MIMO gNB receiver can be used. The MIMO gNB can include K antennas, where each receive antenna chain can be represented by the expression k=1:K. Each receive antenna can be equipped with its own ADC, and the ADC quantization response can depend at least on the carrier frequency ( f c ), bandwidth (B) and the number of resolution bits (b) k Therefore, according to some example embodiments, the k-th ADC response can be represented as a function. ,in It is the signal received at antenna k.

[0039] like Figure 2 As shown, at position 200, the UE generates the TX bit sequence. u It is encoded at position 205, and quadrature amplitude modulation (QAM) is modulated into a symbol vector at position 210. d In 215, vector d Modulated by OFDM to generate a signal s It is pre-distorted at 220 by AIML DPD to generate the signal. At 225, the signal... To generate a signal response Via PA. At 230, via a wireless channel (e.g., the physical channel between the transmitter (UE) and the receiver (gNB)) to the response corresponding to each receiving antenna k. The received signal Resulting signal It is the convolution between the transmitted signal and the channel response, plus additive white Gaussian noise. Damage. At points 235a, 235b, and 235c, the signal... r k Perform sampling, and then respond. Through the corresponding ADC 1 to ADC k. At 240a, 240b, and 240c, the output from the corresponding ADC... Demodulated into a complex vector by OFDM At 245, all complex vectors Combiner C Equilibrium and combination. At 250, for complex vectors... Perform demapping and decoding to reconstruct the original bit sequence obtained at position 255. According to some example embodiments, the decoding process may include error-correcting decoding (e.g., low-density parity-check (LDPC) or Turbo decoding), and / or mapping the demodulated symbols back to binary form.

[0040] According to certain example embodiments, the DPD block can be configured to minimize the PA output signal. and the original OFDM signal s The error between them. However, due to the ADC resolution loss, the receiver may lose focus on the signal contained within it. s This allows for access to complete information within the UE, and therefore the DPD can be adjusted to account for ADC resolution loss. For example, according to some example embodiments, the UE can generate an AIML DPD for both the pre-compensated PA response and the K ADC response.

[0041] Figure 3 This illustrates example generation and configuration of an AIML DPD model according to certain example embodiments. At 300, the UE identifies the need to adjust the DPD. At 305, the UE obtains information about the ADC response from the gNB. Information. At 310, the information obtained from the gNB can include different types and can include one or a combination of any information types described herein and shown in the figures. For example, at 315, the information can include full functions. Or gNB can transmit Some parameters (e.g., bit resolution) At position 320, the information may include an approximate function. Its approximate each OR function It approximates all antennas from the gNB for a given ADC resolution range. At position 325, this information could include a set of sample pairs. The UE can use this set of samples Let's learn functions At 330, this information may include a set of received signal samples and their associated SNRs, which the UE can use to generate an ADC model. For example, sample pairs may include... The UE can use this To approximate the received signal For example, by considering the noise variance is : The inverse AWGN channel, and / or based on the obtained approximation and against Learning function .

[0042] At position 335, when the UE receives a set of sample pairs or sample At this point, the UE and gNB can agree on either at point 340, where the UE is configured to perform tagged UL transmission, or at point 345, where the gNB uses locally generated signals to generate sample data offline. At point 350, the UE can learn functions. For example, AIML can be used to learn a function. According to some example embodiments, the learning function can be found using a reference signal and the related expected output from the function. Therefore, the learning function can be represented as an equation or an AIML training algorithm. At 355, the UE can indicate that it can choose from multiple loss functions and request assistance in making the selection. Once requested, the gNB can provide the requested assistance to select a loss function for training the AIML DPD. If so, at 360, the gNB can continue to assist the UE in selecting a loss function. For example, the gNB can assist the UE in selecting a loss function for training the AIML DPD by providing context-specific recommendations or options tailored to the characteristics of the receiver's ADC and the transmission scenario. Assistance may include suggesting a loss function that minimizes the error in a particular domain, such as reconstructed transmitted bits, transmitted symbols, ADC output relative to the original OFDM signal, or a quantized version of the signal. If not, at 365, the UE can apply the DPD to predistort the signal that has been modulated by OFDM. s .

[0043] Figure 4 An example digital twin of a network element (gNB) front-end is illustrated according to certain example embodiments. For example, in some example embodiments, the UE can use information obtained from the gNB to create... Figure 4 The digital twin of the gNB front end is shown. For example... Figure 4 As shown, a digital twin can be similar to Figure 2 The system shown has had its backend components, including operations 240a to 260, removed. Therefore, Figure 4 Operations 400 to 425 and operations 435a to 435c shown can be similar to Figure 2 Operations 200 to 225 and operations 235a to 235c are shown and described herein. Figure 4 As shown, at 430, it is not Figure 2The wireless channel shown is simplified by ignoring the propagation channel response and replacing it with an identity channel. In some example embodiments, the simplified channel may correspond to an abstraction used to create a digital twin of the gNB front-end during training. The simplified channel can be designed for computational simplicity and can focus on basic elements. For example, the simplified channel may be an identity channel in which the propagation channel response is ignored, thereby simplifying the model to avoid unnecessary complexity during training. According to some example embodiments, from Figure 4 It can be assumed that receiver baseband processing (e.g., using coding redundancy) can resolve channel errors.

[0044] According to certain example embodiments, Figure 3 The learning function 350 (e.g., about The function estimation process may include the UE training the AIML DPD using a loss function. According to some example embodiments, by training the AIML DPD using a loss function, the transmitted bits can be minimized. With reconstructed bits The error between (e.g., the binary cross-entropy function) can be minimized by training the AIML DPD using a loss function, according to other example embodiments. The error between, for example, According to another example embodiment, by training AIMLDPD using a loss function, the ADC output can be minimized. Compared with the original OFDM signal The error between, for example, According to other example embodiments, by training the AIML DPD using a loss function, it is possible to target a set of carrier frequencies. The UE is trained using bandwidth B. Additionally, in some example embodiments, the UE can train one or more models to cover different resolution ranges. (For example, a model for 1-bit resolution, another model for 2-bit to 4-bit resolution, etc.).

[0045] In some example embodiments, when training an AIML DPD using a loss function (where multiple models can be trained to cover different resolution ranges), the UE can associate an ID with each trained model and notify the gNB of the model ID, which includes the complete features of the ID (e.g., model ID X supports ADC resolution b, carrier frequency). In other example embodiments, the UE and gNB can synchronize with each other to determine which DPD model ID to use, such as bandwidth B. In some example embodiments, if the PA (in the UE) and ADC (in the gNB) operations change, both the UE and gNB can request a change in the DPD model ID. In other words, the use of a particular DPD model ID can be dynamic.

[0046] As described herein, the UE can generate an AIML DPD configured as both a pre-compensated PA response and a K ADC response. According to some example embodiments, to accomplish this task, the UE can target the DPD for reconstruction. This allows the UE to reconstruct the information that the gNB receiver can actually access. Using this objective, the UE can train the AIML DPD to minimize... and The error between, for example, In other example embodiments, if the gNB transmits only one set of sample pairs... Then the UE can use a set of sample pairs Learning functions .

[0047] According to certain example embodiments, the UE can learn or derive through various schemes that may include, for example, ML schemes or non-ML schemes. In the ML scheme, the UE can use an ML model to learn a function from sample pairs provided by gNB. This ML model can involve gNB sending labeled sample pairs. ,in It is the signal before the ADC, and This is the ADC output. In the ML scheme, the UE can use the provided sample pairs to train an ML model (e.g., a neural network or regression model) to approximate... Once the model is trained, the UE can use it to adjust the AIML DPD, thereby compensating for both PA and ADC nonlinearities. According to some example implementations, the UE can continuously refine the model as more data arrives from the gNB.

[0048] In non-ML schemes, the UE can use mathematical estimation techniques to derive the ADC response. Mathematical estimation techniques may include, for example, the UE using approximations. Interpolation from sample pairs Construct a lookup table (LUT). In other example embodiments, mathematical estimation techniques may involve the UE using Volterra series or another polynomial extension to model the nonlinear behavior of the ADC and derive an approximate inverse function. In further example embodiments, the UE may use iterative methods (e.g., least squares) to estimate... The DPD model is refined based on the estimated ADC response.

[0049] Figure 5 An example flowchart of a method according to certain example embodiments is shown. In the example embodiments, Figure 5 The method can be performed by a network entity or a group of multiple network elements in a 3GPP system such as LTE or 5G-NR. For example, in an example embodiment, Figure 5 The method can be executed by the UE, similar to Figure 7 One of the devices 10 or 20 shown.

[0050] like Figure 5 As shown, the method may include, at 500, receiving from a network element information associated with the analog-to-digital converter response for each carrier, bandwidth, or antenna chain. The method may also include, at 505, determining at least one digital predistortion model based on the information received from the network element. According to some example embodiments, the digital predistortion model may correspond to an AIML digital predistortion model. The method may further include, at 510, transmitting the determined at least one digital predistortion model to the network element.

[0051] According to some example embodiments, the method may also include sending a capability report message to a network element, including the digital predistortion capability of the user equipment. According to some example embodiments, the method may also include receiving a capability request message from a network element, including a request for a capability of the user equipment. This capability may be associated with the digital predistortion capability of the user equipment. According to other example embodiments, the method may also include performing predistortion of a signal based on at least one determined digital predistortion model.

[0052] In some example embodiments, the method may also include associating an identifier with at least one digital predistortion model. In some example embodiments, the identifier identifies a feature of at least one digital predistortion model. In other example embodiments, the method may further include sending the identifier to a network element.

[0053] According to some example embodiments, the features of at least one digital predistortion model may include at least one of the following: at least one analog-to-digital converter resolution supported by the at least one digital predistortion model, at least one carrier frequency supported by the at least one digital predistortion model, and at least one bandwidth supported by the at least one digital predistortion model. According to some example embodiments, the method may also include: requesting a network element to change at least one digital predistortion model associated with an identifier, or receiving a request from the network element to change at least one digital predistortion model associated with an identifier. According to other example embodiments, the method may further include: requesting a change in the identifier associated with at least one digital predistortion model based on operational changes in the power amplifier of the user equipment and the analog-to-digital converter of the network element.

[0054] In some example embodiments, the information received from the network element may include at least one of the following: some parameters of the analog-to-digital converter (ADC) function or the complete ADC function, an approximation of the ADC function, a set of sample pairs corresponding to the input and the corresponding ADC output, and a set of received signal samples and their associated signal-to-noise ratios. In some example embodiments, the method may also include determining a data collection process in conjunction with the network element. In other example embodiments, the user equipment may be configured for tagged uplink transmissions, or the samples received by the user equipment from the network element may be generated offline using locally generated signals.

[0055] According to some example embodiments, the method may further include: training at least one digital predistortion model based on a selected loss function. According to some example embodiments, the method may further include: receiving assistance from network elements to select a loss function for training the at least one digital predistortion model. According to other example embodiments, the loss function is configured to perform at least one of the following: minimizing the error between transmitted bits and reconstructed bits, minimizing the error between transmitted symbols and reconstructed symbols, minimizing the error between the analog-to-digital converter output and the original transmitted orthogonal frequency division multiplexed signal, and minimizing the error between the analog-to-digital converter output and the quantized version of the original orthogonal frequency division multiplexed signal that has undergone quantization.

[0056] In some example embodiments, the method may also include: reconstructing the analog-to-digital converter (ADC) function via at least one digital predistortion model. In other example embodiments, the method may also include: receiving only one set of sample pairs from network elements, the user equipment using the set of sample pairs to learn the ADC function.

[0057] Figure 6 An example flowchart of another method according to certain example embodiments is shown. In the example embodiments, Figure 6The method can be performed by a network entity or a group of multiple network elements in a 3GPP system such as LTE or 5G-NR. For example, in an example embodiment, Figure 6 The method can be executed by NW or gNB, similar to Figure 7 One of the devices 10 or 20 shown.

[0058] like Figure 6 As shown, the method may include, at 600, sending to the user equipment information associated with the analog-to-digital converter response for each carrier, bandwidth, or antenna chain. The method may also include, at 605, receiving from the user equipment at least one digital predistortion model to be used based on the sent information. The method may further include, at 610, receiving from the user equipment a predistorted signal generated from the at least one digital predistortion model. Furthermore, the method may include, at 615, generating an analog-to-digital converter response based on the predistorted signal.

[0059] According to some example embodiments, the method may also include: receiving a capability report message from a user equipment including the user equipment's digital predistortion capabilities. According to some example embodiments, the method may further include: sending a capability request message to the user equipment including a request for a capability of the user equipment. This capability may be associated with the user equipment's digital predistortion capabilities. According to other example embodiments, at least one digital predistortion model may be associated with an identifier, and the identifier may identify a feature of at least one digital predistortion model.

[0060] In some example embodiments, the method may further include receiving an identifier from a user equipment. In some example embodiments, the features of at least one digital predistortion model may include at least one of the following: at least one analog-to-digital converter resolution supported by the at least one digital predistortion model, at least one carrier frequency supported by the at least one digital predistortion model, and at least one bandwidth supported by the at least one digital predistortion model. In other example embodiments, the method may include receiving from the user equipment a request to change at least one digital predistortion model associated with the identifier; or sending a request to the user equipment to change at least one digital predistortion model associated with the identifier.

[0061] According to some example embodiments, the method may also include: requesting changes to an identifier associated with at least one digital predistortion model based on operational changes in the power amplifier and analog-to-digital converter of the user equipment. According to some example embodiments, the information sent to the user equipment may include at least one of the following: some parameters of the analog-to-digital converter function or the complete analog-to-digital converter function, an approximation of the analog-to-digital converter function, a set of sample pairs corresponding to the input and the corresponding analog-to-digital converter output, and a set of received signal samples and their associated signal-to-noise ratios. According to other example embodiments, the method may also include: determining a data collection process in conjunction with the user equipment.

[0062] In some example embodiments, the method may also include sending assistance to the user equipment for selecting a loss function to train at least one digital predistortion model. In some example embodiments, the method may also include sending only one set of sample pairs to the user equipment, which uses the set of sample pairs to learn the analog-to-digital converter functionality.

[0063] Figure 7 A set of devices 10, 20 according to certain example embodiments is shown. In some example embodiments, devices 10, 20 may be elements in or associated with a communication network. For example, device 10 may be a UE or other similar radio communication computer equipment, and device 20 may be a BS, gNB, network, or other similar computing device.

[0064] In some example embodiments, devices 10, 20 may include one or more processors, one or more computer-readable storage media (e.g., memory, storage device, etc.), one or more radio access components (e.g., modem, transceiver, etc.), and / or a user interface. In some example embodiments, devices 10, 20 may be configured to operate using one or more radio access technologies, such as GSM, LTE, LTE-A, NR, 5G, WLAN, WiFi, NB-IoT, Bluetooth, NFC, MulteFire, and / or any other radio access technology. It should be noted that those skilled in the art will understand that devices 10, 20 may include... Figure 7 Components or features not shown in the diagram.

[0065] like Figure 7 As shown in the example, devices 10, 20 may include or be coupled to processors 12, 22 for processing information and executing instructions or operations. Processors 12, 22 may be any type of general-purpose or special-purpose processor. In practice, as an example, processors 12, 22 may include one or more of a general-purpose computer, a special-purpose computer, a microprocessor, a DSP, a field-programmable gate array (FPGA), an application-specific integrated circuit (ASIC), and a processor based on a multi-core processor architecture. Although in Figure 7 A single processor 12, 22 is shown, but multiple processors may be utilized according to other example embodiments. For example, it should be understood that in some example embodiments, devices 10, 20 may include two or more processors that can form a multiprocessor system capable of supporting multiple processing (e.g., in this case, processor 12 may represent multiple processors). According to some example embodiments, the multiprocessor system may be tightly coupled or loosely coupled (e.g., to form a computer cluster).

[0066] Processors 12 and 22 can perform functions associated with the operation of devices 10 and 20, including, for example, precoding antenna gain / phase parameters, encoding and decoding of individual bits forming communication messages, formatting information, and overall control of devices 10 and 20. Figures 1 to 6 The process and examples are shown below.

[0067] Devices 10 and 20 may further include or be coupled to memories 14 and 24 (internal or external), which may be coupled to processors 12 and 24 respectively for storing information and instructions executable by processors 12 and 24. Memories 14 and 24 may be one or more memories and may be of any type suitable for the local application environment, and may be implemented using any suitable volatile or non-volatile data storage technology, such as semiconductor-based memory devices, magnetic memory devices and systems, optical memory devices and systems, fixed memory and / or removable memory. For example, memories 14 and 24 may consist of random access memory (RAM), read-only memory (ROM), static storage devices such as disks or optical discs, hard disk drives (HDDs), or any other type of non-transitory machine or computer-readable medium, and any combination thereof. Instructions stored in memories 14 and 24 may include program instructions or computer program code that, when executed by processors 12 and 22, enable devices 10 and 20 to perform the tasks described herein.

[0068] In some example embodiments, devices 10, 20 may also include or be coupled to an (internal or external) drive or port configured to accept and read external computer-readable storage media, such as an optical disc, USB drive, flash drive, or any other storage media. For example, the external computer-readable storage media may store computer programs or software for execution by processors 12, 22 and / or devices 10, 20. Figures 1 to 6 Any methods and examples shown.

[0069] In some example embodiments, devices 10, 20 may further include or be coupled to one or more antennas 15, 25 for receiving downlink signals and for transmitting from devices 10, 20 via UL. Devices 10, 20 may also include transceivers 18, 28 configured to transmit and receive information. Transceivers 18, 28 may also include a radio interface (e.g., a modem) coupled to antennas 15, 25. The radio interface may correspond to one or more of various radio access technologies, including GSM, LTE, LTE-A, 5G, NR, WLAN, NB-IoT, Bluetooth, BT-LE, NFC, RFID, UWB, etc. The radio interface may include other components such as filters, converters (e.g., digital-to-analog converters, etc.), symbol demappers, signal shaping components, inverse fast Fourier transform (IFFT) modules, etc., to process symbols (such as OFDMA symbols) carried by the downlink or UL.

[0070] For example, transceivers 18 and 28 may be configured to modulate information onto a carrier waveform for transmission by antennas 15 and 25, and demodulate information received via antennas 15 and 25 for further processing by other elements of devices 10 and 20. In other example embodiments, transceivers 18 and 28 may be capable of directly transmitting and receiving signals or data. Additionally or alternatively, in some example embodiments, device 10 may include input and / or output devices (I / O devices). In some example embodiments, devices 10 and 20 may also include a user interface, such as a graphical user interface or a touchscreen.

[0071] In some example embodiments, memories 14, 34 store software modules that provide functionality when executed by processors 12, 22. These modules may include, for example, an operating system that provides operating system functionality for devices 10, 20. The memories may also store one or more functional modules, such as applications or programs, to provide additional functionality to devices 10, 20. Components of devices 10, 20 may be implemented in hardware or as any suitable combination of hardware and software. According to some example embodiments, devices 10, 20 may optionally be configured to communicate with each other (in any combination) via wireless or wired communication link 70 according to any radio access technology, such as NR.

[0072] According to some example embodiments, processors 12, 22 and memories 14, 24 may be included in or form part of a processing circuit system or control circuit system. Furthermore, in some example embodiments, transceivers 18, 28 may be included in or form part of a transceiver circuit system.

[0073] For example, in some exemplary embodiments, device 10 may be controlled by memory 14 and processor 12 to receive information from network elements associated with the analog-to-digital converter response for each carrier, bandwidth, or antenna chain. Device 10 may also be controlled by memory 14 and processor 12 to determine at least one digital predistortion model based on the information received from network elements. Device 10 may also be controlled by memory 14 and processor 12 to transmit the determined at least one digital predistortion model to network elements.

[0074] In other example embodiments, device 20 may be controlled by memory 24 and processor 22 to send information associated with the analog-to-digital converter response for each carrier, bandwidth, or antenna chain to the user equipment. Device 20 may also be controlled by memory 24 and processor 22 to receive from the user equipment at least one digital predistortion model to be used based on the sent information. Device 20 may also be controlled by memory 24 and processor 22 to receive a predistortion signal generated from at least one digital predistortion model from the user equipment. Furthermore, device 20 may be controlled by memory 24 and processor 22 to generate an analog-to-digital converter response based on the predistortion signal.

[0075] In some example embodiments, the apparatus (e.g., apparatus 10 and / or apparatus 20) may include components for performing the methods, processes, or any variations discussed herein. Examples of such components may include one or more processors, memories, controllers, transmitters, receivers, and / or computer program code for inducing the execution of operations.

[0076] Some example embodiments may relate to an apparatus comprising: components for performing any of the methods described herein, including, for example, components for receiving from network elements information associated with the analog-to-digital converter response of each carrier, bandwidth, or antenna chain. The apparatus may further comprise: components for determining at least one digital predistortion model based on the information received from the network elements. The apparatus may further comprise: components for transmitting the determined at least one digital predistortion model to the network elements.

[0077] Other example embodiments may relate to an apparatus comprising: components for performing any of the methods described herein, including, for example, components for transmitting to a user equipment information associated with the analog-to-digital converter response for each carrier, bandwidth, or antenna chain. The apparatus may also include: components for receiving from the user equipment at least one digital predistortion model to be used based on the transmitted information. The apparatus may further include: components for receiving from the user equipment a predistorted signal generated from the at least one digital predistortion model. Additionally, the apparatus may include: components for generating an analog-to-digital converter response based on the predistorted signal.

[0078] Certain example embodiments described herein provide several technical improvements, enhancements, and / or advantages. For example, in some example embodiments, AIML techniques can be integrated to enhance the DPD algorithm by incorporating models that can adaptively learn and compensate for various system impairments, including those introduced by both the transmitter's PA and the receiver's ADC. According to other example embodiments, AIML-based DPD can provide a more flexible and efficient solution by modeling complex nonlinearities and adapting to changing system conditions. For example, in some example embodiments, AIML DPD can pre-compensate the PA and ADC responses for each receive antenna chain.

[0079] Other example embodiments may quantize the receiver's ADC response ( Integrating this into the DPD design ensures that the transmitter's output signal is aligned with the receiver's capabilities, thus avoiding the unnecessary complexity and inefficiency of signal compensation. Furthermore, certain example implementations can be customized for scalable 6G MIMO systems with low-resolution ADCs. For instance, the challenges of low-resolution ADCs in massive 6G MIMO systems can be addressed, providing a scalable solution that reduces hardware costs and power consumption without significant performance degradation.

[0080] In other example embodiments, collaborative data exchange for model training and synchronization can be provided. For example, the gNB and UE can exchange ADC-related information, such as the full response function, approximation, or sample pairs. This collaborative scheme allows the UE to efficiently train and synchronize AIML-based DPD models. Some example embodiments can also provide dynamic adaptability to AIML. For example, AIML can be used to dynamically adapt the DPD model to changing PA and ADC configurations to ensure robust performance in various operating scenarios.

[0081] According to some example embodiments, improved system performance can also be provided with low-resolution ADCs. For example, by pre-compensating both the PA and ADC nonlinearity, information loss can be minimized, resulting in better signal quality, lower error rates, and improved throughput even with low-resolution ADCs. According to some example embodiments, future-ready designs for 6G standards can be provided. For example, some example embodiments can provide designs that meet the needs of 6G networks, including massive MIMO deployments and high-frequency operation. According to other example embodiments, cost-effective implementations can be provided. For example, low-resolution ADCs can be used without compromising performance and reducing hardware costs, leading to economically viable large-scale deployments.

[0082] In some example embodiments, computational complexity can be reduced by aligning DPD operations with the receiver's quantization behavior, thereby avoiding overcompensation and resulting in lower computational overhead. Additionally, some example embodiments can improve spectral efficiency by focusing on reducing ACLR and improving EVM. This improvement may be relevant to high-density networks. In other example embodiments, simplified standardization and detectability can be provided, where the signaling framework between the gNB and UE for exchanging ADC-related data provides a path for standardization. This can simplify the adoption of certain example embodiments and make such use detectable in a contested environment.

[0083] Certain example embodiments can further provide improved interoperability across various configurations. For example, some example embodiments can support multiple DPD models tailored for various ADC resolutions, bandwidths, and carrier frequencies. This can lead to high interoperability across different network and hardware configurations. Furthermore, other example embodiments can leverage digital twin models to provide training efficiency, where using a simplified channel during training ensures efficient training of the AIML DPD model. Efficient training can be achieved by focusing on PA and ADC features without the overhead of channel effects.

[0084] A computer program product may include one or more computer-executable components that, when the program runs, are configured to perform some example embodiments. The one or more computer-executable components may be at least one piece of software code or a portion thereof. Modifications and configurations required to implement the functionality of certain example embodiments may be executed as routines, which may be implemented as added or updated software routines. Software routines may be downloaded to a device.

[0085] As an example, software or computer program code, or portions thereof, may be in the form of source code, object code, or some intermediate form, and may be stored in some carrier, distribution medium, or computer-readable medium, which may be any entity or device capable of carrying the program. For example, such a carrier may include recording media, computer memory, read-only memory, photoelectric and / or electrical carrier signals, telecommunication signals, and software distribution packages. Depending on the required processing power, a computer program may execute in a single electronic digital computer, or it may be distributed across multiple computers. The computer-readable medium or computer-readable storage medium may be a non-transitory medium.

[0086] In other example embodiments, the function may be performed by hardware or circuitry included in the device (e.g., device 10 or device 20), for example, by using an application-specific integrated circuit (ASIC), a programmable gate array (PGA), a field-programmable gate array (FPGA), or any other combination of hardware and software. In yet another example embodiment, the function may be implemented as a signal, an intangible means that can be carried by an electromagnetic signal downloaded from the Internet or other networks.

[0087] According to certain example embodiments, an apparatus (such as a node, device, or corresponding component) may be configured as a circuit system, a computer, or a microprocessor (such as a single-chip computer element) or a chipset, including at least a memory for providing storage capacity for arithmetic operations and an arithmetic processor for performing arithmetic operations.

[0088] It will be readily understood by those skilled in the art that the present disclosure as described above can be practiced using processes in different sequences and / or using hardware elements in configurations different from the disclosed configuration. Therefore, although the present disclosure has been described based on these exemplary embodiments, it will be apparent to those skilled in the art that certain modifications, variations, and alternative constructions will be readily apparent while remaining within the spirit and scope of the exemplary embodiments. While the above embodiments relate to 5G NR and LTE technologies, the above embodiments can also be applied to any other current or future 3GPP technologies, such as LTE-Advanced and / or fourth-generation (4G) technologies.

[0089] Partial vocabulary list:

[0090] Furthermore, the various implementations of this disclosure can be described with reference to the following terms, and their features can be combined in any reasonable manner.

[0091] Clause 1. An apparatus comprising: at least one processor; and at least one memory including computer program code, which, when executed by the at least one processor, causes the apparatus to at least: receive from a network element information associated with an analog-to-digital converter response for each carrier, bandwidth, or antenna chain; determine at least one digital predistortion model based on the information received from the network element; and transmit the determined at least one digital predistortion model to the network element.

[0092] Clause 2. The apparatus according to Clause 1, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: send a capability report message to the network element including the digital predistortion capability of the apparatus.

[0093] Clause 3. The apparatus according to Clause 2, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: receive from the network element a capability request message including a request for a user equipment capability.

[0094] Clause 4. The apparatus according to any one of Clauses 1 to 3, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: perform predistortion of the signal based on the determined at least one digital predistortion model.

[0095] Clause 5. The apparatus according to any one of Clauses 1 to 4, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: associate an identifier with the at least one digital predistortion model, wherein the identifier identifies a feature of the at least one digital predistortion model; and send the identifier to the network element.

[0096] Clause 6. The apparatus according to Clause 5, wherein the features of the at least one digital predistortion model include at least one of the following: at least one analog-to-digital converter resolution supported by the at least one digital predistortion model, at least one carrier frequency supported by the at least one digital predistortion model, and at least one bandwidth supported by the at least one digital predistortion model.

[0097] Clause 7. The apparatus according to Clause 5, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: request the network element to change the at least one digital predistortion model associated with the identifier, or receive from the network element a request to change the at least one digital predistortion model associated with the identifier.

[0098] Clause 8. The apparatus according to Clause 5, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: request a change in the identifier associated with the at least one digital predistortion model based on operational changes in the power amplifier and analog-to-digital converter of the network element of the apparatus.

[0099] Clause 9. The apparatus according to any one of Clauses 1 to 8, wherein the information received from the network element includes at least one of the following: some parameters of the analog-to-digital converter function or the complete analog-to-digital converter function, an approximation of the analog-to-digital converter function, a set of sample pairs corresponding to the input and the corresponding analog-to-digital converter output, and a set of received signal samples and their associated signal-to-noise ratio.

[0100] Clause 10. The apparatus according to any one of Clauses 1 to 9, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: determine a data collection process together with the network element, wherein the apparatus is configured for tagged uplink transmissions, or wherein samples received by the apparatus from the network element are generated offline using locally generated signals.

[0101] Clause 11. The apparatus according to any one of Clauses 1 to 9, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: train the at least one digital predistortion model based on a selected loss function.

[0102] Clause 12. The apparatus according to Clause 11, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: receive assistance from the network element to select the loss function for training the at least one digital predistortion model.

[0103] Clause 13. The apparatus according to Clause 12, wherein the loss function is configured to perform at least one of the following: minimizing the error between transmitted bits and reconstructed bits, minimizing the error between transmitted symbols and reconstructed symbols, minimizing the error between the analog-to-digital converter output and the original transmitted orthogonal frequency division multiplexed signal, and minimizing the error between the analog-to-digital converter output and the quantized version of the original orthogonal frequency division multiplexed signal that has undergone quantization.

[0104] Clause 14. The apparatus according to any one of Clauses 1 to 13, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: reconstruct the analog-to-digital converter function via the at least one digital predistortion model.

[0105] Clause 15. The apparatus according to any one of Clauses 1 to 14, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: receive only one set of sample pairs from the network element, the apparatus using the set of sample pairs to learn analog-to-digital converter functionality.

[0106] Clause 16. An apparatus comprising: at least one processor; and at least one memory including computer program code, which, when executed by the at least one processor, causes the apparatus to at least: transmit to a user equipment information associated with an analog-to-digital converter response for each carrier, bandwidth, or antenna chain; receive from the user equipment at least one digital predistortion model to be used based on the transmitted information; receive from the user equipment a predistortion signal generated from the at least one digital predistortion model; and generate the analog-to-digital converter response based on the predistortion signal.

[0107] Clause 17. The apparatus according to Clause 16, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: receive from the user equipment a capability report message including the user equipment's digital predistortion capability.

[0108] Clause 18. The apparatus according to Clause 17, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: send a capability request message to the user equipment including a request for capabilities of the user equipment.

[0109] Clause 19. The apparatus according to any one of Clauses 16 to 18, wherein the at least one digital predistortion model is associated with an identifier, and wherein the identifier identifies a feature of the at least one digital predistortion model.

[0110] Clause 20. The apparatus according to Clause 19, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: receive the identifier from the user equipment.

[0111] Clause 21. The apparatus according to Clause 19, wherein the features of the at least one digital predistortion model include at least one of the following: at least one analog-to-digital converter resolution supported by the at least one digital predistortion model, at least one carrier frequency supported by the at least one digital predistortion model, and at least one bandwidth supported by the at least one digital predistortion model.

[0112] Clause 22. The apparatus according to Clause 19, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: receive from the user equipment a request to change the at least one digital predistortion model associated with the identifier; or send to the user equipment a request to change the at least one digital predistortion model associated with the identifier.

[0113] Clause 23. The apparatus according to Clause 19, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: request a change in the identifier associated with the at least one digital predistortion model based on operational changes in the power amplifier of the user equipment and the analog-to-digital converter of the apparatus.

[0114] Clause 24. The apparatus according to any one of Clauses 16 to 23, wherein the information transmitted to the user equipment includes at least one of the following: some parameters of the analog-to-digital converter function or the complete analog-to-digital converter function, an approximation of the analog-to-digital converter function, a set of sample pairs corresponding to the input and the corresponding analog-to-digital converter output, and a set of received signal samples and their associated signal-to-noise ratio.

[0115] Clause 25. The apparatus according to any one of Clauses 16 to 24, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: determine the data collection process together with the user equipment.

[0116] Clause 26. The apparatus according to any one of Clauses 16 to 25, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: send assistance to the user equipment for the selection of a loss function for training the at least one digital predistortion model.

[0117] Clause 27. The apparatus according to any one of Clauses 16 to 26, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: send to the user equipment only one set of sample pairs, the user equipment using the set of sample pairs to learn analog-to-digital converter functionality.

[0118] Clause 28. A method comprising: receiving from a network element information associated with an analog-to-digital converter response for each carrier, bandwidth, or antenna chain; determining at least one digital predistortion model based on the information received from the network element; and transmitting the determined at least one digital predistortion model to the network element.

[0119] Clause 29. The method according to Clause 28 further includes: sending a capability report message to the network element including the digital predistortion capability of the user equipment.

[0120] Clause 30. The method according to Clause 29 further includes: receiving from the network element a capability request message including a request for user equipment capabilities.

[0121] Clause 31. The method according to any one of Clauses 28 to 30 further comprises: performing predistortion of the signal based on the determined at least one digital predistortion model.

[0122] Clause 32. The method according to any one of Clauses 28 to 31 further comprises: associating an identifier with the at least one digital predistortion model, wherein the identifier identifies a feature of the at least one digital predistortion model; and sending the identifier to the network element.

[0123] Clause 33. The method according to Clause 32, wherein the features of the at least one digital predistortion model include at least one of the following: at least one analog-to-digital converter resolution supported by the at least one digital predistortion model, at least one carrier frequency supported by the at least one digital predistortion model, and at least one bandwidth supported by the at least one digital predistortion model.

[0124] Clause 34. The method according to Clause 32 further includes: requesting the network element to change the at least one digital predistortion model associated with the identifier; or receiving from the network element a request to change the at least one digital predistortion model associated with the identifier.

[0125] Clause 35. The method according to Clause 32 further comprises: requesting a change in the identifier associated with the at least one digital predistortion model based on operational changes in the power amplifier of the user equipment and the analog-to-digital converter of the network element.

[0126] Clause 36. The method according to any one of Clauses 28 to 35, wherein the information received from the network element includes at least one of the following: some parameters of the analog-to-digital converter function or the complete analog-to-digital converter function, an approximation of the analog-to-digital converter function, a set of sample pairs corresponding to the input and the corresponding analog-to-digital converter output, and a set of received signal samples and their associated signal-to-noise ratio.

[0127] Clause 37. The method of any one of Clauses 28 to 36 further comprises: determining a data collection process together with the network element, wherein the device is configured for labeled uplink transmission, or wherein samples received by the device from the network element are generated offline using locally generated signals.

[0128] Clause 38. The method according to any one of Clauses 28 to 36 further includes: training the at least one digital predistortion model based on a selected loss function.

[0129] Clause 39. The method according to Clause 38 further comprises: receiving assistance from the network element to select the loss function for training the at least one digital predistortion model.

[0130] Clause 40. The method according to Clause 39, wherein the loss function is configured to perform at least one of the following: minimizing the error between transmitted bits and reconstructed bits, minimizing the error between transmitted symbols and reconstructed symbols, minimizing the error between the analog-to-digital converter output and the original transmitted orthogonal frequency division multiplexed signal, and minimizing the error between the analog-to-digital converter output and the quantized version of the original orthogonal frequency division multiplexed signal that has undergone quantization.

[0131] Clause 41. The method according to any one of Clauses 28 to 40 further includes: reconstructing the analog-to-digital converter function via the at least one digital predistortion model.

[0132] Clause 42. The method according to any one of Clauses 28 to 41 further comprises: receiving only one set of sample pairs from the network element, the user equipment using the set of sample pairs to learn analog-to-digital converter functionality.

[0133] Clause 43. A method comprising: transmitting to a user equipment information associated with an analog-to-digital converter (ADC) response for each carrier, bandwidth, or antenna chain; receiving from the user equipment, based on the transmitted information, at least one digital predistortion model to be used; receiving from the user equipment a predistortion signal generated from the at least one digital predistortion model; and generating the ADC response based on the predistortion signal.

[0134] Clause 44. The method according to Clause 43 further includes: receiving from the user equipment a capability report message including the user equipment's digital predistortion capability.

[0135] Clause 45. The method according to Clause 44 further includes: sending a capability request message to the user equipment including a request for capabilities of the user equipment.

[0136] Clause 46. The method according to any one of Clauses 43 to 45, wherein the at least one digital predistortion model is associated with an identifier, and wherein the identifier identifies a feature of the at least one digital predistortion model.

[0137] Clause 47. The method according to Clause 46 further includes: receiving the identifier from the user equipment.

[0138] Clause 48. The method according to Clause 46, wherein the features of the at least one digital predistortion model include at least one of the following: at least one analog-to-digital converter resolution supported by the at least one digital predistortion model, at least one carrier frequency supported by the at least one digital predistortion model, and at least one bandwidth supported by the at least one digital predistortion model.

[0139] Clause 49. The method according to Clause 46 further comprises: receiving from the user equipment a request to change the at least one digital predistortion model associated with the identifier; or sending to the user equipment a request to change the at least one digital predistortion model associated with the identifier.

[0140] Clause 50. The method according to Clause 46 further comprises: requesting a change in the identifier associated with the at least one digital predistortion model based on operational changes in the power amplifier of the user equipment and the analog-to-digital converter of the device.

[0141] Clause 51. The method according to any one of Clauses 43 to 50, wherein the information sent to the user equipment includes at least one of the following: some parameters of the analog-to-digital converter function or the complete analog-to-digital converter function, an approximation of the analog-to-digital converter function, a set of sample pairs corresponding to the input and the corresponding analog-to-digital converter output, and a set of received signal samples and their associated signal-to-noise ratio.

[0142] Clause 52. The method according to any one of Clauses 43 to 51 further includes: determining the data collection process together with the user equipment.

[0143] Clause 53. The method according to any one of Clauses 43 to 52 further comprises: sending assistance to the user equipment for selecting a loss function for training the at least one digital predistortion model.

[0144] Clause 54. The method according to any one of Clauses 43 to 53 further comprises: sending only one set of sample pairs to the user equipment, the user equipment using the set of sample pairs to learn analog-to-digital converter functionality.

[0145] Clause 55. An apparatus comprising: means for receiving from a network element information associated with an analog-to-digital converter response for each carrier, bandwidth, or antenna chain; means for determining at least one digital predistortion model based on the information received from the network element; and means for transmitting the determined at least one digital predistortion model to the network element.

[0146] Clause 56. The apparatus according to Clause 55 further includes: a component for sending a capability report message including the digital predistortion capability of the apparatus to the network element.

[0147] Clause 57. The apparatus according to Clause 56 further includes: a component for receiving from the network element a capability request message including a request for user equipment capabilities.

[0148] Clause 58. The apparatus according to any one of Clauses 55 to 57 further includes: a component for performing predistortion of the signal based on the determined at least one digital predistortion model.

[0149] Clause 59. The apparatus according to any one of Clauses 55 to 58 further includes: components for associating an identifier with the at least one digital predistortion model, wherein the identifier identifies a feature of the at least one digital predistortion model; and components for transmitting the identifier to the network element.

[0150] Clause 60. The apparatus according to Clause 59, wherein the features of the at least one digital predistortion model include at least one of the following: at least one analog-to-digital converter resolution supported by the at least one digital predistortion model, at least one carrier frequency supported by the at least one digital predistortion model, and at least one bandwidth supported by the at least one digital predistortion model.

[0151] Clause 61. The apparatus according to Clause 59 further includes: means for requesting the network element to change the at least one digital predistortion model associated with the identifier, or means for receiving from the network element a request to change the at least one digital predistortion model associated with the identifier.

[0152] Clause 62. The apparatus according to Clause 59 further includes: a component for requesting a change in the identifier associated with the at least one digital predistortion model based on an operational change in the power amplifier and analog-to-digital converter of the network elements of the apparatus.

[0153] Clause 63. The apparatus according to any one of Clauses 55 to 62, wherein the information received from the network element includes at least one of the following: some parameters of the analog-to-digital converter function or the complete analog-to-digital converter function, an approximation of the analog-to-digital converter function, a set of sample pairs corresponding to the input and the corresponding analog-to-digital converter output, and a set of received signal samples and their associated signal-to-noise ratio.

[0154] Clause 64. The apparatus according to any one of Clauses 55 to 63 further includes: components for determining a data collection process together with the network element, wherein the apparatus is configured for labeled uplink transmission, or wherein samples received by the apparatus from the network element are generated offline using locally generated signals.

[0155] Clause 65. The apparatus according to any one of Clauses 55 to 63 further includes: a component for training the at least one digital predistortion model based on a selected loss function.

[0156] Clause 66. The apparatus according to Clause 65 further includes: a component for receiving assistance from the network element to select the loss function for training the at least one digital predistortion model.

[0157] Clause 67. The apparatus according to Clause 66, wherein the loss function is configured to perform at least one of the following: minimizing the error between transmitted bits and reconstructed bits, minimizing the error between transmitted symbols and reconstructed symbols, minimizing the error between the analog-to-digital converter output and the original transmitted orthogonal frequency division multiplexed signal, and minimizing the error between the analog-to-digital converter output and the quantized version of the original orthogonal frequency division multiplexed signal that has undergone quantization.

[0158] Clause 68. The apparatus according to any one of Clauses 55 to 67 further includes: a component for reconstructing the analog-to-digital converter function via the at least one digital predistortion model.

[0159] Clause 69. The apparatus according to any one of Clauses 55 to 68 further includes: a component for receiving only one set of sample pairs from the network element, the apparatus using the set of sample pairs to learn analog-to-digital converter functionality.

[0160] Clause 70. An apparatus comprising: means for transmitting to a user equipment information associated with an analog-to-digital converter response for each carrier, bandwidth, or antenna chain; means for receiving from the user equipment at least one digital predistortion model to be used based on the transmitted information; means for receiving from the user equipment a predistortion signal generated from the at least one digital predistortion model; and means for generating the analog-to-digital converter response based on the predistortion signal.

[0161] Clause 71. The apparatus according to Clause 70 further includes: a component for receiving from the user equipment a capability report message including the digital predistortion capability of the user equipment.

[0162] Clause 72. The apparatus according to Clause 71 further includes: a component for sending a capability request message to the user equipment including a request for capabilities of the user equipment.

[0163] Clause 73. The apparatus according to any one of Clauses 70 to 72, wherein the at least one digital predistortion model is associated with an identifier, and wherein the identifier identifies a feature of the at least one digital predistortion model.

[0164] Clause 74. The apparatus according to Clause 73 further includes: a component for receiving the identifier from the user equipment.

[0165] Clause 75. The apparatus according to Clause 73, wherein the features of the at least one digital predistortion model include at least one of the following: at least one analog-to-digital converter resolution supported by the at least one digital predistortion model, at least one carrier frequency supported by the at least one digital predistortion model, and at least one bandwidth supported by the at least one digital predistortion model.

[0166] Clause 76. The apparatus according to Clause 73 further includes: means for receiving from the user equipment a request to change the at least one digital predistortion model associated with the identifier, or means for sending to the user equipment a request to change the at least one digital predistortion model associated with the identifier.

[0167] Clause 77. The apparatus according to Clause 73 further includes: a component for requesting a change in the identifier associated with the at least one digital predistortion model based on an operational change in the power amplifier of the user equipment and the analog-to-digital converter of the apparatus.

[0168] Clause 78. The apparatus according to any one of Clauses 70 to 77, wherein the information transmitted to the user equipment includes at least one of the following: some parameters of the analog-to-digital converter function or the complete analog-to-digital converter function, an approximation of the analog-to-digital converter function, a set of sample pairs corresponding to the input and the corresponding analog-to-digital converter output, and a set of received signal samples and their associated signal-to-noise ratio.

[0169] Clause 79. The apparatus according to any one of Clauses 70 to 78 further includes: a component for determining the data collection process together with the user equipment.

[0170] Clause 80. The apparatus according to any one of Clauses 70 to 79 further comprises: a component for sending assistance to the user equipment for selecting a loss function for training the at least one digital predistortion model.

[0171] Clause 81. The apparatus according to any one of Clauses 70 to 80 further comprises: a component for sending only one set of sample pairs to the user equipment, the user equipment using the set of sample pairs to learn analog-to-digital converter functionality.

[0172] Clause 82. A non-transitory computer-readable medium comprising program instructions stored thereon for performing a method according to any one of Clauses 28 to 54.

[0173] Clause 83. An apparatus comprising a circuit system configured to cause the apparatus to perform the method according to any one of Clauses 28 to 54.

Claims

1. A device for communication, comprising: At least one processor; as well as At least one memory, including computer program code, which, when executed by the at least one processor, causes the device to at least: Receive information from network elements that is associated with the analog-to-digital converter response for each carrier, bandwidth, or antenna chain; Based on the information received from the network elements, at least one digital predistortion model is determined; and The determined digital predistortion model is sent to the network element.

2. The apparatus of claim 1, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: Send a capability report message to the network element, including the digital predistortion capability of the device.

3. The apparatus of claim 2, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: Receive capability request messages from the network elements, including requests for user equipment capabilities.

4. The apparatus according to any one of claims 1 to 3, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: The predistortion of the signal is performed based on the determined at least one digital predistortion model.

5. The apparatus according to any one of claims 1 to 3, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: Associating an identifier with the at least one digital predistortion model, wherein the identifier identifies a feature of the at least one digital predistortion model; and The identifier is sent to the network element.

6. The apparatus of claim 5, wherein the feature of the at least one digital predistortion model includes at least one of the following: At least one analog-to-digital converter resolution supported by the at least one digital predistortion model. At least one carrier frequency supported by the at least one digital predistortion model, and At least one bandwidth supported by the at least one digital predistortion model.

7. The apparatus of claim 5, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: The network element is requested to change the at least one digital predistortion model associated with the identifier, or Receive a request from the network element to change the at least one digital predistortion model associated with the identifier.

8. The apparatus of claim 5, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: Based on operational changes in the power amplifier and analog-to-digital converter of the device, a change in the identifier associated with the at least one digital predistortion model is requested.

9. The apparatus according to any one of claims 1 to 3, wherein the information received from the network element comprises at least one of the following: Some parameters of the analog-to-digital converter function or the complete analog-to-digital converter function. The approximation of the function of the analog-to-digital converter. A set of sample pairs corresponding to the input and the corresponding analog-to-digital converter output, and A set of received signal samples and their associated signal-to-noise ratios.

10. The apparatus according to any one of claims 1 to 3, wherein the computer program code, when executed by the at least one processor, further causes the apparatus to at least: Together with the network elements, the data collection process is determined. The device is configured for labeled uplink transmission, or the samples received by the device from the network element are generated offline using locally generated signals.

Images