Hybrid ai-tuned implementations for modeling serdes RX and BER prediction

Abstract

Techniques for training a simulation of a Serializer / Deserializer (SerDes) receiver are described. For example, one implementation of a method includes: based on a channel impulse response (CIR) input, selecting a first receiver equalization model from a plurality of receiver equalization models using a first optimization process, the first receiver equalization model having a corresponding set of model parameters; converting the CIR to an equalized CIR based on the selected receiver equalization model and corresponding set of model parameters; predicting a first bit error rate (BER) based on the first receiver equalization model and corresponding set of model parameters using a second optimization process; and based on a difference between the first BER and an expected or known BER, selecting a second receiver equalization model from the plurality of receiver equalization models using the first optimization process.

Description

Atty. Docket No.: 4906P109167WO01 SPECIFICATION HYBRID AI-TUNED IMPLEMENTATIONS FOR MODELING SERDES RX AND BER PREDICTION CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Application No. 63 / 598,539, filed November 13, 2023, which is hereby incorporated by reference. TECHNICAL FIELD

[0002] Embodiments of the invention relate to the field of communication systems, and in particular, to hybrid AI-tuned implementations for modeling serializer / deserializer (SerDes) receivers and bit error rate (BER) prediction. BACKGROUND

[0003] Serializer / Deserializer (SerDes) technology is employed in modern communication systems for high-speed data transmission. In the design process, developers typically utilize simulation tools (e.g., tools which include input / output buffer information specification (IBIS) - algorithmic modeling interface (AMI) and Channel Operating Margin (COM) capabilities) to model and analyze the performance of SerDes systems. In the case of high-speed SerDes design, developers often face challenges due to the lack of transparent knowledge or readily available models of SerDes from vendors (e.g., on the receiver (Rx) side). Therefore, gaining an initial understanding and characterization of a SerDes system may involve performing measurements, employing simplified simulation tools, and / or relying on existing documentation. These approaches may allow developers to gain insights into parameters (e.g., critical parameters) related to signal integrity, eye diagrams, power consumption, bit error rates (BER), and / or other design considerations.

[0004] However, when developing SerDes chips or different channel designs, utilizing previous measurements and / or models may not be feasible. As a result, developers may fine-tune their designs for each unique scenario, which may include manually tailoring the system parameters and / or equalization techniques to optimize performance. In some situations, a desirable BER outcome might not be sustainable on the physical link (e.g., as designed), prompting the developers to contemplate an overhaul and / or a reconstruction of the physical link. However, the process of manual refinement may involve a significant investment of time and / or resources, which may impede the pace of product development. Therefore, moreAtty. Docket No.: 4906P109167WO01 streamlined and / or efficient approaches may help tackle these issues and may help achieve accurate BER predictions and / or channel designs with favorable outcomes.

[0005] There currently exist certain challenge(s) with existing methods for modeling a SerDes Rx (e.g., when confronted with the multifaceted nature of the technology). These limitations stem from the complex nature of the Rx and the diverse range of challenges encountered in the field. The intrinsic complexity of the SerDes receiver itself makes it difficult to comprehend and / or predict Inter-symbol interference (ISI) (e.g., accurately). ISI, which results in unwanted signal tails overlapping with effective symbols in high-speed communication, may pose a significant obstacle to accurate modeling of SerDes.

[0006] In the context of existing method(s) for characterizing the SerDes receiver and / or estimating BER with new channel configurations and SerDes chips, existing modeling methods and techniques often face shortcomings.

[0007] Some methods are measurement-based. While measurements can provide empirical data about the behavior of a specific SerDes system, they may not scale well to accommodate the rapid testing and iteration typically performed for various channel configurations and SerDes chips. Each new configuration typically involves a new round of measurements, which can be resource-intensive and / or time-consuming.

[0008] Some methods use simplified simulation tools (e.g., IBIS-AMI and COM). While useful for initial assessments, they may not offer the granularity and / or adaptability needed for rapidly changing design scenarios. These tools, for example, may struggle to predict BER accurately for entirely new configurations or chips.

[0009] Some methods use vendor documentation and models. Vendor documentation and pre- existing models often serve as valuable references, but they may not always encompass the full spectrum of potential design variations or specific characteristics of a new SerDes chip or channel setup.

[0010] Additionally, noise, jitter, and crosstalk further compound the difficulties faced by existing modeling techniques. Consequently, existing modeling methods often fall short in capturing the nuanced interactions and intricate dynamics within the SerDes Rx, which leads to inaccurate Bit Error Rate (BER) predictions. Addressing these challenges and improving the accuracy of performance modeling is crucial for optimizing the design and performance of SerDes-based communication systems.

[0011] Equalization operations are often performed in SerDes characterization. Some existing equalization technique(s) can be time-consuming and require specialized expertise, which may involve sweeping through hundreds or more combinations of tap values. Some other existing equalization technique(s) apply artificial intelligence (AI) based approaches (which use machineAtty. Docket No.: 4906P109167WO01 learning (ML) and neural networks) to high-speed channel design and SerDes equalization optimization (See, e.g., M. Ahadi Dolatsara, “A Simplified Constrained Bayesian Optimization Approach to Optimize the Tx Equalization in SerDes Channels,” in IEEE Letters on Electromagnetic Compatibility Practice and Applications, vol. 5, no. 2, pp. 41-47, June 2023, doi: 10.1109 / LEMCPA.2023.3247777 (“Dolatsara”); Hsinho Wu, Masashi Shimanouchi, Mike Peng Li. “COM & IBIS-AMI: How They Relate & Where They Diverge”, Designcon 2019 (“Wu”); Z. Kiguradze et al., "Bayesian Optimization for High-Speed Channel Equalization," 2019 Electrical Design of Advanced Packaging and Systems (EDAPS), Kaohsiung, Taiwan, 2019, pp. 1-3, doi: 10.1109 / EDAPS47854.2019.9011654 (“Kiguradze”); Song S, Sui Y. System level optimization for high-speed serdes: Background and the road towards machine learning assisted design frameworks. Electronics. 2019 Oct 28;8(11):1233 (“Song”)).

[0012] For example, Bayesian optimization, which is a sample-efficient method for optimizing black-box functions (such as signal-to-noise (SNR)-like metrics), has been employed (see Dolatsara, Wu). The Bayesian optimization approach builds a surrogate model to guide the sampling of promising parameters, which expedites the calculation of equalization values efficiently. The Bayesian optimization for equalization optimization has been demonstrated to be more effective when compared to exhaustive search (see Dolatsara, Wu). The random forest method has been used for Tx equalization (see Song).

[0013] However, these approaches do not always account for generalization across chip variations or include BER predictions within the model. Another approach involves a model-free reinforcement learning agent for equalization optimization (see Kiguradze). Nevertheless, RL- based methods often demand a large number of interactions with the environment to discover optimal policies, which can be impractical or costly in certain applications. Hence, there is a need for more data-efficient and / or efficient method(s) for SerDes modeling and BER predictions.

[0014] Evaluation of channel attribute(s) is part of SerDes design. For example, an existing approach using COM requirements has proven to be effective in high-speed SerDes design and involves a meticulous evaluation of channel attributes utilizing a conceptual Rx model. (See Backplanes, Operation Over, and Copper Cables. “IEEE P802. 3bj™ / D3. 0D3. 1 Draft Standard for Ethernet Amendment X2: Physical Layer Specifications and Management Parameters for 100 Gb / s.” (2013)) (“Backplanes”). Nonetheless, this approach does not address the existing performance disparity between real-world conditions and the virtual model. Further, such an existing approach cannot serve as a direct mathematical model for BER prediction.Atty. Docket No.: 4906P109167WO01 BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The invention may best be understood by referring to the following description and accompanying drawings that are used to illustrate embodiments of the invention. In the drawings:

[0016] Figure 1 is a block diagram of a system for modeling a SerDes Rx and predicting BER according to some embodiments.

[0017] Figure 2 illustrates a flow chart of an example COM-compatible differentiable SerDes Rx model according to some embodiments.

[0018] Figure 3 illustrates a block diagram of an exemplary system that implements a hybrid AI-tuned training framework that automatically fits parameters according to some embodiments.

[0019] Figure 4A illustrates a flow chart for training a SerDes receiver model in accordance with some embodiments.

[0020] Figure 4B illustrates a flow chart of a hybrid AI-tuned training framework in accordance with some embodiments.

[0021] Figure 5 illustrates an example COM method.

[0022] Figure 6 illustrates result distributions of a test dataset, including an FFE distribution, a channel distribution and an SNDR distribution.

[0023] Figure 7 illustrates repeated bit error rate (BER) results across various test samples.

[0024] Figure 8 illustrates training results for four different system parameters.

[0025] Figure 9 illustrates a distribution of FOM values on a validation dataset.

[0026] Figure 10 illustrates a distribution of BER log error values on the validation dataset with different COM values.

[0027] Figure 11 illustrates a comparison of BER prediction error for two training rounds.

[0028] Figure 12 illustrates one embodiment of an access network with a plurality of network nodes and a core network with core network node.

[0029] Figure 13 illustrates one example of a user equipment (UE) device.

[0030] Figure 14 illustrates one example of a network node.

[0031] Figure 15 illustrates one embodiment of a host.

[0032] Figure 16 illustrates an example showing communication between a host, a network node, and a UE. DETAILED DESCRIPTION A. Introduction

[0033] Figure 5 illustrates an example COM method according to Institute of Electrical and Electronics Engineers (IEEE) COM model analysis. The calculation of COM is performed asAtty. Docket No.: 4906P109167WO01 follows. At block 502, filtered S-Parameters are converted into channel impulse response to capture system response and, at block 504, an exhaustive search is performed to optimize SNR using equalization settings and Figure Of Merit (FOM) values. The exhaustive search iteratively identifies equalizer settings that yield the highest FOM value, maximizes SNR, and enhances signal quality. Certain optimization operations explore Feed Forward Equalization (FFE) and Continuous Time Linear Equalization (CTLE) combination(s) which provide the highest SNR, where parameters such as tap coefficients, equalization length, gain, pole / zero locations, and filter order are considered. Optimization operations may also include using FOM values to pinpoint optimal equalization configurations from which parameters are derived such as: Transmitter RMS noise (σ_TX), RMS of ISI cursors (σ_ISI), Jitter converted RMS voltage (σ_J), Coding RMS noise voltage (σ_XT), and Receiver noise from eta_0 (σ_N). Another optimization operation is an implementation of the Mueller-Muller algorithm to determine the sample point, denoted as t_s. This algorithm estimates the optimal sampling instant by analyzing the waveform characteristics, enabling precise timing alignment, and minimizing timing errors. Another related optimization operation is calculation of the Signal Amplitude A_s at Sample Point t_s. The amplitude represents the instantaneous voltage or current level at the specified sampling instant and is for quantifying the signal quality. Based on the searched results from the above-mentioned optimization operations, FOM is calculated as follows: FOM = 10 log ^^ i. ^At block 506, a convolution is performed to address ISI, crosstalk, and other noise effects. At operation 507, the Self-Channel Interference (SCI) probability density function (PDF) (sci-PDF) is calculated based on the equalized Symbol Bit Response (SBR) and the optimized equalization settings identified at block 504.

[0034] At operation 508, a jitter PDF (also referred to as dual-dirac PDF) is derived from derivative of Thru SBR waveform. This PDF characterizes the timing variations present in the received signal. At operation 509, random Tx and Rx noise is incorporated into the noise PDF. These noise sources, which can arise from various factors (such as thermal noise), impact the overall system noise. Their effects are considered in the noise PDF. At operation 510, Crosstalk PDF is calculated based on crosstalk noise.

[0035] At operation 520, a cumulative density function (CDF) is computed from combined PDFs 511 for different BER levels. This step is done by discrete numerical computation, where the resolution of the CDF is based on the pre-selected bin width. At operation 522, a COM value is derived using the calculated CDF from operation 520 based on a target BER. The COM valueAtty. Docket No.: 4906P109167WO01 quantifies the margin between the desired signal and the noise present in the system. The COM value serves as a metric (e.g., a crucial metric) to assess system performance and / or robustness. B. Hybrid AI-tuned Implementations for Modeling SerDes Receivers and Bit Error Rate (BER) prediction

[0036] Embodiments described herein accurately characterize a SerDes Rx and predict BER performance while overcoming the challenges associated with manual testing and evaluation of different channel settings for high-speed SerDes systems. Some of these embodiments implement a digital twin model of SerDes Rx to enable efficient and accurate BER predictions, which facilitates characterizing the SerDes Rx performance in a cost-effective manner. In some implementations, differentiable modeling is combined with Bayesian optimizations within a unified framework to implement a hybrid training approach that includes techniques to determine the Rx characteristics including Continuous Time Linear Equalization (CTLE) and intrinsic noise to provide accurate BER predictions.

[0037] Embodiments described herein offer a variety of technical advantages. For example, the hybrid approach as described herein efficiently explores the parameter space, which reduces exhaustive search time. By using a Bayesian optimization, the hybrid approach accelerates this exploration process, which saves time by reducing exhaustive search efforts. Some implementations of the hybrid approach described herein also quickly and accurately predict BER, which provides valuable insights for system design.

[0038] Additionally, at least some of the differentiable modeling techniques described herein explicitly incorporate known physical principles and constraints into the modeling process. Such modeling techniques can leverage prior knowledge about the underlying system, which makes the system more interpretable and less data-hungry. Embodiments using the COM-compatible SerDes Rx differentiable model accelerate and guide passive channel design by suggesting optimal COM margins.

[0039] Furthermore, the set of Rx equalizer models from the model zoo provide the COM- compatible SerDes Rx differentiable model’s generalization capability, allowing the performance of SerDes receivers to be predicted across a wide range of scenarios and operating conditions. These techniques can be easily expanded to a broader scope than described herein including, for example, Tx (transmitter) equalization or different SerDes design parameters.

[0040] The described hybrid training empowers self-evolution within a digital twin model (also referred to as a simulator) of SerDes Rx. This self-evolving capability allows the model to continuously refine and improve its performance over time, adapting to changing requirements and capturing the dynamics of new SerDes designs. Some of these hybrid trainingAtty. Docket No.: 4906P109167WO01 implementations also provide a flexible framework capable of seamlessly incorporating and optimizing improvement(s) within the SerDes communication domain.

[0041] Figure 1 is a block diagram of a system for modeling a SerDes Rx and predicting BER according to some embodiments. The channel measurement of a new channel 101 is processed at the block of link scattering-parameter (S-parameter) modeling 102, which includes at least one of: link S-parameter measurement, device package modeling, termination modeling and s- parameter cascading, and correction and identification. The output of the link S-parameter modeling block 102 is a time-domain channel impulse response (CIR) which is provided to the SerDes Tx block 104 (also referred to as differentiable Tx simulator) for Feed Forward Equalization (FFE) equalizing. A differentiable Tx simulator 104 applies Tx FFE on the CIR to obtain an un-equalized Single Bit Response (Tx SBR) signal.

[0042] In various embodiments described herein, the un-equalized Tx SBR signal is input to a Bayesian Optimization block 106 which performs an evaluation 106B of selected Rx equalizer models 106A. For example, the evaluation 106B may produce one or more FOM values and ISI values (and / or any other type of metrics useful for characterizing the operation of the selected Rx equalizer models 106A). In some embodiments, the Bayesian optimization block 106 selects each Rx equalizer model 106A from a model zoo 107, which includes a collection of Rx equalizer models (e.g., CTLE models).

[0043] If the FOM result produced by the evaluation 106B does not meet or exceed a pre- defined threshold, determined at 108, the Bayesian optimization block 106 attempts to improve the modeling results by selecting another Rx equalizer model 106A from the model zoo 107. Each equalizer model defined in the model zoo 107 may have a different number of parameters and may result in a different level of computational complexity for the evaluation 106B (e.g., for determining the corresponding FOM). The Bayesian optimization block 106 iterates in this manner, selecting new Rx equalizer models 106A until the evaluation 106B produces a FOM which meets the threshold at 108.

[0044] In some implementations, this Bayesian optimization block 106 is performed at least once for each channel of a product (e.g., on a board), many of which are also likely to have optimization challenges. In some implementations, the Bayesian optimization block 106 performs these operations in parallel to accelerate the training (e.g., selecting Rx equalizer models 106A and performing evaluations 106B for each channel in parallel, until the corresponding FOM values meet the specified threshold at 108.

[0045] When the FOM for a selected Rx equalizer model 106A meets the corresponding threshold at 108, the Rx equalizer model, model configuration, and the equalized channel impulse response (CIR) (e.g., the equalized SBR) are both sent (at 109) to the pre-trainedAtty. Docket No.: 4906P109167WO01 differentiable Rx simulator block 110 (hereinafter “differentiable Rx simulator”) for further processing including BER prediction 111. In some embodiments, the differentiable Rx simulator 110 is pretrained using collected measurements with logarithmic BER error as the loss function.

[0046] In some embodiments, the differentiable Rx simulator 110 is a pre-trained, Channel Operation Margin (COM)-compatible differentiable Rx simulator with which the Rx equalizer model and the equalized CIR (e.g., equalized SBR) 109 are integrated to generate corresponding Bit Error Rate (BER) predictions 111. The operations of the differentiable Rx simulator 110 take into account the intricacies of the communication channel, which enables precise and efficient BER prediction without the need for exhaustive simulations.

[0047] As described further below, some implementations of the differentiable Rx simulator 110 utilize a hybrid tuning method within a unified framework of Bayesian optimization and differentiable modeling. This unified framework allows for the seamless integration of intelligent optimization and advanced modeling techniques for improved SerDes system performance. Some hybrid AI-tuned implementations for modeling SerDes Rx and BER prediction include a COM-compatible differentiable SerDes Rx model, Bayesian optimization for Rx equalizers, and / or automatic parameter fitting facilitated by AI hybrid tuning.

[0048] While embodiments are described herein with respect to a SerDes Rx simulator to provide fast and accurate BER prediction for a high-speed link, these embodiments may also be used to provide design guidance for Tx settings as well as margin design, e.g., by effectively learning complex interaction and relationships within the SerDes Rx.

[0049] The differentiable Rx simulator 110 described herein may adopt some aspects of the COM model as previously described to construct the SerDes Rx. By implementing a COM- compatible Rx model, projected SerDes Rx performance is aligned with specified COM margins, which resolves the above-mentioned issue of performance disparity between the real- world conditions and the COM model. In the context of COM analysis, in some embodiments, the differentiable Rx simulator 110 performs noise computation 110A using a plurality of noise probability distribution functions (PDFs), the results of which are combined with a noise Cumulative Density Function (CDF) to determine the margin between the signal’s quality and the noise levels.

[0050] The implementation of a COM-compatible differentiable SerDes Rx model described herein enhances the Rx model’s accuracy by incorporating BER prediction through the utilization of total noise CDF, and treating COM values as fitting parameters in the COM- compatible differentiable SerDes Rx model. By leveraging data-driven AI hybrid tuning techniques, the COM value can be trained and then used as an indicator in the total noise CDF for BER predictions.Atty. Docket No.: 4906P109167WO01

[0051] Specifically, COM values are not directly computed (e.g., as described with respect to operation 522 in Figure 5) as one of the system parameters, along with dual-Dirac amplitude, noise spectral density, Gaussian noise variance, etc. Rather, these parameters are treated as trainable variables that are determined through a set of tested data samples. When the values of these system parameters are obtained, the CDF of total noise can be explicitly computed. This CDF of total noise can then be utilized with the COM value to estimate the BER.

[0052] Figure 2 illustrates a flow chart of an example COM-compatible differentiable SerDes Rx model according to some embodiments. Within the COM-compatible differentiable SerDes Rx model, each computation component may be implemented as a pure function. For example, the transformed CIR (e.g., from s-parameter modeling block 102) is denoted herein as function 200, which is a time-domain vector. The Tx and Rx filters 205 are denoted herein as transfer functions in the frequency domain, "#$^%! and "&$^%!, respectively. The Tx FFE equalizer 207 is denoted as "''(^%!and the Rx equalizer 214 is denoted as "()^%!, where the Continuous Time Linear Equalization (CTLE) is denoted as "*+,(^%!. Due to the CDF computation and BER prediction being based in the time domain, an Inverse Fast Fourier transform (IFFT) is applied on the frequency-domain transfer function(s), and a convolution is then performed for the time domain sequences. In one example, the time-domain equalized CIR is obtained using: ℎ^^^ ! = ℎ^^^^^ !

[0053] In the above example, "()^%! = "''(^%! ⋅ "*+,(^%! and hBCD^⋅! is the DecisionFeedback Equalizer (DFE) function and nBCDis the number of taps in DFE.

[0054] In these embodiments,is transformed from the Sdd21 of the s-parameter measurement from block 102 in Figure 1. The resulting optimal equalizer transfer function from the Bayesian search (as described in detail later herein) is denoted as "(∗)^%!.

[0055] The obtained equalized CIR from the Rx equalizer 214 is denoted as230 and is sent to the differentiable computation block 232 in Figure 2. Blocks 233, 235, 237, and 239 of the differentiable computation block 232 calculate corresponding noise components 234, 236, 238, and 240, respectively, based on ℎ^∗^^ ! 230. In particular, in the illustrated embodiment, at 233, sci-PDF, denoted as GHIJ^K!, is calculated by extracting the ISI noise component fromAtty. Docket No.: 4906P109167WO01 ℎ^∗^^ !; at 235, dual-Dirac PDF, denoted as GLL^K! 236 is calculated by dual-dirac jitter sequence and jitter amplitude; at 237, the random Gaussian noise PDF, denoted as pG(y) 238, is calculated based on gaussian variance. Optionally, at 239, a crosstalk PDF 240 (also referred to as Xtalk PDF) is calculated.

[0056] A combined PDF block 241 uses these values to determine a total noise PDF, G^K!242 (also referred to as combined PDF), which is input to a cumulative distribution function (CDF) block 243 which generates a total noise distribution, M^K! 244 representing the cumulative probability distribution of the noise levels in the given high-speed communication link. As the noise CDF shifts towards higher noise levels (e.g., greater noise amplitude or timing jitter), the likelihood of errors in the received signal increases, resulting in higher BER. The calculated total noise distribution 244 is used to determine the probability associated with different noise levels. @, theUVWX noise is obtained by KZ =[\] , and then FZ ^Xpredicted BER 248 is obtained from the noise CDF as M^K = KZ!.

[0057] Some of the above computation functions (e.g., those involved in the numerical computation of PDF) are inherently non-differentiable. As such, some implementations use approximations to make these functions differentiable with respect to their inputs. For example,in the computation of BER, when searching for value K = KZ in the noise CDF, M^K!, Gaussianapproximation may be applied by first calculating ^WX = ∑exp −^KZ − K!9 / 2d9W. The calculated^WXis a Gaussian approximation for the delta function. The predicted BER 248 is then obtainedviah efg = ⋅ M^K! . These approximations provide an advantage in that thedifferentiable model remains suitable for training and optimization processes while providing reasonable accuracy for capturing the system behavior.

[0058] In response to the predicted BER 248 calculated based on the current model parameters, the loss function block 246 calculates the loss error between the predicted BER 248 and the True BER 250 of the given channel input. In response to this loss error, the differentiable model can apply data-driven gradient descent operations for updating the model parameters until the error converges. The dotted arrows in Figure 2 indicate that the corresponding computations are differentiable between blocks connected by the arrows. In the illustrated embodiment in Figure 2, the computation of parameters for Rx equalizer 214 implements an exhaustive iterative search based on an intermediate FOM metric for each channel input, where the FOM value is not directly differentiable with respect to the equalizer parameters. Because such an exhaustive iterative search may be time consuming and / or inefficient, embodiments described herein implement Bayesian optimization techniques to improve the efficiency of the exhaustive iterative search approach for the Rx equalizer 214.Atty. Docket No.: 4906P109167WO01

[0059] The COM analysis requires optimizing equalizer parameters to maximize signal-to- noise ratio. In some embodiments, if fiJdenotes the parameter set of the equalizer model j, Bayesian optimization is used to find the global optimum of expensive black-box functions0OP = %^k!, where k ∈ fiJ. It finds the global optimum by building a statistical model of%^k! using a Gaussian process (GP): f^x! ∼ GP 5μZ^k!, k^x, xs!?, where tu^k! is the meanfunction and v^k, k′! is the covariance kernel. After evaluating %^k! at some points, the GPposterior distribution is:dependon the observed pointsand valuesThis statistical model is used to select new points k to evaluate by trading off exploration and exploitation. After each evaluation the model is updated. An acquisition function determines the next point by maximizing expected improvement and / or other criteria. By modeling uncertainty and automatically balancing exploration and exploitation, Bayesian optimization can find the optimum using far fewer evaluations than current techniques.

[0060] Because Bayesian optimization with a single fixed equalizer model can converge to a suboptimal solution on channels whose characteristics are poorly matched to the model structure, the Bayesian optimizations described herein may be implemented with a model zoo 107 providing a variety of candidate Rx equalizer models 106A (e.g., to improve robustness and / or adaptability across a diverse range of channels). The model zoo 107 thus provides an advantage by allowing the optimization to switch equalizer models when appropriate. Specifically, a check is performed to determine whether the optimized FOM meets a minimum threshold after each Bayesian optimization run. If the threshold is not met, a different Rx equalizer model 106A is selected from the model zoo 107 and the optimization is repeated using the new model.

[0061] By way of example, and not limitation, in a Bayesian optimization using a model zoo,P = yPF, P9, ... P}{ is the model zoo containing ~ equalizer models. For a new channel, thefollowing operations are performed: (a) Initialize k=1 and select initial model P^from P; (b) Place GP prior on objective function %^k; P^! ~ ^M^0, v^k, k′!!;(c) Generate samples ykF, k9, ... k^{ using acquisition function ^^k!;(d) Observe FOM values K = y%^kF; P^!, %^k9; P^!, ... , %^k^; P^!{;(e) Update GP posterior G^%|^, K!;(f) Find optimal parameters k∗ = ^^T^^k %^k; P^!;(g) If %^k∗; P^! ≥ 0OP#^&^H^u^L, return k∗; else, set v = v + 1 and go to (b),Atty. Docket No.: 4906P109167WO01where %^k; P^! = ^^g objective function under model P^, and v^k, k′! = ^M kerneldefining covariance, and G^%|^, K! = ^M posterior distribution after observing K at ^, and k∗ isthe optimal equalizer parameters. Thus, the additional operation(s) performed in a Bayesian optimization with a model zoo iterate over models P^until the FOM threshold is exceeded. The GP surrogate and acquisition sampling may remain unchanged between iterations (i.e., they are merely conditioned on the selected P^).

[0062] Figure 3 a block diagram of an exemplary system that implements a hybrid AI-tuned training framework that automatically fits parameters according to some embodiments. As previously described, the framework comprises a Bayesian Optimization block 310 and Differentiable BER computation block 332.

[0063] S parameter modeling block 304 converts channel measurements 302 (e.g., data in .s4p format) into an un-equalized channel impulse response (CIR). Differentiable Tx simulator block 306 applies the un-equalized impulse response to the Tx equalizer, implemented by feed forward equalization (FFE) 308, to obtain the Tx-equalized sequence to be processed by the Bayesian optimization block 310.

[0064] In response to the Tx-equalized sequence, the Bayesian optimization block 310 performs operations as described herein to select the optimal receiver Rx equalizers and configuration by evaluating intermediate metrics. In some embodiments, the order in which the Rx equalizers are selected is determined based on the similarity of input features to prior input features for which Rx equalizers were selected (e.g., to choose similar Rx equalizer models for similar input features). Thus, for example, the input features of prior tasks may be cross- correlated to select the best prior Rx equalizer model 106A from the model zoo 312. This technique provides the advantage of reducing complexity associated with the acquisition task for a new model.

[0065] The equalized channel impulse response (CIR) (e.g., the equalized SBR in some embodiments) 316 and Rx equalizer model and model configuration 318 determined through the Bayesian optimization are provided to the differentiable BER computation block 332 to train the COM-compatible differentiable model and obtain system parameters. A gradient flow approach may be employed which involves updating the model’s parameters iteratively, leveraging computed gradients based on a chosen loss function (e.g., logarithmic error). The loss function quantifies the discrepancy between the predicted output of the model and the true output from the labeled data samples. For example, the loss function may be defined as Mean SquaredLogarithmic Error (MSLE), i.e., ‖log^1^^6^(^! − log^M^6^^(^! ‖9. By minimizing this loss,the model parameters are adjusted to improve the accuracy of the predictions. The training process involves feeding the input data samples into the differentiable model and computing theAtty. Docket No.: 4906P109167WO01 corresponding outputs. The gradients of the loss with respect to the model parameters are then calculated using backpropagation operations. These gradients represent the direction and magnitude of the parameter updates that would reduce the loss. By applying an optimization algorithm, such as stochastic gradient descent or Adam, the model parameters are iteratively adjusted in the direction of minimizing loss.

[0066] In some embodiments, the difference between the predicted BER 340 and the actualtested BER 342 is used to generate a log error value 344 (e.g., the log error ‖log^1^^6^(^! −log^M^6^^(^! ‖9 as previously described) provided as feedback to the Bayesian optimizationblock 310 which uses the error data to inform subsequent selections of equalizer models from the model zoo 312 and / or to adjust the threshold value of the objective in the Bayesian optimizations (e.g., the minimum threshold for FOM as described above herein). For example, the Bayesian optimization block 310 may perform subsequent selections of equalizer models and / or set the FOM minimum threshold value with a goal of minimizing the log error value 344.

[0067] Combining the Bayesian optimization block 310 and differentiable BER computation block 332 into a comprehensive framework efficiently optimizes equalization settings and facilitates a self-evolution process for continuous model updates and improvements.

[0068] Figure 4A illustrates a method for training a simulation of a Serializer / Deserializer (SerDes) receiver in accordance with some embodiments described herein. The method may be implanted on the various architectures described herein, but is not limited to any particular hardware or software architecture.

[0069] At 401, based on a channel impulse response (CIR) input, a first receiver equalization model is selected from a plurality of receiver equalization models using a first optimization process, the first receiver equalization model having a corresponding set of model parameters. By way of example, and not limitation, the first optimization process may include selecting receiver equalization models from the plurality of receiver equalization models and evaluating intermediate metrics associated with each receiver equalization model selected until the first receiver equalization model is selected having intermediate metrics which meet a defined threshold. In some implementation, the first optimization process comprises the Bayesian optimization process described herein.

[0070] At 402, the CIR is converted to an equalized CIR based on the selected receiver equalization model and corresponding model configuration (e.g., set of model parameters). For example, the selected equalization model may be used to equalize the CIR to generate the equalized CIR, which is an equalized SBR in some implementations.

[0071] At 403, a first bit error rate (BER) is predicted based on the first receiver equalization model and corresponding set of model parameters using a second optimization process. By wayAtty. Docket No.: 4906P109167WO01 of example, and not limitation, the second optimization process may comprise a gradient flow process for training a differentiable Rx simulator which is to perform the predicting of the first BER. The gradient flow process may iteratively update one or more model parameters of the set of model parameters to reduce the difference between the first BER and an expected or known BER. Gradients of loss corresponding to each iteration may be determined, indicating a direction and magnitude of a subsequent update to the one or more model parameters that would reduce the loss.

[0072] At 404, based on a difference between the first BER and an expected or known BER, a second receiver equalization model is selected from the plurality of receiver equalization models using the first optimization process. By way of example, and not limitation, the difference may guide the selection of a new equalizer model from the model zoo and may also be used to update the threshold value of the objective (e.g., the FOM) in the Bayesian optimization.

[0073] Figure 4B illustrates a method for training a simulation of a Serializer / Deserializer (SerDes) receiver in accordance with some embodiments described herein. The method may be implanted on the various architectures described herein, but is not limited to any particular hardware or software architecture.

[0074] At 411, channel measurement data is processed to generate an un-equalized channel impulse response and, at 412, Tx equalization is applied to the un-equalized impulse response to generate Tx-equalized input for Bayesian optimization.

[0075] At 413, an Rx equalizer model and configuration is selected to process the Tx- equalized input and generate results, which are then evaluated. If the results do not meet a minimum threshold, determined at 415, then the process returns to 413, where a new Rx equalizer model is selected and the results are evaluated.

[0076] When the results meet the minimum threshold at 415, then at 416, a bit error rate (BER) predictor generates a BER prediction based on the provided Rx equalizer mode, model configuration, and equalized CIR (e.g., an equalized SBR in some implementations).At 417, differences between the BER prediction and the true BER values are quantified and, at 420, subsequent Bayesian optimizations may be performed based on the quantified differences, returning to 413, where a new Rx equalizer model and configuration may be selected based on the quantified differences.

[0077] At 418, one or more model parameters of a set of model parameters of the Rx equalizer model are updated in an attempt to reduce the difference between the first BER and an expected or known BER. For example, gradients of loss may be determined in each iteration indicating a direction and magnitude of a subsequent update to the model parameters that would reduce theAtty. Docket No.: 4906P109167WO01 loss. The model parameters may subsequently be updated in accordance with the direction and magnitude.

[0078] At 419, when the accuracy of the prediction is sufficient (e.g., the difference between the predicted BER and the true BER is sufficiently small), the process ends. If the accuracy of the prediction is insufficient following the iterations to update the model parameters at 418, then, at 420, subsequent Bayesian optimizations may be performed based on the quantified differences, returning to 413, where a new Rx equalizer model and configuration may be selected based on the quantified differences. C. Testing Results of Embodiments

[0079] Described below are test results of some embodiments described herein, such as the hybrid AI-tuned training framework for SerDes Rx models as described with respect to Figure 3. Data collection

[0080] The collected datasets include 480 samples from a 50G SerDes Rx with varying parameters such as jitter, insertion loss, and Transmitter SNDR. To ensure reliable input, de- embedded channel measurement data was combined with RX PCB simulated loss data. The collected data format is shown in Table 1 below. The data was collected with a focus on SNDR, FFE, and channel s-parameters, conducting BER tests and repeated BER tests for enhanced reliability. For example, sample 4 is the repeated BER test result for sample 3, where the BER tests share the same feature input. Channel measurement data is in Touchstone format (.s4p) and is measured from 100Mhz to 50Ghz with 100Mhz step. Table 1 - Example of Dataset Sample index CHANNEL type SNDR FFE BER 1 C1.s4p 27dB 2 4.2e-4 2 C1.s4p 27dB 1 4.9e-4 3 C2.s4p 27.5dB 3 5.5e-5 4 C2.s4p 27.5dB 3 5.3e-5 … … … … … 480 C40.s4p 29.5dB 5 4.8e-4

[0081] Three unique features besides the un-equalized sequence for the input data, i.e., SNDR, FFE, channel types, were identified in Figure 6, which shows the distribution of these three features. The transmitter sequence was tested through bit error rate test (BERT). For additional reliability, repeated BER tests were performed on select samples. Figure 7 shows the repeated BER results across the test samples. This distribution offers insights into the actual BERAtty. Docket No.: 4906P109167WO01 logarithmic distance, which establishes benchmarks and defines the lower limit for the predicted BER. This boundary becomes a reference point in evaluating the model’s effectiveness, ensuring that its performance does not fall below this level. Results

[0082] Following the steps shown in Figure 3 the threshold for FOM in Bayesian optimization was set as 12. This is a reasonable value that can guarantee the high-speed link to achieve good performance in BER evaluation. The Bayesian iteration was set to 50 exploration with 10 exploitation. The system parameters were trained using ADAM optimizer for 2000 steps. Figure 8 showing an example of training results for four system parameters. As shown in Figure 8, the COM value was decreasing and converged to 4.5 after 800 training steps, the dual-dirac amplitude absolute value, denoted as A_DD_abs in the figure, was increasing and converged to 0.025 after 1000 steps. There was a negative correlation between noise spectrum variance and random jitter absolute value, denoted as eta_0 and sigma_RJ_abs, respectively in the figure. This is also aligned with the COM mathematical model in Reference [1]. In the inference stage, the hybrid framework was applied with the pre-trained differentiable simulator and pre-trained Bayesian optimization on the validation dataset (the inference stage is described as in Figure 1). Three kinds of Rx equalizer models were in the model zoo for self-evolution implementation. In each round of inference, samples that failed to meet the threshold were sent through the Bayesian optimization block for the second round with different Rx equalizer. Figure 9 illustrates the distribution of FOM value on the validation dataset from first evolution round. Samples with FFE equal to 0-5 all passed the threshold for FOM value. We use ^O^fgg^(^=log^M^6^^(^! − log^1^^6^(^! to illustrate the BER LOG error in the BER inference result.And Figure 10 shows the distribution of BER log error value on the same validation dataset from first round for different COM value. The result was aligned with the training result shown in Figure 8 where COM value converged to 4.5. As the COM value increased from the ground truth, the BER log error was also increasing. It is worth noting that a positive value of BER log error indicates a pessimistic inference where predicted BER is higher than the true BER, and a negative value indicates an optimistic inference where predicted BER is better (lower) than the true BER. An optimistic inference is desired because it will leave more room for the training process. A pessimistic result normally indicates the Rx equalizer is not sufficient to achieve the appropriate noise PDF for BER computation. As shown in Figure 11, samples with FFE equaling to 6 achieved much better BER prediction performance for the second round compared with first round. This is because the Rx equalization configuration only obtained FOM value around 8, which was far away from the desired FOM value (>12). In the second run, FOM value for mostAtty. Docket No.: 4906P109167WO01 samples achieved above 11.8. Therefore, we can see an obvious improvement in the BER prediction at the second round as well. D. Example Environments

[0083] Figure 12 shows an example of a communication system 1200 in accordance with some embodiments. In the example, the communication system 1200 includes a telecommunication network 1202 that includes an access network 1204, such as a radio access network (RAN), and a core network 1206, which includes one or more core network nodes 1208. The access network 1204 includes one or more access network nodes, such as network nodes 1210a and 1210b (one or more of which may be generally referred to as network nodes 1210), or any other similar 3rd Generation Partnership Project (3GPP) access nodes or non-3GPP access points. Moreover, as will be appreciated by those of skill in the art, a network node is not necessarily limited to an implementation in which a radio portion and a baseband portion are supplied and integrated by a single vendor. Thus, it will be understood that network nodes include disaggregated implementations or portions thereof. For example, in some embodiments, the telecommunication network 1202 includes one or more Open-RAN (ORAN) network nodes. An ORAN network node is a node in the telecommunication network 1202 that supports an ORAN specification (e.g., a specification published by the O-RAN Alliance, or any similar organization) and may operate alone or together with other nodes to implement one or more functionalities of any node in the telecommunication network 1202, including one or more network nodes 1210 and / or core network nodes 1208.

[0084] The embodiments described herein for characterizing a SerDes receiver and predicting BER performance may be implemented with respect to any of the SerDes interfaces within the access network 1204 and core network 1205 (including the interconnections between the access network 1204 and core network 1206 and the interconnections between the core network 1206 and the host 1216). These embodiments may be used, for example, to characterize wired high speed communication link channels in the baseband units of a telecommunications network (e.g., the distributed units (DUs) in some implementations). Additionally, the described embodiments may be applied to the wired high speed channels coupled to radio units (RUs) and / or centralized units (CUs) of the telecommunications network, or to any other location in the telecommunications network and / or any type of communication network. Although not specifically described here, the embodiments of this disclosure may also be applied to characterize the performance of wireless channels such as those established between UEs and access networks, as described below with respect to Figure 12. For example, these embodiments may be used to predict the bit error rate of a given wireless channel.Atty. Docket No.: 4906P109167WO01

[0085] Examples of an ORAN network node include an open radio unit (O-RU), an open distributed unit (O-DU), an open central unit (O-CU), including an O-CU control plane (O-CU- CP) or an O-CU user plane (O-CU-UP), a RAN intelligent controller (near-real time or non-real time) hosting software or software plug-ins, such as a near-real time control application (e.g., xApp) or a non-real time control application (e.g., rApp), or any combination thereof (the adjective “open” designating support of an ORAN specification). The network node may support a specification by, for example, supporting an interface defined by the ORAN specification, such as an A1, F1, W1, E1, E2, X2, Xn interface, an open fronthaul user plane interface, or an open fronthaul management plane interface. Moreover, an ORAN access node may be a logical node in a physical node. Furthermore, an ORAN network node may be implemented in a virtualization environment (described further below) in which one or more network functions are virtualized. For example, the virtualization environment may include an O-Cloud computing platform orchestrated by a Service Management and Orchestration Framework via an O-2 interface defined by the O-RAN Alliance or comparable technologies. The network nodes 1210 facilitate direct or indirect connection of user equipment (UE), such as by connecting UEs 1212a, 1212b, 1212c, and 1212d (one or more of which may be generally referred to as UEs 1212) to the core network 1206 over one or more wireless connections.

[0086] Example wireless communications over a wireless connection include transmitting and / or receiving wireless signals using electromagnetic waves, radio waves, infrared waves, and / or other types of signals suitable for conveying information without the use of wires, cables, or other material conductors. Moreover, in different embodiments, the communication system 1200 may include any number of wired or wireless networks, network nodes, UEs, and / or any other components or systems that may facilitate or participate in the communication of data and / or signals whether via wired or wireless connections. The communication system 1200 may include and / or interface with any type of communication, telecommunication, data, cellular, radio network, and / or other similar type of system.

[0087] The UEs 1212 may be any of a wide variety of communication devices, including wireless devices arranged, configured, and / or operable to communicate wirelessly with the network nodes 1210 and other communication devices. Similarly, the network nodes 1210 are arranged, capable, configured, and / or operable to communicate directly or indirectly with the UEs 1212 and / or with other network nodes or equipment in the telecommunication network 1202 to enable and / or provide network access, such as wireless network access, and / or to perform other functions, such as administration in the telecommunication network 1202.

[0088] In the depicted example, the core network 1206 connects the network nodes 1210 to one or more hosts, such as host 1216. These connections may be direct or indirect via one orAtty. Docket No.: 4906P109167WO01 more intermediary networks or devices. In other examples, network nodes may be directly coupled to hosts. The core network 1206 includes one more core network nodes (e.g., core network node 1208) that are structured with hardware and software components. Features of these components may be substantially similar to those described with respect to the UEs, network nodes, and / or hosts, such that the descriptions thereof are generally applicable to the corresponding components of the core network node 1208. Example core network nodes include functions of one or more of a Mobile Switching Center (MSC), Mobility Management Entity (MME), Home Subscriber Server (HSS), Access and Mobility Management Function (AMF), Session Management Function (SMF), Authentication Server Function (AUSF), Subscription Identifier De-concealing function (SIDF), Unified Data Management (UDM), Security Edge Protection Proxy (SEPP), Network Exposure Function (NEF), and / or a User Plane Function (UPF).

[0089] The host 1216 may be under the ownership or control of a service provider other than an operator or provider of the access network 1204 and / or the telecommunication network 1202, and may be operated by the service provider or on behalf of the service provider. The host 1216 may host a variety of applications to provide one or more service. Examples of such applications include live and pre-recorded audio / video content, data collection services such as retrieving and compiling data on various ambient conditions detected by a plurality of UEs, analytics functionality, social media, functions for controlling or otherwise interacting with remote devices, functions for an alarm and surveillance center, or any other such function performed by a server.

[0090] As a whole, the communication system 1200 of Figure 12 enables connectivity between the UEs, network nodes, and hosts. In that sense, the communication system may be configured to operate according to predefined rules or procedures, such as specific standards that include, but are not limited to: Global System for Mobile Communications (GSM); Universal Mobile Telecommunications System (UMTS); Long Term Evolution (LTE), and / or other suitable 2G, 3G, 4G, 5G standards, or any applicable future generation standard (e.g., 6G); wireless local area network (WLAN) standards, such as the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standards (WiFi); and / or any other appropriate wireless communication standard, such as the Worldwide Interoperability for Microwave Access (WiMax), Bluetooth, Z-Wave, Near Field Communication (NFC) ZigBee, LiFi, and / or any low- power wide-area network (LPWAN) standards such as LoRa and Sigfox.

[0091] In some examples, the telecommunication network 1202 is a cellular network that implements 3GPP standardized features. Accordingly, the telecommunications network 1202 may support network slicing to provide different logical networks to different devices that areAtty. Docket No.: 4906P109167WO01 connected to the telecommunication network 1202. For example, the telecommunications network 1202 may provide Ultra Reliable Low Latency Communication (URLLC) services to some UEs, while providing Enhanced Mobile Broadband (eMBB) services to other UEs, and / or Massive Machine Type Communication (mMTC) / Massive IoT services to yet further UEs.

[0092] In some examples, the UEs 1212 are configured to transmit and / or receive information without direct human interaction. For instance, a UE may be designed to transmit information to the access network 1204 on a predetermined schedule, when triggered by an internal or external event, or in response to requests from the access network 1204. Additionally, a UE may be configured for operating in single- or multi-RAT or multi-standard mode. For example, a UE may operate with any one or combination of Wi-Fi, NR (New Radio) and LTE, i.e. being configured for multi-radio dual connectivity (MR-DC), such as E-UTRAN (Evolved-UMTS Terrestrial Radio Access Network) New Radio – Dual Connectivity (EN-DC).

[0093] In the example, the hub 1214 communicates with the access network 1204 to facilitate indirect communication between one or more UEs (e.g., UE 1212c and / or 1212d) and network nodes (e.g., network node 1210b). In some examples, the hub 1214 may be a controller, router, content source and analytics, or any of the other communication devices described herein regarding UEs. For example, the hub 1214 may be a broadband router enabling access to the core network 1206 for the UEs. As another example, the hub 1214 may be a controller that sends commands or instructions to one or more actuators in the UEs. Commands or instructions may be received from the UEs, network nodes 1210, or by executable code, script, process, or other instructions in the hub 1214. As another example, the hub 1214 may be a data collector that acts as temporary storage for UE data and, in some embodiments, may perform analysis or other processing of the data. As another example, the hub 1214 may be a content source. For example, for a UE that is a VR headset, display, loudspeaker or other media delivery device, the hub 1214 may retrieve VR assets, video, audio, or other media or data related to sensory information via a network node, which the hub 1214 then provides to the UE either directly, after performing local processing, and / or after adding additional local content. In still another example, the hub 1214 acts as a proxy server or orchestrator for the UEs, in particular if one or more of the UEs are low energy IoT devices.

[0094] The hub 1214 may have a constant / persistent or intermittent connection to the network node 1210b. The hub 1214 may also allow for a different communication scheme and / or schedule between the hub 1214 and UEs (e.g., UE 1212c and / or 1212d), and between the hub 1214 and the core network 1206. In other examples, the hub 1214 is connected to the core network 1206 and / or one or more UEs via a wired connection. Moreover, the hub 1214 may be configured to connect to an M2M service provider over the access network 1204 and / or toAtty. Docket No.: 4906P109167WO01 another UE over a direct connection. In some scenarios, UEs may establish a wireless connection with the network nodes 1210 while still connected via the hub 1214 via a wired or wireless connection. In some embodiments, the hub 1214 may be a dedicated hub – that is, a hub whose primary function is to route communications to / from the UEs from / to the network node 1210b. In other embodiments, the hub 1214 may be a non-dedicated hub – that is, a device which is capable of operating to route communications between the UEs and network node 1210b, but which is additionally capable of operating as a communication start and / or end point for certain data channels.

[0095] Figure 13 shows a UE 1300 in accordance with some embodiments. As used herein, a UE refers to a device capable, configured, arranged and / or operable to communicate wirelessly with network nodes and / or other UEs. Examples of a UE include, but are not limited to, a smart phone, mobile phone, cell phone, voice over IP (VoIP) phone, wireless local loop phone, desktop computer, personal digital assistant (PDA), wireless cameras, gaming console or device, music storage device, playback appliance, wearable terminal device, wireless endpoint, mobile station, tablet, laptop, laptop-embedded equipment (LEE), laptop-mounted equipment (LME), smart device, wireless customer-premise equipment (CPE), vehicle, vehicle-mounted or vehicle embedded / integrated wireless device, etc. Other examples include any UE identified by the 3rd Generation Partnership Project (3GPP), including a narrow band internet of things (NB-IoT) UE, a machine type communication (MTC) UE, and / or an enhanced MTC (eMTC) UE.

[0096] A UE may support device-to-device (D2D) communication, for example by implementing a 3GPP standard for sidelink communication, Dedicated Short-Range Communication (DSRC), vehicle-to-vehicle (V2V), vehicle-to-infrastructure (V2I), or vehicle- to-everything (V2X). In other examples, a UE may not necessarily have a user in the sense of a human user who owns and / or operates the relevant device. Instead, a UE may represent a device that is intended for sale to, or operation by, a human user but which may not, or which may not initially, be associated with a specific human user (e.g., a smart sprinkler controller). Alternatively, a UE may represent a device that is not intended for sale to, or operation by, an end user but which may be associated with or operated for the benefit of a user (e.g., a smart power meter).

[0097] The UE 1300 includes processing circuitry 1302 that is operatively coupled via a bus 1304 to an input / output interface 1306, a power source 1308, a memory 1310, a communication interface 1312, and / or any other component, or any combination thereof. Certain UEs may utilize all or a subset of the components shown in Figure 13. The level of integration between the components may vary from one UE to another UE. Further, certain UEs may containAtty. Docket No.: 4906P109167WO01 multiple instances of a component, such as multiple processors, memories, transceivers, transmitters, receivers, etc.

[0098] The processing circuitry 1302 is configured to process instructions and data and may be configured to implement any sequential state machine operative to execute instructions stored as machine-readable computer programs in the memory 1310. The processing circuitry 1302 may be implemented as one or more hardware-implemented state machines (e.g., in discrete logic, field-programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), etc.); programmable logic together with appropriate firmware; one or more stored computer programs, general-purpose processors, such as a microprocessor or digital signal processor (DSP), together with appropriate software; or any combination of the above. For example, the processing circuitry 1302 may include multiple central processing units (CPUs).

[0099] In the example, the input / output interface 1306 may be configured to provide an interface or interfaces to an input device, output device, or one or more input and / or output devices. Examples of an output device include a speaker, a sound card, a video card, a display, a monitor, a printer, an actuator, an emitter, a smartcard, another output device, or any combination thereof. An input device may allow a user to capture information into the UE 1300. Examples of an input device include a touch-sensitive or presence-sensitive display, a camera (e.g., a digital camera, a digital video camera, a web camera, etc.), a microphone, a sensor, a mouse, a trackball, a directional pad, a trackpad, a scroll wheel, a smartcard, and the like. The presence-sensitive display may include a capacitive or resistive touch sensor to sense input from a user. A sensor may be, for instance, an accelerometer, a gyroscope, a tilt sensor, a force sensor, a magnetometer, an optical sensor, a proximity sensor, a biometric sensor, etc., or any combination thereof. An output device may use the same type of interface port as an input device. For example, a Universal Serial Bus (USB) port may be used to provide an input device and an output device.

[0100] In some embodiments, the power source 1308 is structured as a battery or battery pack. Other types of power sources, such as an external power source (e.g., an electricity outlet), photovoltaic device, or power cell, may be used. The power source 1308 may further include power circuitry for delivering power from the power source 1308 itself, and / or an external power source, to the various parts of the UE 1300 via input circuitry or an interface such as an electrical power cable. Delivering power may be, for example, for charging of the power source 1308. Power circuitry may perform any formatting, converting, or other modification to the power from the power source 1308 to make the power suitable for the respective components of the UE 1300 to which power is supplied.Atty. Docket No.: 4906P109167WO01

[0101] The memory 1310 may be or be configured to include memory such as random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read- only memory (EEPROM), magnetic disks, optical disks, hard disks, removable cartridges, flash drives, and so forth. In one example, the memory 1310 includes one or more application programs 1314, such as an operating system, web browser application, a widget, gadget engine, or other application, and corresponding data 1316. The memory 1310 may store, for use by the UE 1300, any of a variety of various operating systems or combinations of operating systems.

[0102] The memory 1310 may be configured to include a number of physical drive units, such as redundant array of independent disks (RAID), flash memory, USB flash drive, external hard disk drive, thumb drive, pen drive, key drive, high-density digital versatile disc (HD-DVD) optical disc drive, internal hard disk drive, Blu-Ray optical disc drive, holographic digital data storage (HDDS) optical disc drive, external mini-dual in-line memory module (DIMM), synchronous dynamic random access memory (SDRAM), external micro-DIMM SDRAM, smartcard memory such as tamper resistant module in the form of a universal integrated circuit card (UICC) including one or more subscriber identity modules (SIMs), such as a USIM and / or ISIM, other memory, or any combination thereof. The UICC may for example be an embedded UICC (eUICC), integrated UICC (iUICC) or a removable UICC commonly known as ‘SIM card.’ The memory 1310 may allow the UE 1300 to access instructions, application programs and the like, stored on transitory or non-transitory memory media, to off-load data, or to upload data. An article of manufacture, such as one utilizing a communication system may be tangibly embodied as or in the memory 1310, which may be or comprise a device-readable storage medium.

[0103] The processing circuitry 1302 may be configured to communicate with an access network or other network using the communication interface 1312. The communication interface 1312 may comprise one or more communication subsystems and may include or be communicatively coupled to an antenna 1322. The communication interface 1312 may include one or more transceivers used to communicate, such as by communicating with one or more remote transceivers of another device capable of wireless communication (e.g., another UE or a network node in an access network). Each transceiver may include a transmitter 1318 and / or a receiver 1320 appropriate to provide network communications (e.g., optical, electrical, frequency allocations, and so forth). Moreover, the transmitter 1318 and receiver 1320 may be coupled to one or more antennas (e.g., antenna 1322) and may share circuit components, software or firmware, or alternatively be implemented separately.Atty. Docket No.: 4906P109167WO01

[0104] In the illustrated embodiment, communication functions of the communication interface 1312 may include cellular communication, Wi-Fi communication, LPWAN communication, data communication, voice communication, multimedia communication, short- range communications such as Bluetooth, near-field communication, location-based communication such as the use of the global positioning system (GPS) to determine a location, another like communication function, or any combination thereof. Communications may be implemented in according to one or more communication protocols and / or standards, such as IEEE 802.11, Code Division Multiplexing Access (CDMA), Wideband Code Division Multiple Access (WCDMA), GSM, LTE, New Radio (NR), UMTS, WiMax, Ethernet, transmission control protocol / internet protocol (TCP / IP), synchronous optical networking (SONET), Asynchronous Transfer Mode (ATM), QUIC, Hypertext Transfer Protocol (HTTP), and so forth.

[0105] Regardless of the type of sensor, a UE may provide an output of data captured by its sensors, through its communication interface 1312, via a wireless connection to a network node. Data captured by sensors of a UE can be communicated through a wireless connection to a network node via another UE. The output may be periodic (e.g., once every 15 minutes if it reports the sensed temperature), random (e.g., to even out the load from reporting from several sensors), in response to a triggering event (e.g., when moisture is detected an alert is sent), in response to a request (e.g., a user initiated request), or a continuous stream (e.g., a live video feed of a patient).

[0106] As another example, a UE comprises an actuator, a motor, or a switch, related to a communication interface configured to receive wireless input from a network node via a wireless connection. In response to the received wireless input the states of the actuator, the motor, or the switch may change. For example, the UE may comprise a motor that adjusts the control surfaces or rotors of a drone in flight according to the received input or to a robotic arm performing a medical procedure according to the received input.

[0107] A UE, when in the form of an Internet of Things (IoT) device, may be a device for use in one or more application domains, these domains comprising, but not limited to, city wearable technology, extended industrial application and healthcare. Non-limiting examples of such an IoT device are a device which is or which is embedded in: a connected refrigerator or freezer, a TV, a connected lighting device, an electricity meter, a robot vacuum cleaner, a voice controlled smart speaker, a home security camera, a motion detector, a thermostat, a smoke detector, a door / window sensor, a flood / moisture sensor, an electrical door lock, a connected doorbell, an air conditioning system like a heat pump, an autonomous vehicle, a surveillance system, a weather monitoring device, a vehicle parking monitoring device, an electric vehicle charging station, a smart watch, a fitness tracker, a head-mounted display for Augmented Reality (AR) orAtty. Docket No.: 4906P109167WO01 Virtual Reality (VR), a wearable for tactile augmentation or sensory enhancement, a water sprinkler, an animal- or item-tracking device, a sensor for monitoring a plant or animal, an industrial robot, an Unmanned Aerial Vehicle (UAV), and any kind of medical device, like a heart rate monitor or a remote controlled surgical robot. A UE in the form of an IoT device comprises circuitry and / or software in dependence of the intended application of the IoT device in addition to other components as described in relation to the UE 1300 shown in Figure 13.

[0108] As yet another specific example, in an IoT scenario, a UE may represent a machine or other device that performs monitoring and / or measurements, and transmits the results of such monitoring and / or measurements to another UE and / or a network node. The UE may in this case be an M2M device, which may in a 3GPP context be referred to as an MTC device. As one particular example, the UE may implement the 3GPP NB-IoT standard. In other scenarios, a UE may represent a vehicle, such as a car, a bus, a truck, a ship and an airplane, or other equipment that is capable of monitoring and / or reporting on its operational status or other functions associated with its operation.

[0109] In practice, any number of UEs may be used together with respect to a single use case. For example, a first UE might be or be integrated in a drone and provide the drone’s speed information (obtained through a speed sensor) to a second UE that is a remote controller operating the drone. When the user makes changes from the remote controller, the first UE may adjust the throttle on the drone (e.g. by controlling an actuator) to increase or decrease the drone’s speed. The first and / or the second UE can also include more than one of the functionalities described above. For example, a UE might comprise the sensor and the actuator, and handle communication of data for both the speed sensor and the actuators.

[0110] Figure 14 shows a network node 1400 in accordance with some embodiments. As used herein, network node refers to equipment capable, configured, arranged and / or operable to communicate directly or indirectly with a UE and / or with other network nodes or equipment, in a telecommunication network. Examples of network nodes include, but are not limited to, access points (APs) (e.g., radio access points), base stations (BSs) (e.g., radio base stations, Node Bs, evolved Node Bs (eNBs) and NR NodeBs (gNBs)), O-RAN nodes or components of an O-RAN node (e.g., O-RU, O-DU, O-CU).

[0111] Base stations may be categorized based on the amount of coverage they provide (or, stated differently, their transmit power level) and so, depending on the provided amount of coverage, may be referred to as femto base stations, pico base stations, micro base stations, or macro base stations. A base station may be a relay node or a relay donor node controlling a relay. A network node may also include one or more (or all) parts of a distributed radio base station such as centralized digital units, distributed units (e.g., in an O-RAN access node) and / or remoteAtty. Docket No.: 4906P109167WO01 radio units (RRUs), sometimes referred to as Remote Radio Heads (RRHs). Such remote radio units may or may not be integrated with an antenna as an antenna integrated radio. Parts of a distributed radio base station may also be referred to as nodes in a distributed antenna system (DAS).

[0112] Other examples of network nodes include multiple transmission point (multi-TRP) 5G access nodes, multi-standard radio (MSR) equipment such as MSR BSs, network controllers such as radio network controllers (RNCs) or base station controllers (BSCs), base transceiver stations (BTSs), transmission points, transmission nodes, multi-cell / multicast coordination entities (MCEs), Operation and Maintenance (O&M) nodes, Operations Support System (OSS) nodes, Self-Organizing Network (SON) nodes, positioning nodes (e.g., Evolved Serving Mobile Location Centers (E-SMLCs)), and / or Minimization of Drive Tests (MDTs).

[0113] The network node 1400 includes a processing circuitry 1402, a memory 1404, a communication interface 1406, and a power source 1408. The network node 1400 may be composed of multiple physically separate components (e.g., a NodeB component and a RNC component, or a BTS component and a BSC component, etc.), which may each have their own respective components. In certain scenarios in which the network node 1400 comprises multiple separate components (e.g., BTS and BSC components), one or more of the separate components may be shared among several network nodes. For example, a single RNC may control multiple NodeBs. In such a scenario, each unique NodeB and RNC pair, may in some instances be considered a single separate network node. In some embodiments, the network node 1400 may be configured to support multiple radio access technologies (RATs). In such embodiments, some components may be duplicated (e.g., separate memory 1404 for different RATs) and some components may be reused (e.g., a same antenna 1410 may be shared by different RATs). The network node 1400 may also include multiple sets of the various illustrated components for different wireless technologies integrated into network node 1400, for example GSM, WCDMA, LTE, NR, WiFi, Zigbee, Z-wave, LoRaWAN, Radio Frequency Identification (RFID) or Bluetooth wireless technologies. These wireless technologies may be integrated into the same or different chip or set of chips and other components within network node 1400.

[0114] The processing circuitry 1402 may comprise a combination of one or more of a microprocessor, controller, microcontroller, central processing unit, digital signal processor, application-specific integrated circuit, field programmable gate array, or any other suitable computing device, resource, or combination of hardware, software and / or encoded logic operable to provide, either alone or in conjunction with other network node 1400 components, such as the memory 1404, to provide network node 1400 functionality.Atty. Docket No.: 4906P109167WO01

[0115] In some embodiments, the processing circuitry 1402 includes a system on a chip (SOC). In some embodiments, the processing circuitry 1402 includes one or more of radio frequency (RF) transceiver circuitry 1412 and baseband processing circuitry 1414. In some embodiments, the radio frequency (RF) transceiver circuitry 1412 and the baseband processing circuitry 1414 may be on separate chips (or sets of chips), boards, or units, such as radio units and digital units. In alternative embodiments, part or all of RF transceiver circuitry 1412 and baseband processing circuitry 1414 may be on the same chip or set of chips, boards, or units.

[0116] The memory 1404 may comprise any form of volatile or non-volatile computer- readable memory including, without limitation, persistent storage, solid-state memory, remotely mounted memory, magnetic media, optical media, random access memory (RAM), read-only memory (ROM), mass storage media (for example, a hard disk), removable storage media (for example, a flash drive, a Compact Disk (CD) or a Digital Video Disk (DVD)), and / or any other volatile or non-volatile, non-transitory device-readable and / or computer-executable memory devices that store information, data, and / or instructions that may be used by the processing circuitry 1402. The memory 1404 may store any suitable instructions, data, or information, including a computer program, software, an application including one or more of logic, rules, code, tables, and / or other instructions capable of being executed by the processing circuitry 1402 and utilized by the network node 1400. The memory 1404 may be used to store any calculations made by the processing circuitry 1402 and / or any data received via the communication interface 1406. In some embodiments, the processing circuitry 1402 and memory 1404 is integrated.

[0117] The communication interface 1406 is used in wired or wireless communication of signaling and / or data between a network node, access network, and / or UE. As illustrated, the communication interface 1406 comprises port(s) / terminal(s) 1416 to send and receive data, for example to and from a network over a wired connection. The communication interface 1406 also includes radio front-end circuitry 1418 that may be coupled to, or in certain embodiments a part of, the antenna 1410. Radio front-end circuitry 1418 comprises filters 1420 and amplifiers 1422. The radio front-end circuitry 1418 may be connected to an antenna 1410 and processing circuitry 1402. The radio front-end circuitry may be configured to condition signals communicated between antenna 1410 and processing circuitry 1402. The radio front-end circuitry 1418 may receive digital data that is to be sent out to other network nodes or UEs via a wireless connection. The radio front-end circuitry 1418 may convert the digital data into a radio signal having the appropriate channel and bandwidth parameters using a combination of filters 1420 and / or amplifiers 1422. The radio signal may then be transmitted via the antenna 1410. Similarly, when receiving data, the antenna 1410 may collect radio signals which are then converted into digital data by the radio front-end circuitry 1418. The digital data may be passedAtty. Docket No.: 4906P109167WO01 to the processing circuitry 1402. In other embodiments, the communication interface may comprise different components and / or different combinations of components.

[0118] In certain alternative embodiments, the network node 1400 does not include separate radio front-end circuitry 1418, instead, the processing circuitry 1402 includes radio front-end circuitry and is connected to the antenna 1410. Similarly, in some embodiments, all or some of the RF transceiver circuitry 1412 is part of the communication interface 1406. In still other embodiments, the communication interface 1406 includes one or more ports or terminals 1416, the radio front-end circuitry 1418, and the RF transceiver circuitry 1412, as part of a radio unit (not shown), and the communication interface 1406 communicates with the baseband processing circuitry 1414, which is part of a digital unit (not shown).

[0119] The antenna 1410 may include one or more antennas, or antenna arrays, configured to send and / or receive wireless signals. The antenna 1410 may be coupled to the radio front-end circuitry 1418 and may be any type of antenna capable of transmitting and receiving data and / or signals wirelessly. In certain embodiments, the antenna 1410 is separate from the network node 1400 and connectable to the network node 1400 through an interface or port.

[0120] The antenna 1410, communication interface 1406, and / or the processing circuitry 1402 may be configured to perform any receiving operations and / or certain obtaining operations described herein as being performed by the network node. Any information, data and / or signals may be received from a UE, another network node and / or any other network equipment. Similarly, the antenna 1410, the communication interface 1406, and / or the processing circuitry 1402 may be configured to perform any transmitting operations described herein as being performed by the network node. Any information, data and / or signals may be transmitted to a UE, another network node and / or any other network equipment.

[0121] The power source 1408 provides power to the various components of network node 1400 in a form suitable for the respective components (e.g., at a voltage and current level needed for each respective component). The power source 1408 may further comprise, or be coupled to, power management circuitry to supply the components of the network node 1400 with power for performing the functionality described herein. For example, the network node 1400 may be connectable to an external power source (e.g., the power grid, an electricity outlet) via an input circuitry or interface such as an electrical cable, whereby the external power source supplies power to power circuitry of the power source 1408. As a further example, the power source 1408 may comprise a source of power in the form of a battery or battery pack which is connected to, or integrated in, power circuitry. The battery may provide backup power should the external power source fail.Atty. Docket No.: 4906P109167WO01

[0122] Embodiments of the network node 1400 may include additional components beyond those shown in Figure 14 for providing certain aspects of the network node’s functionality, including any of the functionality described herein and / or any functionality necessary to support the subject matter described herein. For example, the network node 1400 may include user interface equipment to allow input of information into the network node 1400 and to allow output of information from the network node 1400. This may allow a user to perform diagnostic, maintenance, repair, and other administrative functions for the network node 1400.

[0123] Figure 15 is a block diagram of a host 1500, which may be an embodiment of the host 1216 of Figure 12, in accordance with various aspects described herein. As used herein, the host 1500 may be or comprise various combinations hardware and / or software, including a standalone server, a blade server, a cloud-implemented server, a distributed server, a virtual machine, container, or processing resources in a server farm. The host 1500 may provide one or more services to one or more UEs.

[0124] The host 1500 includes processing circuitry 1502 that is operatively coupled via a bus 1504 to an input / output interface 1506, a network interface 1508, a power source 1510, and a memory 1512. Other components may be included in other embodiments. Features of these components may be substantially similar to those described with respect to the devices of previous figures, such as Figures 12 and 13, such that the descriptions thereof are generally applicable to the corresponding components of host 1500.

[0125] The memory 1512 may include one or more computer programs including one or more host application programs 1514 and data 1516, which may include user data, e.g., data generated by a UE for the host 1500 or data generated by the host 1500 for a UE. Embodiments of the host 1500 may utilize only a subset or all of the components shown. The host application programs 1514 may be implemented in a container-based architecture and may provide support for video codecs (e.g., Versatile Video Coding (VVC), High Efficiency Video Coding (HEVC), Advanced Video Coding (AVC), MPEG, VP9) and audio codecs (e.g., FLAC, Advanced Audio Coding (AAC), MPEG, G.711), including transcoding for multiple different classes, types, or implementations of UEs (e.g., handsets, desktop computers, wearable display systems, heads-up display systems). The host application programs 1514 may also provide for user authentication and licensing checks and may periodically report health, routes, and content availability to a central node, such as a device in or on the edge of a core network. Accordingly, the host 1500 may select and / or indicate a different host for over-the-top services for a UE. The host application programs 1514 may support various protocols, such as the HTTP Live Streaming (HLS) protocol, Real-Time Messaging Protocol (RTMP), Real-Time Streaming Protocol (RTSP), Dynamic Adaptive Streaming over HTTP (MPEG-DASH), etc.Atty. Docket No.: 4906P109167WO01

[0126] Virtualizing means creating virtual versions of apparatuses or devices which may include virtualizing hardware platforms, storage devices and networking resources. As used herein, virtualization can be applied to any device described herein, or components thereof, and relates to an implementation in which at least a portion of the functionality is implemented as one or more virtual components. Some or all of the functions described herein may be implemented as virtual components executed by one or more virtual machines (VMs) implemented in one or more virtual environments hosted by one or more of hardware nodes, such as a hardware computing device that operates as a network node, UE, core network node, or host. Further, in embodiments in which the virtual node does not require radio connectivity (e.g., a core network node or host), then the node may be entirely virtualized. In some embodiments, the virtualization environment includes components defined by the O-RAN Alliance, such as an O-Cloud environment orchestrated by a Service Management and Orchestration Framework via an O-2 interface.

[0127] Applications (which may alternatively be called software instances, virtual appliances, network functions, virtual nodes, virtual network functions, etc.) are run in the virtualization environment to implement some of the features, functions, and / or benefits of some of the embodiments disclosed herein.

[0128] Software may be executed by the processing circuitry to instantiate one or more virtualization layers (also referred to as hypervisors or virtual machine monitors (VMMs)), provide VMs, and / or perform any of the functions, features and / or benefits described in relation with some embodiments described herein. The virtualization layer may present a virtual operating platform that appears like networking hardware to the VMs.

[0129] The VMs comprise virtual processing, virtual memory, virtual networking or interface and virtual storage, and may be run by a corresponding virtualization layer. Different embodiments of the instance of a virtual appliance may be implemented on one or more of VMs, and the implementations may be made in different ways. Virtualization of the hardware is in some contexts referred to as network function virtualization (NFV). NFV may be used to consolidate many network equipment types onto industry standard high volume server hardware, physical switches, and physical storage, which can be located in data centers, and customer premise equipment.

[0130] In the context of NFV, a VM may be a software implementation of a physical machine that runs programs as if they were executing on a physical, non-virtualized machine. Each of the VMs, and that part of hardware that executes that VM, be it hardware dedicated to that VM and / or hardware shared by that VM with others of the VMs, forms separate virtual network elements. Still in the context of NFV, a virtual network function is responsible for handlingAtty. Docket No.: 4906P109167WO01 specific network functions that run in one or more VMs on top of the hardware and corresponds to an application.

[0131] Figure 16 shows a communication diagram of a host 1602 communicating via a network node 1604 with a UE 1606 over a partially wireless connection in accordance with some embodiments. Example implementations, in accordance with various embodiments, of the UE (such as a UE 1212a of Figure 12 and / or UE 1300 of Figure 13), network node (such as network node 1210a of Figure 12 and / or network node 1400 of Figure 14), and host (such as host 1216 of Figure 12 and / or host 1500 of Figure 15) discussed in the preceding paragraphs will now be described with reference to Figure 16.

[0132] Like host 1500, embodiments of host 1602 include hardware, such as a communication interface, processing circuitry, and memory. The host 1602 also includes software, which is stored in or accessible by the host 1602 and executable by the processing circuitry. The software includes a host application that may be operable to provide a service to a remote user, such as the UE 1606 connecting via an over-the-top (OTT) connection 1650 extending between the UE 1606 and host 1602. In providing the service to the remote user, a host application may provide user data which is transmitted using the OTT connection 1650.

[0133] The network node 1604 includes hardware enabling it to communicate with the host 1602 and UE 1606. The connection 1660 may be direct or pass through a core network (like core network 1206 of Figure 12) and / or one or more other intermediate networks, such as one or more public, private, or hosted networks. For example, an intermediate network may be a backbone network or the Internet.

[0134] The UE 1606 includes hardware and software, which is stored in or accessible by UE 1606 and executable by the UE’s processing circuitry. The software includes a client application, such as a web browser or operator-specific “app” that may be operable to provide a service to a human or non-human user via UE 1606 with the support of the host 1602. In the host 1602, an executing host application may communicate with the executing client application via the OTT connection 1650 terminating at the UE 1606 and host 1602. In providing the service to the user, the UE's client application may receive request data from the host's host application and provide user data in response to the request data. The OTT connection 1650 may transfer both the request data and the user data. The UE's client application may interact with the user to generate the user data that it provides to the host application through the OTT connection 1650.

[0135] The OTT connection 1650 may extend via a connection 1660 between the host 1602 and the network node 1604 and via a wireless connection 1670 between the network node 1604 and the UE 1606 to provide the connection between the host 1602 and the UE 1606. The connection 1660 and wireless connection 1670, over which the OTT connection 1650 may beAtty. Docket No.: 4906P109167WO01 provided, have been drawn abstractly to illustrate the communication between the host 1602 and the UE 1606 via the network node 1604, without explicit reference to any intermediary devices and the precise routing of messages via these devices.

[0136] As an example of transmitting data via the OTT connection 1650, in step 1608, the host 1602 provides user data, which may be performed by executing a host application. In some embodiments, the user data is associated with a particular human user interacting with the UE 1606. In other embodiments, the user data is associated with a UE 1606 that shares data with the host 1602 without explicit human interaction. In step 1610, the host 1602 initiates a transmission carrying the user data towards the UE 1606. The host 1602 may initiate the transmission responsive to a request transmitted by the UE 1606. The request may be caused by human interaction with the UE 1606 or by operation of the client application executing on the UE 1606. The transmission may pass via the network node 1604, in accordance with the teachings of the embodiments described throughout this disclosure. Accordingly, in step 1612, the network node 1604 transmits to the UE 1606 the user data that was carried in the transmission that the host 1602 initiated, in accordance with the teachings of the embodiments described throughout this disclosure. In step 1614, the UE 1606 receives the user data carried in the transmission, which may be performed by a client application executed on the UE 1606 associated with the host application executed by the host 1602.

[0137] In some examples, the UE 1606 executes a client application which provides user data to the host 1602. The user data may be provided in reaction or response to the data received from the host 1602. Accordingly, in step 1616, the UE 1606 may provide user data, which may be performed by executing the client application. In providing the user data, the client application may further consider user input received from the user via an input / output interface of the UE 1606. Regardless of the specific manner in which the user data was provided, the UE 1606 initiates, in step 1618, transmission of the user data towards the host 1602 via the network node 1604. In step 1620, in accordance with the teachings of the embodiments described throughout this disclosure, the network node 1604 receives user data from the UE 1606 and initiates transmission of the received user data towards the host 1602. In step 1622, the host 1602 receives the user data carried in the transmission initiated by the UE 1606.

[0138] One or more of the various embodiments improve the performance of OTT services provided to the UE 1606 using the OTT connection 1650, in which the wireless connection 1670 forms the last segment.

[0139] In an example scenario, factory status information may be collected and analyzed by the host 1602. As another example, the host 1602 may process audio and video data which may have been retrieved from a UE for use in creating maps. As another example, the host 1602 mayAtty. Docket No.: 4906P109167WO01 collect and analyze real-time data to assist in controlling vehicle congestion (e.g., controlling traffic lights). As another example, the host 1602 may store surveillance video uploaded by a UE. As another example, the host 1602 may store or control access to media content such as video, audio, VR or AR which it can broadcast, multicast or unicast to UEs. As other examples, the host 1602 may be used for energy pricing, remote control of non-time critical electrical load to balance power generation needs, location services, presentation services (such as compiling diagrams etc. from data collected from remote devices), or any other function of collecting, retrieving, storing, analyzing and / or transmitting data.

[0140] In some examples, a measurement procedure may be provided for the purpose of monitoring data rate, latency and other factors on which the one or more embodiments improve. There may further be an optional network functionality for reconfiguring the OTT connection 1650 between the host 1602 and UE 1606, in response to variations in the measurement results. The measurement procedure and / or the network functionality for reconfiguring the OTT connection may be implemented in software and hardware of the host 1602 and / or UE 1606. In some embodiments, sensors (not shown) may be deployed in or in association with other devices through which the OTT connection 1650 passes; the sensors may participate in the measurement procedure by supplying values of the monitored quantities exemplified above, or supplying values of other physical quantities from which software may compute or estimate the monitored quantities. The reconfiguring of the OTT connection 1650 may include message format, retransmission settings, preferred routing etc.; the reconfiguring need not directly alter the operation of the network node 1604. Such procedures and functionalities may be known and practiced in the art. In certain embodiments, measurements may involve proprietary UE signaling that facilitates measurements of throughput, propagation times, latency and the like, by the host 1602. The measurements may be implemented in that software causes messages to be transmitted, in particular empty or ‘dummy’ messages, using the OTT connection 1650 while monitoring propagation times, errors, etc.

[0141] A computer implemented method is described herein for training a machine learning- based simulation of a receiver of a Serializer / Deserializer (SerDes) system. One embodiment of the method comprises: based on a channel impulse response, selecting a receiver equalization model from a plurality of receiver equalization models using an optimization process; based on the selected receiver equalization model, converting the channel impulse response to an equalized channel impulse response; training a supervised machine learning model to predict an error bit rate using the equalized channel impulse response and selected receiver equalization model; and based on an error of the prediction, selecting a second receiver equalization model from the plurality of receiver equalization models using the optimization process.Atty. Docket No.: 4906P109167WO01

[0142] Although the computing devices described herein (e.g., UEs, network nodes, hosts) may include the illustrated combination of hardware components, other embodiments may comprise computing devices with different combinations of components. It is to be understood that these computing devices may comprise any suitable combination of hardware and / or software needed to perform the tasks, features, functions and methods disclosed herein. Determining, calculating, obtaining or similar operations described herein may be performed by processing circuitry, which may process information by, for example, converting the obtained information into other information, comparing the obtained information or converted information to information stored in the network node, and / or performing one or more operations based on the obtained information or converted information, and as a result of said processing making a determination. Moreover, while components are depicted as single boxes located within a larger box, or nested within multiple boxes, in practice, computing devices may comprise multiple different physical components that make up a single illustrated component, and functionality may be partitioned between separate components. For example, a communication interface may be configured to include any of the components described herein, and / or the functionality of the components may be partitioned between the processing circuitry and the communication interface. In another example, non-computationally intensive functions of any of such components may be implemented in software or firmware and computationally intensive functions may be implemented in hardware.

[0143] In certain embodiments, some or all of the functionality described herein may be provided by processing circuitry executing instructions stored on in memory, which in certain embodiments may be a computer program product in the form of a non-transitory computer- readable storage medium. In alternative embodiments, some or all of the functionality may be provided by the processing circuitry without executing instructions stored on a separate or discrete device-readable storage medium, such as in a hard-wired manner. In any of those particular embodiments, whether executing instructions stored on a non-transitory computer- readable storage medium or not, the processing circuitry can be configured to perform the described functionality. The benefits provided by such functionality are not limited to the processing circuitry alone or to other components of the computing device, but are enjoyed by the computing device as a whole, and / or by end users and a wireless network generally.

Claims

Atty. Docket No.: 4906P109167WO01 CLAIMS What is claimed is:

1. A method for training a simulation of a Serializer / Deserializer (SerDes) receiver, the method comprising: based on a channel impulse response (CIR) input, selecting a first receiver equalization model from a plurality of receiver equalization models using a first optimization process, the first receiver equalization model having a corresponding set of model parameters (401); converting the CIR to an equalized CIR based on the selected receiver equalization model and corresponding set of model parameters (402); predicting a first bit error rate (BER) based on the first receiver equalization model and corresponding set of model parameters using a second optimization process (403); and based on a difference between the first BER and an expected or known BER, selecting a second receiver equalization model from the plurality of receiver equalization models using the first optimization process (404).

2. The method of claim 1, wherein the first optimization process comprises: selecting receiver equalization models from the plurality of receiver equalization models (413); and evaluating intermediate metrics associated with each receiver equalization model selected until the first receiver equalization model is selected having intermediate metrics which meet a defined threshold (415).

3. The method of claim 2, further comprising: updating the defined threshold of one or more of the intermediate metrics based on the difference between the first BER and an expected or known BER (417, 420).

4. The method of claim 2 or 3, wherein the intermediate metrics include a Figure of Merit (FOM) value (310, 504), wherein evaluating includes determining if the FOM value meets the defined threshold.

5. The method of any of claims 1 to 4, wherein the second optimization process (403) comprises a gradient flow process for training a differentiable Rx simulator which is to perform the predicting of the first BER.Atty. Docket No.: 4906P109167WO01 6. The method of claim 5, wherein the gradient flow process comprises iteratively updating one or more model parameters of the set of model parameters to reduce the difference between the first BER and an expected or known BER.

7. The method of claim 6, wherein iteratively updating the one or more model parameters to reduce the difference between the first BER and an expected or known BER further comprises: determining gradients of loss corresponding to each iteration of updating the one or more model parameters, the gradients of loss indicating a direction and magnitude of a subsequent update to the one or more model parameters that would reduce the loss; and performing the subsequent update to the one or more model parameters.

8. The method of claim 7, wherein backpropagation operations are performed for determining the gradients of loss.

9. The method of claim 1, wherein predicting the first BER comprises: configuring a plurality of noise spectral parameters (234, 236, 238, 240), at least a first noise spectral parameter of the plurality of noise spectral parameters configured based on a noise component of the equalized CIR (230); generating a total noise distribution (244) based on the plurality of noise spectral parameters; and using the total noise distribution for predicting the BER (248).

10. The method of claim 9, wherein the spectral parameters comprise a plurality of probability distribution functions (PDFs) (235, 237, 239, 241) corresponding to a plurality of noise sources and wherein the total noise distribution comprises a cumulative distribution function (CDF) of total noise.

11. The method of claim 2 wherein the plurality of receiver equalization models are to be provided to the first optimization process in a model zoo (312).

12. The method of any of claims 1 to 11 wherein different receiver equalization models of the plurality of receiver equalization models comprise different types or numbers of parameters.

13. The method of any of claims 1 to 12, wherein the first optimization process is to be performed at least once for each channel of a plurality of channels.Atty. Docket No.: 4906P109167WO01 14. The method of claim 13 wherein a plurality of instances of the first optimization process are to be executed in parallel for the plurality of channels.

15. A machine-readable medium having program code stored thereon which, when executed by a machine, causes the machine to perform operations for training a simulation of a Serializer / Deserializer (SerDes) receiver, the operations comprising: based on a channel impulse response (CIR) input, selecting a first receiver equalization model from a plurality of receiver equalization models using a first optimization process, the first receiver equalization model having a corresponding set of model parameters (401); converting the CIR to an equalized CIR based on the selected receiver equalization model and corresponding set of model parameters (402); predicting a first bit error rate (BER) based on the first receiver equalization model and corresponding set of model parameters using a second optimization process (403); and based on a difference between the first BER and an expected or known BER, selecting a second receiver equalization model from the plurality of receiver equalization models using the first optimization process (404).

16. The machine-readable medium of claim 15, wherein the first optimization process comprises: selecting receiver equalization models from the plurality of receiver equalization models (413); and evaluating intermediate metrics associated with each receiver equalization model selected until the first receiver equalization model is selected having intermediate metrics which meet a defined threshold (415).

17. The machine-readable medium of claim 16, further comprising program code to cause the machine to perform the additional operations of: updating the defined threshold of one or more of the intermediate metrics based on the difference between the first BER and an expected or known BER (417, 420).

18. The machine-readable medium of claim 16 or 17, wherein the intermediate metrics include a Figure of Merit (FOM) value (310, 504), wherein evaluating includes determining if the FOM value meets the defined threshold.Atty. Docket No.: 4906P109167WO01 19. The machine-readable medium of any of claims 15 to 18, wherein the second optimization process (403) comprises a gradient flow process for training a differentiable Rx simulator which is to perform the predicting of the first BER.

20. The machine-readable medium of claim 19, wherein the gradient flow process comprises iteratively updating one or more model parameters of the set of model parameters to reduce the difference between the first BER and an expected or known BER.

21. The machine-readable medium of claim 20, wherein iteratively updating the one or more model parameters to reduce the difference between the first BER and an expected or known BER further comprises: determining gradients of loss corresponding to each iteration of updating the one or more model parameters, the gradients of loss indicating a direction and magnitude of a subsequent update to the one or more model parameters that would reduce the loss; and performing the subsequent update to the one or more model parameters.

22. The machine-readable medium of claim 21, wherein backpropagation operations are performed for determining the gradients of loss.

23. The machine-readable medium of claim 15, wherein predicting the first BER comprises: configuring a plurality of noise spectral parameters (234, 236, 238, 240), at least a first noise spectral parameter of the plurality of noise spectral parameters configured based on a noise component of the equalized CIR (230); generating a total noise distribution based on the plurality of noise spectral parameters; and using the total noise distribution for predicting the BER (248).

24. The machine-readable medium of claim 23, wherein the spectral parameters comprise a plurality of probability distribution functions (PDFs) (235, 237, 239, 241) corresponding to a plurality of noise sources and wherein the total noise distribution comprises a cumulative distribution function (CDF) of total noise.

25. The machine-readable medium of claim 16 wherein the plurality of receiver equalization models are to be provided to the first optimization process in a model zoo (312).Atty. Docket No.: 4906P109167WO01 26. The machine-readable medium of any of claims 15 to 25 wherein different receiver equalization models of the plurality of receiver equalization models comprise different types or numbers of parameters.

27. The machine-readable medium of any of claims 15 to 26, wherein the first optimization process is to be performed at least once for each channel of a plurality of channels.

28. The machine-readable medium of claim 27 wherein a plurality of instances of the first optimization process are to be executed in parallel for the plurality of channels.

Images