Complex Parameter Estimation Research Articles

Variational Bayesian Estimation of Quantile Nonlinear Dynamic Latent Variable Models with Possible Nonignorable Missingness

Our study presents an innovative variational Bayesian parameter estimation method for the Quantile Nonlinear Dynamic Latent Variable Model (QNDLVM), particularly when dealing with missing data and nonparametric priors. This method addresses the computational inefficiencies associated with the traditional Markov chain Monte Carlo (MCMC) approach, which struggles with large datasets and high-dimensional parameters due to its prolonged computation times, slow convergence, and substantial memory consumption. By harnessing the deterministic variational Bayesian framework, we convert the complex parameter estimation into a more manageable deterministic optimization problem. This is achieved by leveraging the hierarchical structure of the QNDLVM and the principle of efficiently optimizing approximate posterior distributions within the variational Bayesian framework. We further optimize the evidence lower bound using the coordinate ascent algorithm. To specify propensity scores for missing data manifestations and covariates, we adopt logistic and probit models, respectively, with conditionally conjugate mean field variational Bayes for logistic models. Additionally, we utilize Bayesian local influence to analyze the Ecological Momentary Assessment (EMA) dataset. Our results highlight the variational Bayesian approach’s notable accuracy and its ability to significantly alleviate computational demands, as demonstrated through simulation studies and practical applications.

Enhanced parameter estimation in multiparametric arterial spin labeling using artificial neural networks.

Multiparametric arterial spin labeling (MP-ASL) can quantify cerebral blood flow (CBF) and arterial cerebral blood volume (CBVa). However, its accuracy is compromised owing to its intrinsically low SNR, necessitating complex and time-consuming parameter estimation. Deep neural networks (DNNs) offer a solution to these limitations. Therefore, we aimed to develop simulation-based DNNs for MP-ASL and compared the performance of a supervised DNN (DNNSup), physics-informed unsupervised DNN (DNNUns), and the conventional lookup table method (LUT) using simulation and in vivo data. MP-ASL was performed twice during resting state and once during the breath-holding task. First, the accuracy and noise immunity were evaluated in the first resting state. Second, CBF and CBVa values were statistically compared between the first resting state and the breath-holding task using the Wilcoxon signed-rank test and Cliff's delta. Finally, reproducibility of the two resting states was assessed. Simulation and first resting-state analyses demonstrated that DNNSup had higher accuracy, noise immunity, and a six-fold faster computation time than LUT. Furthermore, all methods detected task-induced CBF and CBVa elevations, with the effect size being larger with the DNNSup (CBF, p = 0.055, Δ = 0.286; CBVa, p = 0.008, Δ = 0.964) and DNNUns (CBF, p = 0.039, Δ = 0.286; CBVa, p = 0.008, Δ = 1.000) than that with LUT (CBF, p = 0.109, Δ = 0.214; CBVa, p = 0.008, Δ = 0.929). Moreover, all the methods exhibited comparable and satisfactory reproducibility. DNNSup outperforms DNNUns and LUT with respect to estimation performance and computation time.

Dual-Polarization Ambient Backscatter Communications and Signal Detection.

Ambient backscatter communication (AmBC), an emerging mechanism for batteryless communications that can utilize ambient radio-frequency signals to modulate information and thus reduce power consumption, has attracted considerable attention and has been considered as a critical technology in green "Internet of Things" sensor networks due to its ultra-low power consumption. This paper presents the first a complete dual-polarization AmBC (DPAm) system model, which can extend AmBC into polarization diversity and improve the data-transmission rate of backscatter symbols. We proposed two scenarios: direct dual-polarization-based DPAm node structures and polarization-conversion-based DPAm node structures. In addition, we consider a parallel backscatter mode with differential coding and develop corresponding detectors, which also give the analytical detection thresholds. Moreover, we consider a simultaneous backscatter mode with Manchester coding in order to avoid complex-parameter estimation. To address the power imbalance problem of the DPAm system that arises because the polarization deflection angle would cause the power level to change with different polarization patterns, we also develop a power-average detector and a clustering detector. Simulation results show the throughput performance on each DPAm node structure with each detector, demonstrating the feasibility and efficiency of the proposed DPAm nodes and detectors. Compared with single-polarization AmBC (SPAm), the proposed DPAm node can achieve higher throughput in most cases. Finally, the clustering detector is shown to be more robust to short training sequences and complex environments.

Using deep learning to accelerate magnetic resonance measurements of molecular exchange.

Real-time monitoring and quantitative measurement of molecular exchange between different microdomains are useful to characterize the local dynamics in porous media and biomedical applications of magnetic resonance. Diffusion exchange spectroscopy (DEXSY) is a noninvasive technique for such measurements. However, its application is largely limited by the involved long acquisition time and complex parameter estimation. In this study, we introduce a physics-guided deep neural network that accelerates DEXSY acquisition in a data-driven manner. The proposed method combines sampling pattern optimization and physical parameter estimation into a unified framework. Comprehensive simulations and experiments based on a two-site exchange system are conducted to demonstrate this new sampling optimization method in terms of accuracy, repeatability, and efficiency. This general framework can be adapted for other molecular exchange magnetic resonance measurements.

FPGA Implementation of Modified Reconfigurable Adaptive Gain Scheduling Controller

This article aims to put forward a modified type of adaptive gain scheduling that will be able to deal with the immeasurable and unpredictable variations of system variables by adapting its value at each time instance to follow any change in the input and overcome any disturbance applied to the system without the need to predetermine gains values. In addition, the inverse neural controller will precede the gain scheduling to eliminate the need for complex system linear zing and parameter estimation. Therefore, the problems of needing complex mathematics for system linearization and gains calculations have been solved. The performance of the presented controller was tested by comparing the step response of a DC-motor controlled via the proposed technique and the response of that motor when controlled by the inverse neural controller and PID controller. MATLAB/Simulink has been used for making the simulations and obtaining the results. In addition, the FPGA implementation of the proposed controller has been presented. The results showed a remarkable improvement in the transient response of the system for all of the rising time, delay time, settling time, peak overshoot, and steady-state error.

Identification and parameter estimation algorithm of radar signal subtle features

With the rapid development of electronic technology and its widespread application in modern warfare, the modulation methods of individual radar signal sources are more flexible, the parameters are diverse, and the modern battlefield electromagnetic signal environment is complex, so that radar signal recognition based on traditional conventional parameters Technology cannot meet real needs. Among the many radar signal modulation recognition algorithms that have appeared in recent years, neural network theory and intra-pulse analysis algorithms are widely used, but they each have their own shortcomings, and a single method cannot meet the needs of increasingly complex modulation signal recognition and parameter estimation needs. . Based on the above background, the research content of this paper is to identify the subtle characteristics of radar signals and their parameter estimation algorithms. In this paper, two types of typical modulation signals in intentional modulation of radar signals, namely phase-coded (PSK) signals and frequency-modulated (FM) signals, are proposed. A SVEFD algorithm based on classification from coarse to fine. According to the characteristics of the 3 dB bandwidth of the frequency spectrum of the PSK signal and the FM signal, the coarse classification between the classes is performed first. Finally, through experimental simulations, the results show that the correct recognition probability of the SVEFD algorithm at a lower signal-to-noise ratio is much higher than the m algorithm and also higher than the SVD algorithm. When the signal-to-noise ratio is greater than 1 dB, the average recognition rate reaches 94%, which proves that The effectiveness of the proposed SVEFD algorithm. When the signal-to-noise ratio is .1 dB, in addition to maintaining the correct recognition rate of more than 90% for the LFM and COSTAS signals, the other six types of signals have low correct recognition rates, so the SVEFD algorithm has advantages in comparison.

Adaptive Testing Based on Moment Estimation

Adaptive testing (AT) is a software testing approach that uses a feedback mechanism to enhance test effectiveness. Its testing strategy can be adjusted online by using the testing data collected during the software testing process. However, it requires complex parameter estimation which results in excessive computational overhead that may hinder the applicability of AT. In this paper, we propose an approach called AT based on moment estimation (AT-ME) to address this problem. The proposed approach uses moment estimation to serve as the algorithm of parameter estimation, which reduces the complexity of AT-ME. In addition, a dynamic length for testing action is set to limit the number of decisions without influencing the test effectiveness. The proposed approach has been validated on the Siemens test suite, which includes seven real programs. The experiments show that AT-ME can reduce the computational overhead of AT without compromising overall testing efficiency. Results demonstrate that AT-ME is a feasible and effective AT strategy.

Research on Numerically Solving the Inverse Problem Based on L1 Norm

In this paper, the numerical solution method based on L1 norm for inverse problem is studied. First, according to the theory of the L1 norm and the characteristics of the problem to be solved, a cost function is constructed. Further, for complex parameter estimation problems and derivative discontinuities, the regularization method is used to construct the cost function and the Huber function is used in the derivative discontinuity. Finally, the problem is solved numerically by the semi-smooth Newton method. The experimental results show that the method based on L1 norm is an effective method under the interference of non-Gaussian noises. For the parameter estimation of complex models, the results based on the L1 norm method are very closer to the real parameters when there is impulse noise. For parameter estimation under non-Gaussian noise, the L1 norm estimation method has significant advantages over the L2 norm method.

A Nonparametric Paired Echo Suppression Method for Helicopter-Borne SAR Imaging

The helicopter-borne synthetic aperture radar (SAR) system suffers high-frequency vibrations caused by the rotor rotation. This micro-vibration induces phase modulation, which gives rise to paired echoes and seriously degrades the quality of SAR images due to the unexpected ghost phenomenon. In this letter, a nonparametric paired echo suppression method for helicopter-borne SAR imaging is proposed. By constructing a modified vibration error model and introducing the autofocus method for the vibration phase error estimation, it can suppress the paired echoes directly, which avoids the complex vibration parameter estimation in the traditional methods. The effectiveness of the proposed method is demonstrated via simulations and real data processing.

A Sensorless Rotational Speed-Based Control System for Continuous Flow Left Ventricular Assist Devices

Continuous Flow Left Ventricular Assist Devices (CFLVAD) are circulatory support devices that are implanted in patients with end-stage heart failure. We developed a novel control algorithm for CFLVAD to maintain physiologic perfusion while avoiding ventricular suction using only the intrinsic pump measurement of pump speed and without utilizing model-based estimation. The controller objective is to maintain a differential pump speed setpoint. A mathematical model of the circulatory system coupled with a model of a CFLVAD was used to test the control algorithm in silico. Robustness and efficacy were evaluated by comparing the proposed control algorithm to constant speed control, differential pump pressure control, mean aortic pressure control, and ventricular end diastolic pressure control during (1) rest and exercise conditions, (2) a rapid eight-fold increase in pulmonary vascular resistance under rest and exercise, (3) transitions from rest to exercise, and exercise to rest, (4) safe mode during left ventricular asystole, and (5) RPM measurement noise of 1% to 10% for (1) to (4). The control algorithm provided adequate perfusion while preventing ventricular suction for all test conditions. Performance did not deteriorate significantly with pump speed measurement noise of up to 6%. The safe mode successfully detected asystole and maintained adequate perfusion to sustain life even when the differential pump speed was low. Maintaining a constant differential pump speed can simultaneously achieve physiologic perfusion and suction prevention without needing unreliable, direct measurements of flow or pressure, or complex parameter or model-based estimation techniques.

Liquidity stress testing: a model for a portfolio of credit lines

In this paper, we demonstrate how cash outflows due to credit lines can be modeled in a liquidity stress test. Our model is based on bootstrapping from a portfolio time series of daily credit line drawdowns. Key features of our model are (i) that it does not rely on any distributional assumptions or any complex parameter estimation, ie, the model risk is low; (ii) that it is intuitive and straightforward to implement; (iii) that it is very flexible and allows the portfolio's free amount to be reduced during the stress test horizon; and (iv) that it calculates an outflow profile with daily granularity. In a detailed simulation study, we demonstrate the reasonable behavior of the model output toward changes in several input parameters.

An algorithmic calibration approach to identify globally optimal parameters for constraining the DayCent model

The accurate calibration of complex biogeochemical models is essential for the robust estimation of soil greenhouse gases (GHG) as well as other environmental conditions and parameters that are used in research and policy decisions. DayCent is a popular biogeochemical model used both nationally and internationally for this purpose. Despite DayCent's popularity, its complex parameter estimation is often based on experts’ knowledge which is somewhat subjective. In this study we used the inverse modelling parameter estimation software (PEST), to calibrate the DayCent model based on sensitivity and identifiability analysis. Using previously published N2O and crop yield data as a basis of our calibration approach, we found that half of the 140 parameters used in this study were the primary drivers of calibration differences (i.e. the most sensitive) and the remaining parameters could not be identified given the data set and parameter ranges we used in this study. The post calibration results showed improvement over the pre-calibration parameter set based on, a decrease in residual differences 79% for N2O fluxes and 84% for crop yield, and an increase in coefficient of determination 63% for N2O fluxes and 72% for corn yield. The results of our study suggest that future studies need to better characterize germination temperature, number of degree-days and temperature dependency of plant growth; these processes were highly sensitive and could not be adequately constrained by the data used in our study. Furthermore, the sensitivity and identifiability analysis was helpful in providing deeper insight for important processes and associated parameters that can lead to further improvement in calibration of DayCent model.

Planar coordinate transformation and its parameter estimation in the complex number field

In this paper, the planar coordinate transformation model in the complex field is constructed. It is more concise than the traditional planar coordinate transformation model in the real field, because it combines the x and y coordinates as a whole. In order to carry out the complex parameter estimation, the paper introduces the complex least squares and the complex Gauss-Jacobi combinatorial algorithm. A simulative numerical case study and an actual one are given to investigate the correctness and efficiency of the presented approaches. The results show that the complex least squares for the complex transformation model obtains the identical estimation as the conventional least squares for the real transformation model, thus it is a reliable least squares for the complex parameter estimation. However, for the case of gross error, it becomes invalid. On the contrary, the complex Gauss-Jacobi combinatorial algorithm is capable of detecting the gross error based on its parameter estimations of numerous combinations. Consequently, it is verified as a good robust estimation by the numerical case study.

On the approximation of empirical data for service system simulations

The difficulties associated with parameter estimation for phase-type approximations of empirical data distributions in queue modelling are well known. While significant progress has been achieved in improving such approximations, difficulties in parameter estimation still limit the extent to which queue modelling is applied in practice. This paper presents a simplified technique for approximating empirical data in service system simulations, based on the specialised Cox phase-type distribution. When utilised in simulation modelling of a service system, the specialised Cox distribution is shown to provide improved approximations to the combined waiting and service time distribution without the need for complex parameter estimation techniques. This approach should enable much greater flexibility in the application of queue modelling to service systems.

Improving the calibration strategy of the physically-based model WaSiM-ETH using critical events

The use of a physically-based hydrological model for streamflow forecasting is limited by the complexity in the model structure and the data requirements for model calibration. The calibration of such models is a difficult task, and running a complex model for a single simulation can take up to several days, depending on the simulation period and model complexity. The information contained in a time series is not uniformly distributed. Therefore, if we can find the critical events that are important for identification of model parameters, we can facilitate the calibration process. The aim of this study is to test the applicability of the Identification of Critical Events (ICE) algorithm for physically-based models and to test whether ICE algorithm-based calibration depends on any optimization algorithm. The ICE algorithm, which uses the data depth function, was used herein to identify the critical events from a time series. Low depth in multivariate data is an unusual combination and this concept was used to identify the critical events on which the model was then calibrated. The concept is demonstrated by applying the physically-based hydrological model WaSiM-ETH on the Rems catchment, Germany. The model was calibrated on the whole available data, and on critical events selected by the ICE algorithm. In both calibration cases, three different optimization algorithms, shuffled complex evolution (SCE-UA), parameter estimation (PEST) and robust parameter estimation (ROPE), were used. It was found that, for all the optimization algorithms, calibration using only critical events gave very similar performance to that using the whole time series. Hence, the ICE algorithm-based calibration is suitable for physically-based models; it does not depend much on the kind of optimization algorithm. These findings may be useful for calibrating physically-based models on much fewer data. Editor D. Koutsoyiannis; Associate editor A. Montanari Citation Singh, S.K., Liang, J.Y., and Bárdossy, A., 2012. Improving calibration strategy of physically-based model WaSiM-ETH using critical events. Hydrological Sciences Journal, 57 (8), 1487–1505.

A novel P systems based optimization algorithm for parameter estimation of proton exchange membrane fuel cell model

Accurate kinetic models are of great significance for the simulation and analysis for hydrogen fuel cells. The proton exchange membrane (PEM) fuel cell is a complex nonlinear, multi-variable system. The mathematical modeling of PEM fuel cell usually leads to nonlinear parameter estimation problems which often contain more than one minimum. In this paper, a novel bio-inspired P systems based optimization algorithm, named BIPOA, is proposed to solve PEM fuel cell model parameter estimation problems. In BIPOA, the nested membrane structure and new rules such as adaptive mutation rule, partial migration rule and autophagy rule are combined to improve the algorithm's global search capacities and convergence accuracy. Studies on some benchmark test functions indicate that the BIPOA outperforms the other two methods (PSOPS and GAs) in both convergence speed and accuracy. In addition, experimental results reveal that the model predictive outputs are in better agreement with the actual experimental data. Therefore, the BIPOA is a helpful and reliable technique for estimating the PEM fuel cell model parameters and is available to other complex parameter estimation problems of fuel cell models.

Trading link utilization for queueing delays: An adaptive approach

Understanding the relationship between queueing delays and link utilization for general traffic conditions is an important open problem in networking research. Difficulties in understanding this relationship stem from the fact that it depends on the complex nature of arriving traffic and the problems associated with modelling such traffic. Existing AQM schemes achieve a ''low delay'' and ''high utilization'' by responding early to congestion without considering the exact relationship between delay and utilization. However, in the context of exploiting the delay/utilization tradeoff, the optimal choice of a queueing scheme's control parameter depends on the cost associated with the relative importance of queueing delay and utilization. The optimal choice of control parameter is the one that maximizes a benefit that can be defined as the difference between utilization and cost associated with queuing delay. We present two practical algorithms, Optimal Drop-Tail (ODT) and Optimal BLUE (OB), that are designed with a common performance goal: namely, maximizing this benefit. Their novelty lies in fact that they maximize the benefit in an online manner, without requiring knowledge of the traffic conditions, specific delay-utilization models, nor do they require complex parameter estimation. Packet level ns2 simulations are given to demonstrate the efficacy of the proposed algorithms and the framework in which they are designed.

Parameter determination and response analysis of viscoelastic material

The parameter determination of viscoelastic material is a multi-variable, multi-aim nonlinear optimization problem, which made the optimization process very complicated. In this paper a hybrid optimal algorithm was proposed to determine the viscoelastic parameters in the constitutive relation according to the experimentally obtained mechanical properties. This algorithm merges the Broydon–Fletcher–Goldfarb–Shanno search into a genetic algorithm framework as a basic operator in order to enhance the local search capability. The proposed hybrid algorithm not only can reduce the iterative times greatly but can abolish the limitation of initial parameter values. Nonlinear material characteristic curve-fitting was carried out using the proposed algorithm and other existing approaches. And the comparison results show this algorithm is accurate and effective. The numerical simulation and experimental study of viscoelastic cantilever beam also indicates that the finite element formulation and the calculative viscoelastic model parameters are reliable. The proposed optimization method can be extended to further complex parameter estimation researches.

On the similarity of the sensitivity functions of methane combustion models

It is widely known that detailed kinetic mechanisms with identical reaction steps but with very different rate parameters may provide similar simulation results in combustion calculations. This phenomenon is related to the similarity of sensitivity functions, which arises if low-dimensional manifolds in the space of variables, and autocatalytic processes are present. We demonstrated the similarity of sensitivity functions for adiabatic explosions and burner-stabilized laminar flames of stoichiometric methane–air mixtures. The cause of similarities was investigated by calculating the dimension of the corresponding manifolds, and the pseudo-homogeneous property of the sensitivity ordinary differential equation (ODE). The methane explosion model showed global similarity, which means that different parameter sets could provide the same simulation results. This was demonstrated by numerical experiments, in which two significantly different parameter sets resulted in identical concentration profiles for all species. This phenomenon is important from a practical point of view in the fields of ‘validation’ of complex reaction mechanisms and parameter estimation of chemical kinetic systems.

A statistical assessment on the stochastic relationship between biomarker concentrations and environmental exposures

Biological monitoring has gradually developed into a powerful tool in the identification and quantification of exposures to occupational and/or environmental hazards in environmental and occupational health studies. Aggregate individual exposure to pollutants and evidence for exploring dose-response relationship in the human bodies can be assessed through biomarker measurements. The existence of inter-individual differences among a study population, however, often hampers the relationship assessment between exposure and the biomarker. In this paper, a statistical random effects model identified from a simplified one-compartment pharmacokinetic model is applied to establish the dynamic relationship between a biomarker and its corresponding external exposure. This model avoids the complex parameter estimation problem encountered using a physiologically based pharmacokinetic (PBPK) model, and incorporates inter-individual variations often ignored in the usual regression approach. In addition, the relevant parameters for the generic kinetic process can be estimated directly. As a guideline for preliminary sampling strategy, tables of required sample sizes and the number of repeated measurements to achieve the desired statistical power and test efficiency are given. The currently established biological exposure indices (BEIs) for benzene and methyl chloroform are employed to illustrate the impact of inter- individual variations on the percentages of protection for workers exposed to the threshold limit value (TLV) of the corresponding chemical.