Improve Estimation Performance Research Articles

A quantified approach of predicting suitability of using the Unscented Kalman Filter in a non-linear application

A mathematical framework to predict the Unscented Kalman Filter (UKF) performance improvement relative to the Extended Kalman Filter (EKF) using a quantitative measure of non-linearity is presented. It is also shown that the range of performance improvement the UKF can attain, for a given minimum probability depends on the Non-linearity Indices of the corresponding system and measurement models. Three distinct non-linear estimation problems are examined to verify these relations. A launch vehicle trajectory estimation problem, a satellite orbit estimation problem and a re-entry vehicle position estimation problem are examined to verify these relations. Using these relations, a procedure is suggested to predict the estimation performance improvement offered by the UKF relative to the EKF for a given non-linear system and measurement without designing, implementing and tuning the two Kalman Filters.

Research on sparse channel estimation algorithm based on MIMO OFDM system

In the application of MIMO OFDM system, the traditional pilot-assisted channel estimation algorithm, such as LS, MMSE algorithm, does not take advantage of the characteristics of channel sparsity, which results in the reduction of spectrum utilization due to the use of a large number of pilots. Under the premise of compressed sensing, this paper uses channel sparsity to study the AMP channel estimation algorithm based on MIMO OFDM system, and introduces the principles and steps of the AMP algorithm in detail. Through theoretical analysis and simulation, it is shown that the AMP channel estimation algorithm proposed in this paper has a significant improvement in spectrum utilization and estimation performance compared with the traditional method, which is more effective and greatly reduces the calculation time.

Computationally Efficient Ambiguity-Free Two-Dimensional DOA Estimation Method for Coprime Planar Array: RD-Root-MUSIC Algorithm

While the two-dimensional (2D) spectral peak search suffers from expensive computational burden in direction of arrival (DOA) estimation, we propose a reduced-dimensional root-MUSIC (RD-Root-MUSIC) algorithm for 2D DOA estimation with coprime planar array (CPA), which is computationally efficient and ambiguity-free. Different from the conventional 2D DOA estimation algorithms based on subarray decomposition, we exploit the received data of the two subarrays jointly by mapping CPA to the full array of the CPA (FCPA), which contributes to the enhanced degrees of freedom (DOFs) and improved estimation performance. In addition, due to the ambiguity-free characteristic of the FCPA, the extra ambiguity elimination operation can be avoided. Furthermore, we convert the 2D spectral search process into 1D polynomial rooting via reduced-dimension transformation, which substantially reduces the computational complexity while preserving the estimation accuracy. Finally, numerical simulations demonstrate the superiority of the proposed algorithm.

A comprehensive review for optimal placement of phasor measurement unit for network observability

Real time synchronized phasor measurement in power network is obtained by the improvement in monitoring, control and, protection of the power system. In recent time, the installation ratio of Phasor Measurement Units (PMUs) is constantly increasing for the real time measurement throughout worldwide. The increment in the number of PMU installation is to only focus on the improvement of system state estimation (SE) performance. However, the expensive nature of the metering device requires huge amount of installation cost with the other communication facilities, therefore an optimal placement of PMU is necessary. Different techniques have been designed and used to overcome this matter. The paper presents numerous optimization algorithms such as, Mathematical programming, Heuristic, and Meta-Heuristic techniques which are specially used for the optimization of PMU placement with complete network observability. Furthermore, each PMU technique is explained, and performances are compared for the most appropriate and optimal placement of PMU methods, which can be recommended for a future work to get complete network observability.

A Generalized VAMP-Based Channel Estimator for Uplink Quantized Massive MIMO Systems

In this study, we propose an improved channel estimator based on vector approximate message passing (VAMP) for quantized millimeter wave (mmWave) massive multiple-input and multiple-output (MIMO) systems. Using channel sparsity and the orthogonality of steering matrix, the proposed VAMP-on method designed for the on-grid scenario not only improves estimation performance but also has better convergence and inverse-free implementation. Furthermore, to achieve better estimation performance under the off-grid scenario, we provide a VAMP-off algorithm to update steering matrix and channel coefficients in an alternating fashion. Finally, simulation results are provided to demonstrate the advantages of the proposed approaches.

Estimating dynamic binary choice models using irregularly spaced panel data

This paper investigates the irregular spacing problem when estimating dynamic binary choice models with panel data. We show that traditional approaches tend to underestimate the state dependence parameter. We propose a new gap dummy-variable estimator to improve estimation performance.

Model-Based Reinforcement Learning for Physical Systems Without Velocity and Acceleration Measurements

In this letter, we propose a derivative-free model learning framework for Reinforcement Learning (RL) algorithms based on Gaussian Process Regression (GPR). In many mechanical systems, only positions can be measured by the sensing instruments. Then, instead of representing the system state as suggested by the physics with a collection of positions, velocities, and accelerations, we define the state as the set of past position measurements. However, the equation of motions derived by physical first principles cannot be directly applied in this framework, being functions of velocities and accelerations. For this reason, we introduce a novel derivative-free physically-inspired kernel, which can be easily combined with nonparametric derivative-free Gaussian Process models. Tests performed on two real platforms show that the considered state definition combined with the proposed model improves estimation performance and data-efficiency w.r.t. traditional models based on GPR. Finally, we validate the proposed framework by solving two RL control problems for two real robotic systems.

Adaptive least squares channel estimation for visible light communications based on tap detection

Visible light communication (VLC) is one of the promising technologies for future wireless communications and has attracted increasing research interest in recent years. In this paper, we propose an adaptive least squares (LS) channel estimation algorithm for VLC systems using the direct current biased optical orthogonal frequency division multiplexing (DCO-OFDM) scheme. The proposed algorithm is obtained by employing the tap detection that provides the maximum probability of correctly separating non-zero and zero-value channel taps, yielding significantly improved estimation performance. We prove that the Cramér–Rao lower bound of the proposed algorithm is lower than that of the conventional LS algorithm. Finally, numerical results demonstrate that the proposed algorithm achieves better performance in terms of the received constellation, mean squared error (MSE) and bit error rate (BER).

Future Uncertainty-Based Control for Relative Navigation in GPS-Denied Environments

In a nonlinear system where the separation principle does not hold, control actions at the current time can have a significant effect on estimation accuracy in the future. To properly control this type of system with a model predictive controller requires an accurate estimate of the future uncertainty of the system. To compute this estimate, we propose a definition of the nonlinear observability matrix with respect to a future state and establish its connection with the filtering Cramer–Rao lower bound at the future state. We also define a sensitivity function that correlates noise on control inputs to errors on the state estimate. As a demonstrative example, we consider the problem of relative navigation of two unmanned aerial systems in Global Positioning System denied environments. We demonstrate using Monte Carlo simulations that by controlling to optimize our derived observability and controllability metrics, we achieve significantly improved estimation performance over a trajectory optimized for an uncertainty estimate that does not consider input noise.

A Novel Multichannel Long Short-Term Memory Method With Time Series for Soil Temperature Modeling

Soil temperature (Ts) plays an important role in earth sciences. The temporal and spatial variations of Ts are affected by several factors. To acquire the deterministic component and the stochastic component of time series data and further improve estimation performance, a multichanne long short-term memory model (MC-LSTM) is developed to address the challenge of accurate Ts estimation. Specifically, we connect multiple channels in an LSTM structure and an autoregressive integrated moving average model (ARIMA) directly to the output layer to characterize the deterministic part and the stochastic component of time series data. Meanwhile, to improve the correlations in the time series by increasing the number of memory units, we also connect the units in the different steps directly to the output layer in the LSTM structure to learn the long-term pattern of historical Ts by connecting the unit in an earlier step directly to the output layer, to learn the features that occurred in a short-term pattern by connecting the unit to the later step captures. In this article we test the hypothesis that Ts estimation is performed at depths of 5, 10 and 15 cm over 6, 12, and 24 hours. To achieve this, the input data are composed of half-hour data from the Ts of two synoptic stations (Laegern and Fluehli) in Switzerland. Meanwhile, the estimation accuracy is verified by three performance criteria; RMSE, MAE and R2. As expected, the proposed model achieves the highest relative R2 values of 0.9965 and the lowest values of RMSE = 0.3414 and MAE = 0.2310 for Ts estimation over 6 hours at Fluehli Station (10 cm soil depth) when compared with other state-of-the-art machine learning methods. Consequently, the proposed model can serve as an alternative approach for Ts profile estimation.

Joint Array Diagnosis and Channel Estimation for RIS-Aided mmWave MIMO System

In this paper, we consider a reconfigurable intelligent surface (RIS) aided millimeter wave (mmWave) multiple-input multiple-output (MIMO) system. The system can obtain the huge gain via joint active beamforming at the base station (BS) and passive beamforming at the RIS. However, due to weather and atmospheric effects, outdoor RIS antenna elements are subject to full or partial blockages from a plethora of particles like dirt, salt, ice, and water droplets. These blockages can cause an approximate squared power/SNR loss for the system. Different from the conventional array diagnosis, the RIS has no signal processing capability. Thus, we propose the joint array diagnosis and channel estimation techniques containing two stages to solve the problem. At the first stage the channel parameters at user equipment (UE) and BS are estimated using an iterative reweighted (IR) method. At the second stage, the array blockage coefficient vector and the effective sparse channel parameters at RIS are jointly estimated via solving a two-timescale non-convex optimization problem. We propose two algorithms, i.e., a batch algorithm (BA) and a two-timescale online joint array diagnosis and channel estimation (TOJADCE) algorithm to solve the problem and compare the performance of these two algorithms. Finally, to speed up the convergence of long-term variable and improve estimation performance, we propose a noise reduction (NR) algorithm. The simulations verify the effectiveness of our proposed algorithms.

Cramér-Rao Bounds for Direction Estimation in the Presence of Circular and Noncirular Signals

In direction finding, the use of noncircularity of signals arriving at a sensor array can allow us to improve estimation performance. It will be more realistic, rather than the assumption of all noncircular sources, to consider some sources transmitting circular signals. In the presence of both circular and noncircular signals, this paper presents the deterministic Cramer-Rao bound (CRB) of direction estimation. The derivation of the CRB is quite general so that the bounds with unknown signal structures and with known waveforms can be readily obtained from it. Moreover, we show that the more information on signal structures is available the less the CRB is. And we present general conditions under which the CRBs for the direction parameters obtained from different Fisher information matrices are equal. Based on these basic theoretical results, the bounds with different information are analytically compared. Numerical results confirm the theoretical results.

INS/gravity gradient aided navigation based on gravitation field particle filter

Abstract Swarm intelligence method is an effective way to improve the particle degradation and sample depletion of the traditional particle filter. This paper proposes a particle filer based on the gravitation field algorithm (GF-PF), and the gravitation field algorithm is introduced into the resampling process to improve particle degradation and sample depletion. The gravitation field algorithm simulates the solar nebular disk model, and introduces the virtual central attractive force and virtual rotation repulsion force between particles. The particles are moves rapidly to the high-likelihood region under action of the virtual central attractive force. The virtual rotation repulsion force makes the particles keep a certain distance from each other. These operations improve estimation performance, avoid overlapping of particles and maintain the diversity of particles. The proposed method is applied into INS/gravity gradient aided navigation, by combining the sea experimental data of an inertial navigation system. Compared with the particle swarm optimization particle filter(PSO-PF) and artificial physics optimized particle filter (APO-PF), the GF-PF has higher position estimate accuracy and faster convergence speed with the same experimental conditions.

SPICE-ML Algorithm for Direction-of-Arrival Estimation.

Sparse iterative covariance-based estimation, an iterative direction-of-arrival approach based on covariance fitting criterion, can simultaneously estimate the angle and power of incident signal. However, the signal power estimated by sparse iterative covariance-based estimation approach is inaccurate, and the estimation performance is limited to direction grid. To solve the problem above, an algorithm combing the sparse iterative covariance-based estimation approach and maximum likelihood estimation is proposed. The signal power estimated by sparse iterative covariance-based estimation approach is corrected by a new iterative process based on the asymptotically minimum variance criterion. In addition, a refinement procedure is derived by minimizing a maximum likelihood function to overcome the estimation accuracy limitation imposed by direction grid. Simulation results verify the effectiveness of the proposed algorithm. Compared with sparse iterative covariance-based estimation approach, the proposed algorithm can achieve more accurate signal power and improved estimation performance.

The sequential algorithm for combined state of charge and state of health estimation of lithium-ion battery based on active current injection

When the State of Charge, State of Health, and parameters of a Lithium-ion battery are estimated simultaneously, estimation accuracy is hard to be ensured due to uncertainties in the estimation process. A sequential algorithm, which uses frequency-scale separation and estimates parameters/states sequentially by injecting currents with different frequencies, is proposed in this paper to improve estimation performance. Specifically, by incorporating a high-pass filter, the parameters can be independently characterized by injecting high-frequency and medium-frequency currents, respectively. Using the estimated parameters, battery capacity and State of Charge can then be estimated concurrently. Experimental results show that the estimation accuracy of the proposed sequential algorithm is much better than the concurrent algorithm where all parameters/states are estimated simultaneously, and the computational cost can also be reduced. Finally, experiments are conducted at different temperatures to verify the effectiveness of the proposed algorithm for varying battery capacities.

Graph convolutional neural network for multi-scale feature learning

Automatic deformable 3D modeling is computationally expensive, especially when considering complex position, orientation and scale variations. We present a volume segmentation framework to utilize local and global regularizations in a data-driven approach. We introduce automated correspondence search to avoid manually labeling landmarks and improve scalability. We propose a novel marginal space learning technique, utilizing multi-resolution pooling to obtain local and contextual features without training numerous detectors or excessively dense patches. Unlike conventional convolutional neural network operators, graph-based operators allow spatially related features to be learned on the irregular domain of the multi-resolution space, and a graph-based convolutional neural network is proposed to learn representations for position and orientation classification. The graph-CNN classifiers are used within a marginal space learning framework to provide efficient and accurate shape pose parameter hypothesis prediction. During segmentation, a global constraint is initially non-iteratively applied, with local and geometric constraints applied iteratively for refinement. Comparison is provided against both classical deformable models and state-of-the-art techniques in the complex problem domain of segmenting aortic root structure from computerized tomography scans. The proposed method shows improvement in both pose parameter estimation and segmentation performance.

Estimating Biomechanical Time-Series with Wearable Sensors: A Systematic Review of Machine Learning Techniques.

Wearable sensors have the potential to enable comprehensive patient characterization and optimized clinical intervention. Critical to realizing this vision is accurate estimation of biomechanical time-series in daily-life, including joint, segment, and muscle kinetics and kinematics, from wearable sensor data. The use of physical models for estimation of these quantities often requires many wearable devices making practical implementation more difficult. However, regression techniques may provide a viable alternative by allowing the use of a reduced number of sensors for estimating biomechanical time-series. Herein, we review 46 articles that used regression algorithms to estimate joint, segment, and muscle kinematics and kinetics. We present a high-level comparison of the many different techniques identified and discuss the implications of our findings concerning practical implementation and further improving estimation accuracy. In particular, we found that several studies report the incorporation of domain knowledge often yielded superior performance. Further, most models were trained on small datasets in which case nonparametric regression often performed best. No models were open-sourced, and most were subject-specific and not validated on impaired populations. Future research should focus on developing open-source algorithms using complementary physics-based and machine learning techniques that are validated in clinically impaired populations. This approach may further improve estimation performance and reduce barriers to clinical adoption.

Soft‐decision schemes for radar estimation of elevation at low grazing angles

In modern radars, the problem of estimating elevation angle at low grazing angles is typically solved using super-resolution techniques. These techniques often require one to provide an estimate of the number of waveforms impinging the array, which one can accomplish using model selection techniques. In this study, the authors investigate the performance of an alternative approach, based on the Bayesian-like model averaging. The Bayesian approach exploits the fact that the parameters of the model related to multipath signals are nuisance ones, which allows one to avoid the estimation of the number of waveforms and improves estimation performance. The method is introduced for the classical conditional maximum-likelihood estimator and extended to its, recently proposed, robustified version. The authors find, however, that the robustified estimator includes its own soft-decision mechanism and benefits from the averaging only for low levels of model uncertainty.

Forming Distributed State Estimation Network From Decentralized Estimators

In this paper, we focus on the distributed state estimation of nonlinear systems comprised of several subsystems. We assume that a decentralized state estimation system already exists for the nonlinear system, where the local estimators can be of different types. In order to achieve improved estimation performance, the existing decentralized estimators may be connected together via a communication network to form a distributed state estimation network. We propose a systematic approach to take advantage of the existing decentralized estimators potentially of different types to form a distributed state estimation network without performing a complete redesign of the estimation system. Specifically, a compensator is designed for each subsystem, and is connected to the corresponding decentralized estimator to obtain an augmented estimator (AE). The AEs for the subsystems communicate with each other to exchange subsystem state estimates and measurements via a communication network every sampling time. We derive sufficient conditions on the convergence and boundedness of the estimation error of the proposed distributed estimation network. The proposed approach is demonstrated via the application to two chemical process examples and one hybrid tank plant.

Fast Depth Estimation in a Single Image Using Lightweight Efficient Neural Network.

Depth estimation is a crucial and fundamental problem in the computer vision field. Conventional methods re-construct scenes using feature points extracted from multiple images; however, these approaches require multiple images and thus are not easily implemented in various real-time applications. Moreover, the special equipment required by hardware-based approaches using 3D sensors is expensive. Therefore, software-based methods for estimating depth from a single image using machine learning or deep learning are emerging as new alternatives. In this paper, we propose an algorithm that generates a depth map in real time using a single image and an optimized lightweight efficient neural network (L-ENet) algorithm instead of physical equipment, such as an infrared sensor or multi-view camera. Because depth values have a continuous nature and can produce locally ambiguous results, pixel-wise prediction with ordinal depth range classification was applied in this study. In addition, in our method various convolution techniques are applied to extract a dense feature map, and the number of parameters is greatly reduced by reducing the network layer. By using the proposed L-ENet algorithm, an accurate depth map can be generated from a single image quickly and, in a comparison with the ground truth, we can produce depth values closer to those of the ground truth with small errors. Experiments confirmed that the proposed L-ENet can achieve a significantly improved estimation performance over the state-of-the-art algorithms in depth estimation based on a single image.