Probabilistic principal component regression (PPCR) has been introduced for soft sensor modeling as a probabilistic projection method, which is effective in handling data collinearity and random noises. However, the linear limitation of relationships may cause its performance deterioration when applied to nonlinear processes. Therefore, novel weighted PPCR (WPPCR) algorithm proposed this paper sensing In WPPCR, by including most relevant samples local modeling, different weights will be...

Fingerprint matching is a key issue in research of an automatic fingerprint identification system. On the basis Delaunay triangulation (DT) computational geometry, we proposed algorithm based on DT net this paper. It uses matching, and then develops to find reference minutiae pairs (RMPs). Using topological structure set, formed with as vertexes. From nets input set template select out certain which have similar structures RMPs for aligning, carried point pattern. The experiment conducted...

This paper presents an effective constrained multiple model particle filtering (CMMPF) for bearings-only maneuvering target tracking. In the proposed algorithm, process of tracking is factorized into two sub-problems: 1) motion estimation and model-conditioned state according to Rao-Blackwellised theorem 2) dynamic system modeled by switching models in a jump Markov framework. To estimate set, modified sequential importance resampling method used draw particles, which can be restricted...

In order to solve the drawback that navigation error of Sandia Inertial Terrain Aided Navigation (SITAN) algorithm diverges easily if initial System (INS) is big. The contour matching for optimization proposed. can be reduced and match position points formed. And then SITAN introduced optimal estimation obtained. makes full use characteristics INS has high precision heading small during a short period. distance function differences are constantly found between two contours Then precise...

Traditional pilot-based channel estimation techniques are LS-based, MMSE-based and DFT-based methods. The LS-based performance is worse than MMSE DFT due to the large noise impact. Although has superior performance, difficult implement because of matrix inversion during calculation making computational complexity large. In reality, methods usually used. algorithm lower that algorithm, better LS algorithm. However, traditional only eliminates outside cyclic prefix in impulse response...

Abstract Multi‐passive‐sensor systems are a common means for the target tracking and their bearings processing is prerequisite stable control nonlinear filtering. This study proposes mathematical methodology that based on incorporating deterministic unscented transition rules into stochastic sequential importance sampling frame makes use of soft spatiotemporal constraint comprise multiview epipolar geometry numerical regularization to solve correspondence problem. A prototype...

In order to solve the problem that iterated closest contour point(ICCP) algorithm diverges easily when initial INS error is large, terrain matching (TERCOM) firstly used reduce error, then ICCP obtain best position. Two difference as measurement of Kalman filter, corrected and optimal estimate obtained. The correlative analysis MSD only introduced in coarse stage, sliding window precise stage improve efficiency. Simulations are performed results show proposed combinational process more...

To track the bearings-only maneuvering target tracking accurately online, soft measurement constraints are implemented into Unscented Kalman filtering (UKF). deal with constraints, Lasso regularization is added as obstacle function. In doing this, sampled sigma points can be restricted feasible region. enhance sampling efficiency, global optimal solution acquired by a heuristic optimizer. smooth outliers, posterior distribution approximated Gaussian mixture consists of original and modified...

This letter proposes a novel spatiotemporal learning model via mixture importance Gaussian filtering (MIGF). In the MIGF, we explore causal mapping between target's true parameters and latest measurement, present an approach for elegantly combining deterministic Gaussian-Hermite integral with stochastic sampling method. formulation allows use of soft-constrained sparse regularization to reduce truncation error improve weight adaptivity. We also provide effective invariant updating rule learn...

Particle filtering (PF) schemes are a set of simulation-based techniques relying on proposal distributions which have crucial impact their performance. In this paper, we introduce novel constrained Extended Kalman particle filter (CEPF), the is used to generate distribution. The algorithm integrates nonlinear state constraint information and latest measurement simultaneously into dynamic system transition density. proposed selects particles with higher likelihood propagate next time ste by...