Searcharxiv⌕ Search

arXiv · 2611.21045

Nonlinear Joint State and Noise Estimation under Packet Dropouts and Additive Measurement Corruption

Abstract

This paper investigates joint state and noise covariance estimation for nonlinear systems subject to dual reliability impairments: random packet dropouts and additive measurement corruption. This scenario poses challenging nonconjugate inference, as clean and corrupted measurements share an unknown baseline covariance $\boldsymbol{R}_k$ with fixed known corruption covariance $\boldsymbol{E}_k$. Packet reception is observable, whereas the authenticity of received measurements remains latent and must be inferred. To resolve this intractable estimation problem, we cast joint posterior inference as a variational optimization problem and develop a conjugate variational inference adaptive Kalman filter with dual reliability (CVIAKF-DR). The proposed method retains analytical updates for the prediction covariance and two reliability probabilities, while adopting CVI to estimate the state and baseline measurement covariance. The covariance update combines the analytical contribution from clean measurements with Monte Carlo gradient estimates for corrupted measurements. Numerical results on a re-entry tracking benchmark demonstrate the superior accuracy of the proposed method in both state and noise covariance estimation. Remarkably, even built upon the linear Kalman filtering (KF) framework that is originally designed for linear systems, our method achieves more accurate estimation performance than dedicated nonlinear filtering algorithms based on the cubature Kalman filter (CKF).

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Peng Sun, Ruoyu Wang, Xue Luo. 2026-11-16. Nonlinear Joint State and Noise Estimation under Packet Dropouts and Additive Measurement Corruption. https://searcharxiv.com/abs/2611.21045

Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.

KEEP EXPLORING

Related papers

Deep Mixture Models: A Statistical Review with R

Deep mixture models extend finite mixture models through hierarchical latent factor structures with shared mixture parameters. This focused review examines deep Gaussian mixture models (DGMMs), robust deep mixture models (RDMMs), deep skew-$t$ mixture models (DStMMs), and mixed deep Gaussian mixture models (MDGMMs) within a common statistical framework. We compare their pathway distributions, measurement models, and latent architectures, highlighting the implications for likelihood evaluation and inference. For continuous measurements, we discuss Gaussian conditional calculations and the scale-mixture methods underlying Student-$t$ and skew-$t$ pathways. For mixed measurements, we consider response-specific likelihoods and numerical integration for discrete variables. Alongside the review, we introduce \texttt{MixedDeepMM}, an R package providing a unified interface for fitting and evaluating deep mixture models with Gaussian, Student-$t$, and skew-$t$ kernels, supporting continuous and mixed measurements and single- and two-head architectures. We illustrate its use on the UCI Wine and Cryotherapy datasets, reporting clustering measures with standard errors across repeated fits and visualising consensus partitions. For mixed data, Gower silhouette widths complement external clustering agreement measures. We conclude by discussing computational challenges, model selection, and uncertainty assessment.

stat.CO↗

Multilevel Filtering on Manifolds with Light Schrödinger Bridges

In this article we consider the filtering of partially observed diffusions when the observations, made at discrete times, are noiseless. Each observation then confines the hidden state to a manifold defined by the data, whose type, for instance a sphere or an affine subspace, may change from one observation time to the next. In practice, the filter is approximated by particle filters, which require the diffusion to be discretized in time. Multilevel Monte Carlo reduces the cost of this discretization only if the particle approximations at consecutive discretization levels stay close, which standard particle filters do not ensure. We develop a new particle filter based on light Schrödinger bridges, which works between changing manifolds and can be coupled across discretization levels. A key insight is that the previous state enters the light Schrödinger bridge only through a few features, so that the bridge only has to be fitted on the new manifold, regardless of the type of the previous one. We bound the bias of this particle filter and the gap between its approximations at consecutive levels, and show numerically, on spheres and on manifolds that change type, that this gap is much smaller than with standard multilevel particle filters, and that, with the cost of the fits included, a hybrid estimator that uses it at the fine levels reduces the cost at small tolerances.

stat.CO↗

A Hybrid NUTS-Gibbs Sampler with State Space Marginalization for Estimation of Dynamic Structural Equation Models with Binomial Outcomes

Dynamic structural equation models (DSEMs) are widely used for analyzing intensive longitudinal data (ILD). Although many ILD have categorical (Bernoulli or binomially distributed) responses, currently available algorithms for estimating DSEMs are limited to using the probit link and the Bernoulli distribution, i.e., the response can only be success or failure for a single trial and not the number of successes in multiple trials. They also scale poorly with increasing model complexity and sample size. Here, we present a hybrid sampler---alternating between one step of the No-U-Turn Sampler (NUTS) and one Gibbs step---which solves both of these problems: the Gibbs step naturally handles Pólya-Gamma distributed latent variables arising from binomially distributed responses with a logit link, and the NUTS step utilizes a Kalman filter to exactly marginalize over latent states, making models with dynamics defined in terms of latent traits more computationally feasible. We demonstrate the efficiency of the proposed sampler in simulation experiments, and also study conditions under which it fails. We finally illustrate its use in an example application of predicting panic attacks.

stat.CO↗