CS Paper Radar

Multi-Threshold Sampling: Local and Global Parameter Identifiability

Ao Qiu, Qingguo Xie

eess.SP physics.ins-det ·submitted 2026-10-20 05:26 UTC · v1
theory

Abstract

Multi-threshold (MT) sampling records threshold-crossing times for parameter estimation in applications including positron emission tomography, oil well logging, and photon-counting X-ray imaging. We study whether the distribution of recorded crossings uniquely determines the signal parameters and how many thresholds are required for this identifiability. For a fixed MT sampler, we show that the ranks of distribution derivatives with respect to the parameters cannot increase through crossing, timing, and selection. Under suitable regularity assumptions, we characterize local identifiability by a full-rank Jacobian of finitely many event probabilities defined on recorded crossings and establish a necessary and sufficient global identifiability criterion under additional rank and topological assumptions. For a nonhomogeneous Poisson signal model with four intensity parameters, a fixed known single-photon response, and deterministic MT sampling, we prove that four cumulative probabilities of the first recorded crossing time establish local and global identifiability under their respective conditions using a single sufficiently small positive threshold. This attains the minimum number of thresholds for the specified model and sampler. Numerical experiments illustrate these results and reveal weak parameter sensitivity and substantial finite-sample estimation errors when estimation uses these probabilities. The results establish parameter uniqueness while highlighting the role of statistical sensitivity in accurate estimation.

Closest papers

Nearest neighbours by meaning across the whole index.

Multi-Threshold Sampling: Signal Space, Sampling Operators, and First Recorded Crossing-Time Distributions

eess.SP physics.ins-det ·2026-09-10 ·Ao Qiu, Qingguo Xie
theory similarity 0.75
Abstract

Multi-threshold (MT) sampling records crossing times for parameter estimation and waveform reconstruction. We develop signal and observation spaces and stochastic crossing, timing, and selection operators with associated Markov kernels. Finite point measures accommodate variable crossing counts and preserve multiplicity after quantization. For nonhomogeneous Poisson photon arrivals with shared threshold noise, shared timing jitter before quantization, and identity selection, we derive exact first recorded crossing-time distributions. A Poisson series covers general admissible paths with a time-uniform truncation bound. For continuous paths, interval extrema give the exact cumulative distribution function (CDF) and a Fourier representation with contact corrections. A biexponential arrival-intensity and response model admits a two-dimensional first-crossing integral equation on arbitrary finite observation windows, with a unique bounded measurable solution and a Poisson-tail approximation bound. For continuous causal responses with finitely many analytic pieces and bounded first and second branch derivatives, we prove a conditional standard-deviation (SD) limit at first mean crossings with positive slope under relative threshold-noise and fine timing scales. For the biexponential model, its coefficient increases strictly on a parameter-uniform interval and, under a shape condition, up to the first mean peak. Scintillation experiments compare the calculated CDF and conditional SDs with independent waveform simulations on a window extending beyond the single-photon response peak.

0.41

A Nonuniform-Sampling Theory for Static Multi-Threshold Sampling

eess.SP cs.ITphysics.ins-detphysics.med-ph ·2026-08-27 ·Ao Qiu, Qingguo Xie
theory similarity 0.60
Abstract

Multi-Threshold (MT) sampling fixes voltage levels and records when a waveform crosses them. The resulting signal-generated event stream is governed not by a prescribed clock but by the nonuniform geometry of the realized crossing set. Within classical nonuniform sampling, we develop a retained-event MT theory for static, uniformly spaced thresholds. For a realized retained MT time set, Beurling's weak-limit theorem gives the exact necessary-and-sufficient condition for stable sampling of $PW_Ω$. For a fixed signal class, the same principle gives the a priori criterion: uniform stable MT sampling is equivalent to Bernstein uniqueness for every set in the class weak-limit hull. This criterion reduces to computable forms under structured priors. Periodic templates give finite fiber/rank tests, periodic-gap streams give a Shannon-type complete-interpolation boundary and a stable-sampling density rule, and finite-state gap priors reduce noncritical certification to a maximum-cycle-mean condition. Kadec-type local certificates, retained-density bounds, and finite-matrix tests provide practical sufficient conditions for finite active records, while conditional perturbation bounds quantify how jitter, front-end noise, and threshold error consume a clean sampling margin under event correspondence. The theory also identifies the unavoidable boundary of this model: unrestricted finite-energy retained MT records have zero full-line lower Beurling density and therefore cannot provide a universal prior-free sampling theorem for $PW_Ω$. Within retained-event point sampling, the MT recovery question is resolved precisely: exact recovery is governed by event-set geometry, computable guarantees require structural priors or finite-dimensional/tail assumptions, and the unrestricted static finite-energy problem is impossible.

0.26

One Photon, Many Worlds: Posteriors and Predictions with Single-Photon Cameras

eess.IV cs.CV ·2026-10-02 ·Haejoon Lee, Mohit Gupta, Vijayakumar Bhagavatula, Aswin C. Sankaranarayanan
similarity 0.53
Abstract · 18 pages, 18 figures

Single-photon avalanche diode (SPAD) cameras operate fundamentally differently from conventional cameras due to their photon-counting nature. Each frame produces a binary image: pixels report zero if no photons arrived during exposure, and one if one or more photons arrived. Reconstructing a scene or inferring its properties from a single binary frame is difficult because many different images could produce the same measurement; thus, the inverse problem is fundamentally one-to-many. As we gather more binary measurements, the inherent uncertainty associated with the inverse problem and any associated inference diminishes. With sufficient photon counts, photon noise becomes negligible relative to the signal mean, enabling near-deterministic scene recovery and inference. This work characterizes the transition from stochastic to near-deterministic scene understanding as photon budget increases, analyzing how the stochasticity in photon arrival affects downstream inference tasks. Technically, we develop a conditional generative framework based on a Hypergeometric frame-thinning process for accumulated binary SPAD measurements. Generative models capture the one-to-many nature of photon-starved inverse problems, enabling empirical characterization of how this ambiguity diminishes with increasing measurements and its impact on downstream tasks like character recognition, QR code decoding, and facial analysis.

0.20

Locally calibrated and mesh-free inference for spatial point distributions: closed-form null, contamination law, and detectability threshold

cs.IT stat.ME ·2026-09-15 ·Henock Mwanza Lubukayi, Mechack Kabanga Ntolo
theory similarity 0.52
Abstract · 10 pages, 5 figures

Local inference for spatial point distributions is dominated by Monte Carlo calibration. We develop an alternative based on the Tweedie--Miyasawa identities of empirical Bayes, which relate locally weighted moments of a point distribution under a Gaussian kernel to derivatives of its log-intensity in scale space. We first establish a rigidity theorem showing that the structure of these identities forces the Gaussian kernel. Under complete spatial randomness, we derive a closed-form null distribution for a bounded inter-scale contrast, yielding a calibrated simulation-free pointwise test. In experiments, the measured type I error is 0.070 at a nominal level of 0.05, with a calibration cost 199 times smaller than Monte Carlo for the same local statistic. We then derive a contamination law for structures of dimension m and width w embedded in a uniform background, together with an explicit detectability threshold. For a filament in three dimensions, the threshold is 16 pi. The resulting scale-resolved local dimension estimator has no free parameters. Applications to California seismicity, a trefoil knot, and 10,071 SDSS galaxies show that the method separates local structures across scales without a spatial mesh and reproduces published cosmic web fractions. All experiments are reproducible from a single public script.

0.29

Optimal Gating for Dead-Time-Constrained Single-Photon Detectors

eess.SP ·2026-10-12 ·Frederic J. N. Jorgensen, Steven G. Johnson
similarity 0.51
Abstract · 30 pages, 4 figures

We develop a Fisher-information framework for optimizing single-photon detector gating in periodic experiments, in contrast to previous more empirical or heuristic gating choices. After each photon detection, such a detector is insensitive for a dead time, so that an early background detection can obscure a later signal. Gating preserves availability for informative photons by selectively deactivating the detector. In our high-pile-up simulation of distance-attenuated lidar, our practical data-independent gating design reduces the worst-case asymptotic range variance by more than half relative to the free-running acquisition (which does not use any gating). More generally, we characterize achievable detector-active frequencies as a convex polytope and construct an adaptive rule that asymptotically attains an optimal "oracle" design without prior knowledge of the physical parameters. We also derive practical criteria that quantify the potential gains of gating and/or photon attenuation over unattenuated free-running acquisition.

0.26

Provably adaptive sampling with uniform and remasking discrete diffusion models

cs.LG cs.ITmath.STstat.ML ·2026-08-24 ·Daniil Dmitriev, Zhihan Huang, Yuting Wei
similarity 0.50
Abstract · 37 pages, 4 figures

Discrete diffusion models offer a promising alternative to autoregressive generation by enabling parallel updates, but their sampling efficiency can depend strongly on the choice of the forward process and the sampler. For the uniform forward process, existing lower bounds for the standard $τ$-leaping sampler scale linearly with the ambient dimension $d$, raising the question of whether this dependence is intrinsic to the forward process. We answer this question in the negative. We consider a first-order sampler based on the leave-one-out denoiser for uniform and remasking processes whose coordinate updates can be performed in parallel. In both cases, the sampler can correct denoising mistakes during the sampling process, which becomes necessary when many coordinates are updated together. Our main result establishes an adaptive sampling guarantee: up to logarithmic factors, $N = O(\mathrm{DTC}(X_0) / \varepsilon)$ discretization steps suffice to achieve sampling error $O(\varepsilon_{\mathrm{score}}+\varepsilon)$, where $\varepsilon_{\mathrm{score}}$ is the error in score estimation. Thus, the sampling complexity is governed by the intrinsic dependence structure of the target distribution, as measured by its dual total correlation $\mathrm{DTC}(X_0)$, rather than directly by the ambient dimension $d$. Our analysis proceeds through a Bayes-optimal auxiliary sampler that separates discretization error from score-estimation error. We also derive an exact information-theoretic representation of the discretization error in terms of the mutual information between different coordinates of the forward process at different times. This representation applies to general forward processes and, in the uniform and remasking cases, can be controlled by $\mathrm{DTC}(X_0)$. Numerical experiments on structured synthetic distributions illustrate the predicted dimension-adaptive behavior.

0.60

Estimation with Quantized Parameter Side-Information

cs.IT ·2026-10-08 ·Mathis Wetterwald, Ofer Shayevitz, Michèle Wigger
published similarity 0.50
Abstract · 5 pages, conference paper

This paper presents upper and lower bounds on the minimax risk of a parameter estimation problem where the estimator not only observes independent and identically distributed observations but also a quantized version of the parameter, where the quantization has a limitation on the number of levels, but can be optimized otherwise. Our upper and lower bounds have similar decay rates for large number of observation samples and quantization levels.

0.40

Junction Geometry Survives an Unidentifiable Blur

cs.CV physics.ins-det ·2026-09-30 ·Sohail, Sarkar, Shakuntala Baichoo
unusual similarity 0.49
Abstract · 37 pages, 13 figures, 11 tables

Thin filaments cross in microscopy, and dispersed spectra cross on X-ray detectors. At a junction, where these structures cross or branch, how many straight arms meet, and which way do they go? Unknown blur obscures the answer. At a balanced crossing, distinct Gaussian blurs can even produce the same observation law. We show that the arms can still be recovered. For separated rays of nonvanishing weight and an unknown centered Gaussian-mixture blur, a distance combining arm count, directions, relative weights and supporting lines is bounded by a constant times the Euclidean difference in cell probabilities on a sufficiently fine fixed grid. The bound is uniform over a bounded model class, including unidentifiable blurs. From $n$ positions or photon counts, the minimax error is of order $n^{-1/2}$. This rate is proved for a global estimator. In practice we fit rays and blur by constrained local likelihood, reducing the BIC count when two detectors agree on fewer arms in the microscopy studies. Across three studies with protocols fixed before evaluation, on 360 crossings and branches assembled from measured single-molecule localizations, the fit gets the arm count and every direction within $10^\circ$ 24 to 37 percentage points more often than a tuned fixed-scale Hessian ridge detector, averaged over 25 to 400 localizations. On acute angles with correct counts, mean absolute error is $3.5^\circ$ to $4.3^\circ$, versus $10.4^\circ$ to $15.5^\circ$. On Chandra spectra, 38 of 90 fits at 800 photons find all four arms within $10^\circ$; among these, median error is $0.023^\circ$.

0.29