CS Paper Radar

Compressed Single-Tone Frequency Estimation With Unknown Complex Gain: Singular Rates and Global Identifiability

Armon Rasooli

eess.SP ·submitted 2026-08-31 11:30 UTC · v1
theory

Abstract

Fixed linear compression can preserve local Fisher information for a sinusoid yet destroy global frequency identification, while a dark response can make local estimation nonregular. We study both failures for a single complex tone observed through a fixed complex-linear sketch with unknown nonzero complex gain and pre-sketch white Gaussian noise. Near an isolated analytic dark frequency, we separate radial signal vanishing from optimized projective contact between the two signed frequency branches. When the gain magnitude is constrained to a fixed nondegenerate interval, finite contact is equivalent to local quotient identifiability and minimax consistency. The sharp mean-square-error rate is determined by the sum of the radial and contact orders; infinite contact produces exact local aliases. Globally, we formulate a worst-frequency efficientinformation objective on whitened row spaces. We solve it exactly for every even output rank, prove uniqueness and quantitative rigidity of the symmetric-edge projector, and solve the evenaperture co-rank-one case. The rank-two optimum is aliased, and a winding obstruction gives a positive lower bound on the information price of global identification. From three outputs onward, sketches that are globally identifying with an immersive projective response are open and dense and have zero price at the level of suprema; for every even rank of at least four, the exact optimizer has this property. Thus local information preservation, singular recoverability, and global identifiability obey distinct compression laws within one estimation model.

Closest papers

Nearest neighbours by meaning across the whole index.

Minimax-Optimal Robust Identification of Continuous-Time Systems: Handling Narrow-Band Disturbances in the Frequency Domain

eess.SY math.ST ·2026-09-20 ·Jonas Gillberg, Fredrik Gustafsson
similarity 0.55
Abstract · 17 pages, 13 figures

The high-frequency spectral roll-off of continuous-time ARMA (CARMA) models can magnify the effect of narrow-band disturbances when aliasing is weak, making standard maximum-likelihood Whittle estimation sensitive to affected ordinates. We show that a logarithmic transformation $r_k = \log ρ_k$ converts the Whittle scale problem into a Gumbel location problem, connecting robust spectral estimation to the classical minimax theory of Huber and Rieder. A piecewise centering correction---closed-form for small $b$, implicit closed-form for the practitioner range---preserves Fisher consistency for any clipping level without numerical optimisation. Clipping the Gumbel score symmetrically and transforming back yields a two-sided clipped Gumbel score (the standard Rieder--Hampel bounded-influence form) whose normalised influence curve is proved locally asymptotically minimax under shrinking gross-error contamination. The efficiency loss is quantified by a single scalar $K(b)$: at $b = 1.5$, only $16\%$ nominal asymptotic variance overhead. In the stated AR(2) Monte Carlo design, the sample-size trends are compatible with the asymptotic rate, and at $b=1.5$ the bias reductions are about $40\%$, $90\%$, and $96\%$ for $a_1$, $a_2$, and $λ$, respectively.

0.26

Fisher Information and the Minimum Mean-Square Error for Locating Closely Spaced Point Sources

cs.IT ·2026-10-23 ·Chuling Wen, Jian Lu, Qingtang Jiang, Charles K. Chui
theory similarity 0.53
Abstract

We determine the optimal average error of locating closely spaced sources from noisy superpositions of a known smooth response with unknown complex coefficients. Relative phases control how changes in one location can be offset by the other unknowns, so local Fisher information alone does not determine the attainable error. For every fixed number $K\ge2$ of sources with independent uniform phases, we prove matching Bayes risk bounds jointly in separation and noise, for all sufficiently small separations and all noise levels. Odd source counts have two risk regimes; even counts have an intermediate regime in which mean squared error is linear in the noise standard deviation. A projected least-squares estimator attains these orders without knowing the phases or the prior density. Bounded nonuniform phase laws preserve the orders under stated support conditions, whereas priors concentrated near poorly identified configurations can change the exponents. A common finite comparison yields both a global inverse bound and paths controlled by the original prior. These give sufficient conditions under which the risk has the order of the pointwise capped Cramér--Rao bound, and a lower-tail calculation explains the power and logarithmic regimes. We also give conditional guarantees when the cluster center, separation, and source count are unknown, with an additional low-noise requirement for count selection. A calibrated covariance factor model illustrates the same criterion beyond source localization.

0.26

Sharp Minimax Limits and Compatibility Spectra for Critical Near-DFT Index-Only Frequency Estimation

eess.SP ·2026-09-01 ·Armon Rasooli, Mohammad Sadegh Narimani
similarity 0.52
Abstract

An $M$-channel discrete Fourier transform (DFT) channelizer routes an on-grid sinusoid to a single output. When each frame reports only one energy-proportional channel index, signed sub-bin frequency estimation becomes nonregular: dark-channel probability is quadratic in the offset, whereas orientation enters cubically. We study $N$ independent labeled reports under a known uniform-replacement probability $ε_N$ and a single deterministic unitary shared by all frequencies and frames, constrained to routing defect $τ/N$. If $\sqrt{N}ε_N \to λ< \infty$, we establish an attained global-in-frequency minimax limit at the critical scales $N^{-1/4}$ for frequency offset, $N^{-1/2}$ for unitary perturbation and replacement, and $N^{-1}$ for routing defect. The effective unitary tangent is a symmetric complete-graph edge field modulo one centering nuisance, and its first variation is a radius-dependent weighted divergence. For $λ>0$, evaluation at $k$ distinct normalized offset magnitudes yields an exact Fourier-Cauchy nullity spectrum: $\lfloor(M-1)/2\rfloor$ evaluations are necessary and sufficient, for every choice of distinct magnitudes, to certify neutrality at all magnitudes. The terminal nullspace has a greatest-common-divisor dimension formula and positive-definite aggregate curvature. Consequently, exact DFT routing is uniquely minimax within the complete critical tangent class for $M=3$, whereas every positive critical defect budget strictly improves the minimax constant for $M\geq4$. The analysis also yields a smallest-prime curvature-visibility law and, for $M\geq5$, discontinuous compatibility geometry but a continuous minimax value at the zero critical replacement floor $λ=0$.

0.24

Unsupervised Instantaneous Phase and Frequency Tracking by Inverse Voice Synthesis

eess.AS cs.SD ·2026-10-02 ·Chin-Yun Yu, György Fazekas
similarity 0.51
Abstract · Submitted to ICASSP 2027

Knowledge-driven neural vocoders struggle to learn reliable fundamental frequency end-to-end, because spectral objectives provide weak supervision of periodic structure and lack phase information. We address this with a source-filter model whose alias-free additive source makes the instantaneous phase of the glottal cycle explicit; differentiating it yields the instantaneous frequency, and thus $F_0$, without an external tracker. Waveform error supervises only the deterministic harmonic path, while a spectral loss covers the full signal. On M4Singer and LM-SSD, the reconstruction is phase-aligned, reaching a signal-to-reconstruction-error ratio of 8.1 dB, while neural baselines remain negative. However, GOLF, given an external $F_0$, still reaches lower spectral distortion. On LM-SSD, the recovered $F_0$ attains the highest overall accuracy of any method tested, including supervised neural pitch trackers applied off the shelf, and the glottal closure instants come within 0.53 points of REAPER's identification rate, without any $F_0$ label.

0.50

Compact and Infinite-Order Error Analysis for Null-Space SVD Estimation

cs.LG math.ST ·2026-08-31 ·Xin Li, Jonathan Cohen, Rami Puzis
theory similarity 0.50
Abstract · 31 pages

We study null-space estimation from a noisy matrix. For a simple left null space, we first derive an exact compact expression for the error of the smallest left singular vector. We then give an all-order series for the SVD vector and projector, followed by compact and consistently truncated series forms for the fixed-realization empirical risk and conditional population generalization risk. The recursion extends to a multiple-dimensional null space by following the complete invariant subspace. The convergence radius is not inferred from an error plot: it is computed independently from the nearest complex exceptional point that joins a retained eigenvalue branch to its complement. A reduced-nullity experiment shows that moving this spectral boundary can increase the radius, although the improvement is not monotone in the retained nullity. For individually ordered null directions under Gaussian training with \(τ\geq m\), we prove that the Wishart splitting matrix \(W\) gives a strict second-order empirical ranking. Gaussian averaging equalizes the leading generalization risks at both small and very large noise, while a column-swap theorem proves strict expected generalization ranking for an isotropic signal subspace. For unequal spikes, an exact population-overlap criterion and a simultaneous \(99\%\) Monte Carlo confidence certificate explain the observed intermediate ranking. A sixth-order risk correction improves the lower-crossover estimate in the reported experiment. This equal--ranked--equal phenomenon is a finite-sample diagnostic related to spectral mixing, but its tolerance crossings, the exceptional-point radius, and the asymptotic BBP threshold are three distinct quantities.

0.38

A Mathematical Theory of Near-Field Super-Resolution

cs.IT math.NT ·2026-09-25 ·Sajad Daei, Gabor Fodor, Mikael Skoglund
similarity 0.50
Abstract

Finite-aperture near-field sensing leads to a super-resolution geometry fundamentally different from the translation-invariant Fourier setting. In the Fresnel regime, wavefront curvature introduces a range-dependent quadratic aperture phase, so distinguishability is governed by incomplete quadratic exponential sums of the form \( \sum_{n=0}^{N_r-1} a_n e^{i(ω_1 n+ω_2 n^2)} \), rather than by angular separation alone. We develop a deterministic recovery theory for sparse measures with ranges on a finite grid and continuous angles, and introduce a support-uniform quadratic-phase aperture criterion replacing classical minimum separation. Under this criterion, total-variation minimization exactly recovers every sparse measure in the admissible support class, uniformly over all nonzero complex amplitudes. The proof develops nonasymptotic support-uniform bounds for finite quadratic sums and combines them with a gauged Hermite dual certificate controlling interpolation, local curvature, and off-support leakage. We further construct a finite-harmonic Bessel-Vandermonde lift with explicit truncation error. In the far-field limit, the quadratic phase disappears and the theory reduces to Fourier-type angular super-resolution.

0.27

Spectra of Multivariate Vandermonde Matrices and Min-max Error Bounds for the High-dimensional Super-resolution Problem

cs.IT math.SPmath.NA ·2026-10-17 ·Nuha Diab, Dmitry Batenkov
theory similarity 0.49
Abstract · 38 pages, 10 figures

We study multivariate rectangular Vandermonde matrices $U_N$ whose rows are indexed by the grid $G_N=\{-N,\ldots,N\}^d, $ and whose columns correspond to irregularly spaced points on the unit $d$-dimensional torus. Motivated by the problem of super-resolution (recovery of spike trains from low frequency noisy measurements), we consider configurations in which the points form several clusters, with points within each cluster separated by at most $h$ and distinct clusters are well-separated. We show that the column subspaces associated with different clusters are nearly orthogonal, i.e. for $N \rightarrow \infty, Nh \ll 1$, the minimal principal angle between any such pair is at least $ {π\over 2} - \frac{C_1}{N} - C_2Nh - \frac{C_3}{\sqrt{N}}, $ for some constants $C_1,C_2,C_3$ independent of $N$ and $h$. This near-orthogonality enables the spectral analysis of such $U_N$ to be reduced to the analysis of each individual cluster. As a consequence, we derive sharp scaling for all singular values of $U_N$, revealing how the geometry of clustered points controls the conditioning of the high-dimensional super-resolution problems. Then we establish min-max error bounds for the recovery of clustered point sources, requiring only an upper bound on the largest cluster size and no knowledge of the geometry within the clusters. These bounds coincide with the optimal rates known in the one-dimensional setting. Furthermore, when the clustered points are in generic position (i.e. do not lie on a low-dimensional algebraic variety) our results indicate that super-resolution in high dimensions is more stable than in one dimension.

0.40

The Manifold-Index Model: provably and efficiently avoiding the statistical and computational curse of dimensionality

cs.LG stat.MLmath.ST ·2026-11-15 ·Yantao Wu, Mauro Maggioni
similarity 0.49
Abstract · 89 pages, 17 figures, 8 tables

We study vector-valued regression with predictors filling a tube around an unknown $m$-dimensional manifold $\mathcal{M}$ in $\mathbb{R}^D$. In the Manifold-Index Model, the mean response is $F=f\circΠ_{\mathcal{M}}$, where $Π_{\mathcal{M}}$ is the nearest-point projection onto $\mathcal{M}$ and $f:\mathcal{M}\to\mathbb{R}^p$ is a possibly non-injective link function. Multiscale ANisotropic lOcal Regression (MANOR) estimates tangent planes by multiscale PCA of predictors with similar responses, then refines them by pooling outer products of local slopes. The neighborhoods narrow tangentially while retaining complete normal fibers. We assume a Lipschitz link differential and bounded response noise. For $p\ge m$, under our geometric and sampling conditions, $n$ independent observations give mean tangent-space error of order $n^{-2/(m+4)}$ and integrated squared prediction error at the $m$-dimensional nonparametric rate $n^{-4/(m+4)}$, up to logarithmic factors. In a scaling regime with fixed intrinsic dimension, model and scale parameters, and ratio of manifold volume to the $m$-th power of its reach, the required sample size grows polynomially in $D+p$, and computation is polynomial in $n$, $D$ and $p$. Numerical experiments on curves with non-monotone links and surfaces with vector-valued responses compare observed convergence rates with the theoretical bounds; a Langevin dynamics example tests prediction of hitting probabilities under an approximate manifold-index model.

0.29