Abstract:Consider the problem of detecting one of M i.i.d. Gaussian signals corrupted in white Gaussian noise. Conventionally, matched filters are used for detection. We first show that the outputs of the matched filter form a set of asymptotically optimal sufficient statistics in the sense of maximizing the error exponent of detecting the true signal. In practice, however, M may be large which motivates the design and analysis of a reduced set of N statistics which we term approximate sufficient statistics. Our construction of these statistics is based on a small set of filters that project the outputs of the matched filters onto a lower-dimensional vector using a sensing matrix. We consider a sequence of sensing matrices that has the desiderata of row orthonormality and low coherence. We analyze the performance of the resulting maximum likelihood (ML) detector, which leads to an achievable bound on the error exponent based on the approximate sufficient statistics; this bound recovers the original error exponent when N = M. We compare this to a bound that we obtain by analyzing a modified form of the Reduced Dimensionality Detector (RDD) proposed by Xie, Eldar, and Goldsmith [IEEE Trans. on Inform. Th., 59(6):3858-3874, 2013]. We show that by setting the sensing matrices to be column-normalized group Hadamard matrices, the exponents derived are ensemble-tight, i.e., our analysis is tight on the exponential scale given the sensing matrices and the decoding rule. Finally, we derive some properties of the exponents, showing, in particular, that they increase linearly in the compression ratio N/M.

On the MMSE Estimation of Norm of a Gaussian Vector under Additive White Gaussian Noise with Randomly Missing Input Entries

Estimation in Gaussian Noise: Properties of the Minimum Mean-Square Error

Estimation of Non-Gaussian Random Variables in Gaussian Noise: Properties of the MMSE.

On MMSE estimation from quantized observations in the nonasymptotic regime

Adaptive robust estimation in sparse vector model

Asymptotic Theory of Least Distances Estimate in Multivariate Linear Models

Diffusion Sparse Least-Mean Squares over Networks

Optimal spectral norm rates for noisy low-rank matrix completion

An MMSE Lower Bound via Poincaré Inequality

Precise Error Analysis of Regularized M-estimators in High-dimensions

On the Error Exponent of Approximate Sufficient Statistics for M-ary Hypothesis Testing

On the Equivalence of Maximum SNR and MMSE Estimation: Applications to Additive Non-Gaussian Channels and Quantized Observations

Analytic MMSE Bounds in Linear Dynamic Systems with Gaussian Mixture Noise Statistics

Linear and Nonlinear MMSE Estimation in One-Bit Quantized Systems under a Gaussian Mixture Prior

Fundamental limits of Non-Linear Low-Rank Matrix Estimation

Alternating Direction Method of Multipliers Based on $\ell_{2,0}$-norm for Multiple Measurement Vector Problem

Estimation of the L2-Norm and Testing in Sparse Linear Regression with Unknown Variance

Estimation with Norm Regularization

Estimation beyond Missing (Completely) at Random

Sparsity meets correlation in Gaussian sequence model

Non-Parametric Field Estimation using Randomly Deployed, Noisy, Binary Sensors