In this paper, we derive an online estimator for sparse parameter vectors which, unlike the LASSO approach, does not require the tuning of any hyperparameters. The algorithm is based on a covariance matching approach and is equivalent to a weighted version of the square-root LASSO. The computational complexity of the estimator is of the same order as that of the online versions of regularized least-squares (RLS) and LASSO. We provide a numerical comparison with feasible and infeasible implementations of the LASSO and RLS to illustrate the advantage of the proposed online hyperparameter-free estimator.
17 March 2017 in IEEE Transactions on Wireless Communications
Clock synchronization is ubiquitous in wireless systems for communication, sensing, and control. In this paper, we design a scalable system in which an indefinite number of passively receiving wireless units can synchronize to a single master clock at the level of discrete clock ticks. Accurate synchronization requires an estimate of the node positions to compensate the time-of-flight transmission delay in line-of-sight environments. If such information is available, the framework developed here takes position uncertainties into account. In the absence of such information, as in indoor scenarios, we propose an auxiliary localization mechanism. Furthermore, we derive the Cramer-Rao bounds for the system, which show that it enables synchronization accuracy at sub-nanosecond levels. Finally, we develop and evaluate an online estimation method, which is statistically efficient.
10 February 2015 in IEEE Signal Processing Magazine
The estimation of multiple parameters is a common task in signal processing. The Cramer-Rao bound (CRB) sets a statistical lower limit on the resulting errors when estimating parameters from a set of random observations. It can be understood as a fundamental measure of parameter uncertainty [1], [2]. As a general example, suppose denotes the vector of sought parameters and that the random observation model can be written as y = xi + w, (1) where xi is a function or signal parameterized by i and w is a zero-mean Gaussian noise vector. Then the CRB for i has the following notable properties: 1) For a fixed i, the CRB for i decreases as the dimension of y increases. 2) For a fixed y, if additional parameters i u are estimated, then the CRB for i increases as the dimension of i u increases. 3) If adding a set of observations yu requires estimating additional parameters, i u then the CRB for i decreases as the dimension of yu increases, provided the dimension of i u does not exceed that of yu [3]. This property implies both 1) and 2) above. 4) Among all possible distributions of w with a fixed covariance matrix, the CRB for i attains its maximum when w is Gaussian, i.e., the Gaussian scenario is the "worst case" for estimating θ [4]-[6].
6 June 2017 in IEEE Journal of Oceanic Engineering
In this paper, we review and compare the performance of two recently introduced hyperparameter-free sparse signal processing methods namely, the sparse iterative covariance-based estimation method and the sparse Bayesian learning-based relevance vector machine method, for direction-of-arrival (DOA) tracking of multiple signals using an array of sensors. The methods are presented to the readers, in a tutorial style for easy understanding. Hyperparameter-free sparsity-based methods are attractive in practice since tuning of regularization parameters (hyperparameters) is not necessary as they are automatically estimated from the data. The DOA tracking problem is formulated as a snapshot-by-snapshot estimation problem and the implementation of the methods are discussed in detail. A simulation study using a uniform-linear-array is carried out to evaluate the performance of the methods in terms of the root-mean-squared error of the DOA estimates and the probability of resolution with the goal of determining when one is to be preferred over the other. The algorithms are also applied on passive sonar data from the 1997 High-Frequency (HF$\mathbf {97}$) ocean acoustic experiment to demonstrate their usability in a real underwater scenario, as well as their robustness to the modeling assumptions made. We draw new conclusions about the main features of these methods that are important to the underwater acoustic practitioners.
20 November 2020 in IEEE Open Journal of Signal Processing
We consider a general statistical learning problem where an unknown fraction of the training data is corrupted. We develop a robust learning method that only requires specifying an upper bound on the corrupted data fraction. The method minimizes a risk function defined by a non-parametric distribution with unknown probability weights. We derive and analyse the optimal weights and show how they provide robustness against corrupted data. Furthermore, we give a computationally efficient coordinate descent algorithm to solve the risk minimization problem. We demonstrate the wide range applicability of the method, including regression, classification, unsupervised learning and classic parameter estimation, with state-of-the-art performance.
21 August 2017 in IEEE Transactions on Signal Processing
We show that the recently proposed (enhanced) principal-singular-vector utilization for modal analysis (PUMA) estimator for array processing [C. Qian, L. Huang, N. Sidiropoulos, and H. C. So, “Enhanced PUMA for direction-of-arrival estimation and its performance analysis,” IEEE Trans. Signal Process., vol. 64, no. 16, pp. 4127–4137, Aug. 2016], minimizes the same criterion function as the well-established method of direction estimation (MODE) estimator.