The low-rank matrix reconstruction (LRMR) approach is widely used in direction-of-arrival (DOA) estimation. As the rank norm penalty in an LRMR is NP-hard to compute, the nuclear norm (or the trace norm for a positive semidefinite matrix) has been often employed as a convex relaxation of the rank norm. However, solving a nuclear norm convex problem may lead to a suboptimal solution of the original rank norm problem. In this paper, we propose to apply a family of nonconvex penalties on the singular values of the covariance matrix as the sparsity metrics to approximate the rank norm. In particular, we formulate a nonconvex minimization problem and solve it by using a locally convergent iterative reweighted strategy in order to enhance the sparsity and resolution. The problem in each iteration is convex and hence can be solved by using the optimization toolbox. Convergence analysis shows that the new method is able to obtain a suboptimal solution. The connection between the proposed method and the sparse signal reconstruction is explored showing that our method can be regarded as a sparsity-based method with the number of sampling grids approaching infinity. Two feasible implementation algorithms that are based on solving a duality problem and deducing a closed-form solution of the simplified problem are also provided for the convex problem at each iteration to expedite the convergence. Extensive simulation studies are conducted to show the superiority of the proposed methods.

In this paper, a new indoor localization algorithm with received signal strength indicators (RSSI) fingerprints by a kernel-based learning technique is proposed. In the offline phase, after the training set pre-processing by the iterative self-organizing data analysis techniques algorithm (ISODATA), the measurement-label set is utilized for classification learning by the c-support vector classification approach. Moreover, each measurement-position training subset is chosen for regressing learning with hybrid kernel and cross-validation techniques. In the online phase, based on the classification result of the received RSSI measurement, the corresponding regression function is chosen for target position estimation. Compared with existing learning techniques, it can increase the training ability of a localization system dramatically. The field tests show that, without adding any extra computation complexity, the proposed algorithm can obtain more accurate position estimation than other existing learning-based localization approaches achieve.

This correspondence proposes a localization method for mixed far-field (FF) and near-field (NF) sources based on cross arrays where the two-dimensional (2-D) direction-of-arrival (DOA) and range estimation are considered. For 2-D DOA estimation, we first construct a range-free cumulant-based vector and employ the covariance matching criterion and Vandermonde decomposition theorem to retrieve the DOA information. We then build a sparse model with respect to the range parameter and solve an $\ell_1$ -norm minimization problem to obtain the range estimates. Our method enjoys the following advantages: 1) our method can locate mixed FF and NF sources; 2) the azimuth, elevation and range can be automatically paired; 3) array aperture loss can be avoided. Simulation results show the effectiveness of our method.

A new extreme learning machine (ELM) localization technique that uses received signal strength indicator fingerprints only is proposed for multifloor environments. This structured scheme forms multiple individual ELMs for the floors as well as for the geographically formed data clusters of each floor. Multifloor environments often have huge amount of training and online measurement data. To maximize efficiency, we develop a data preprocessing algorithm, aiming to: 1) efficiently extract out only the essential information from the vast amount of data sets and reduce the data dimension and 2) transform the floor-level data sets and positioning data sets of each floor into a proper structure that is suitable for the proposed ensemble ELM technique. The proposed solution is unique in that its offline phase exploits multiple individual ELMs for all floors to generate a set of floor-level classification functions with the preprocessed training data sets, and for each floor, it exploits multiple ELMs for the data clusters to generate a set of position regression functions. The online phase executes a coarse localization step to estimate the floor by using the floor-level classification functions and a refined step to estimate the position on the floor by using the position regression functions. The proposed algorithm and several existing algorithms are implemented to perform localization using the same measured datasets in a multistory building. For both floor estimation and localization on the floor, it outperforms existing schemes. For most cases, the performance gap is substantial.

Although fingerprint-based methods could achieve high location accuracy for device-free indoor localization, it requires cost-expensive massive labeling. In order to fully exploit the previously collected CSI fingerprints for various indoor localization tasks, this letter proposes a meta-learning approach for indoor localization, with which an excellent initialization solution for a new localization task is resorted and further training with a few labeled samples results in sound performance improvement over the state-of-the-art deep learning (DL) approach. By recognizing the unequal contribution of an individual training task to the target task, a novel weighted loss is introduced under the framework of meta-learning. Experiments show that the proposed meta-learning approach can gain improvement of 20% Root-Mean-Square-Error (RMSE) enhancement over the DL approach.