Haoli Zhao

Global navigation satellite systems (GNSS) provide efficient positioning services for location-aware Internet of Things (IoT) devices. However, GNSS non-line-of-sight (NLOS) signals can result in severe positioning errors in urban canyon areas. Existing deep learning-based NLOS signal classification methods cannot appropriately model the spatiotemporal information of NLOS interference, resulting in limited accuracy across multiple locations. This study presents a spatiotemporal information-driven model that can capture environmental characteristics and signal temporal information simultaneously to improve NLOS classification accuracy across multiple locations. First, a visualization analysis of the signal distribution across multiple locations demonstrates the impact of environmental characteristics. In addition, the significance of both the spatial environmental features and the signal temporal features for NLOS classification is clarified by constructing a tree diagram of the dataset. Second, we propose an airspace attention mechanism module and a long short-term memory (LSTM)-based temporal feature extraction module to model both types of features, respectively. Third, the learnable sparse regularizer is utilized to reduce feature redundancy and thereby realize a sparse representation, which improves model generalization performance. Finally, the spatiotemporal information-driven cross-attention model is developed to perform NLOS classification, which uses a cross-attention fusion strategy to integrate the two modules. We use real-world datasets collected across multiple urban canyon locations to test our model. Experiments show that the proposed model can achieve 98% classification accuracy across multiple locations. Generalization performance in unknown environments can be improved over 7% compared to several state-of-the-art models.

The BeiDou Navigation Satellite System (BDS) provides real-time absolute location services to users around the world and plays a key role in the rapidly evolving field of autonomous driving. In complex urban environments, the positioning accuracy of BDS often suffers from large deviations due to non-line-of-sight (NLOS) signals. Deep learning (DL) methods have shown strong capabilities in detecting complex and variable NLOS signals. However, these methods still suffer from the following limitations. On the one hand, supervised learning methods require labeled samples for learning, which inevitably encounters the bottleneck of difficulty in constructing databases with a large number of labels. On the other hand, the collected data tend to have varying degrees of noise, leading to low accuracy and poor generalization performance of the detection model, especially when the environment around the receiver changes. In this article, we propose a novel deep neural architecture named convolutional denoising autoencoder network (CDAENet) to detect NLOS in urban forest environments. Specifically, we first design a denoising autoencoder based on unsupervised DL to reduce the long time series signal dimension and extract the deep features of the data. Meanwhile, denoising autoencoders improve the model’s robustness in identifying noisy data by introducing a certain amount of noise into the input data. Then, an MLP algorithm is used to identify the non-linearity of the BDS signal. Finally, the performance of the proposed CDAENet model is validated on a real urban forest dataset. The experimental results show that the satellite detection accuracy of our proposed algorithm is more than 95%, which is about an 8% improvement over existing machine-learning-based methods and about 3% improvement over deep-learning-based approaches.

In dynamic and complex environments, the positioning accuracy of global navigation satellite system (GNSS) will be seriously reduced. Deep reinforcement learning (DRL) has been found to give effective dynamic policy learning for complex GNSS positioning correction tasks. However, catastrophic interference in DRL models caused by the high correlation between successive positioning states, together with instability in gradient backpropagation in deep neural networks (DNNs), produces inaccurate DRL value approximation thereby degrades GNSS positioning performance. In this paper, we develop a dictionary learning-based reinforcement learning (RL) algorithm with the nonconvex log regularizer for GNSS positioning correction. To avoid DNN instability problems, a dictionary learning-structured RL model is proposed. It has a feed-forward learning architecture obviating the need for gradient backpropagation. The nonconvex log regularizer for dictionary learning reduces the correlation between states and thereby alleviates interference in RL. This provides sparse representations, which can more effectively capture features and produce representations with lower biases than convex regularizers. Furthermore, the nonconvex optimization is made efficient through a decomposition scheme that generates an explicit closed-form solution using the proximal operator. Finally, based on the proposed dictionary learning-structured RL model, a novel positioning correction method is developed to enhance GNSS positioning accuracy. The experimental results indicate that the proposed method outperforms state-of-the-art sparse coding-based RL methods in benchmark environments. Moreover, the proposed method effectively improves GNSS positioning accuracy relative to the Kalman filter method and the weighted least squares method.

Deep Reinforcement Learning (DRL) has been applied to satellite navigation and positioning applications. Its performance relies heavily on the function-approximation capability of Deep Neural Networks (DNNs). However, existing DRL models suffer from catastrophic interference, resulting in inaccurate function approximation. The sparse-coding-based DRL is an effective method to mitigating this interference, but existing methods involve two challenging issues: 1) the value function estimation network suffers from instability problems with gradient backpropagation, including gradient explosion and gradient vanishing, 2) existing methods are limited to using hand-crafted sparse regularizers that produce only static sparsity, which may be difficult to apply in various dynamic Reinforcement Learning (RL) environments. In this paper, we propose a novel Dictionary Learning (DL)-structured RL model with Adaptive-Sparsity Regularizer (ASR) that alleviates the catastrophic interference and enables accurate value function approximation, thereby improving the RL performance. To alleviate the interference and avoid the instability problems in RL, a feed-forward DL-structured RL model is constructed to predict the value function without the need for gradient backpropagation. To learn data-driven sparse representations with adaptive sparsity, we propose to use the learnable sparse regularizer ASR in the model, where the key hyperparameters of ASR can be trained to be adaptive to variable RL environments. To optimize the model efficiently, the model parameters are first pretrained in the pretraining stage, with only the value weights used for value function approximation needing to be fine-tuned for actual RL applications in the control training stage. Our comparative experiments in benchmark environments demonstrate that the proposed method can outperform existing state-of-the-art sparse-coding-based RL algorithms. In terms of accumulated rewards (used to measure the quality of the learned policy), the improvement was over 63% in Cart Pole environment and nearly 10% for Puddle World. Furthermore, the proposed algorithm can maintain its relatively high performance in the presence of noise up to 20 dB.

Deep reinforcement learning (DRL), which highly depends on the data representation, has shown its potential in many practical decision-making problems. However, the process of acquiring representations in DRL is easily affected by interference from models, and moreover leaves unnecessary parameters, leading to control performance reduction. In this article, we propose a double sparse DRL via multilayer sparse coding and nonconvex regularized pruning. To alleviate interference in DRL, we propose a multilayer sparse-coding-structural network to obtain deep sparse representation for control in reinforcement learning. Furthermore, we employ a nonconvex log regularizer to promote strong sparsity, efficiently removing the unnecessary weights with a regularizer-based pruning scheme. Hence, a double sparse DRL algorithm is developed, which can not only learn deep sparse representation to reduce the interference but also remove redundant weights while keeping the robust performance. The experimental results in five benchmark environments of the deep $q$ network (DQN) architecture demonstrate that the proposed method with deep sparse representations from the multilayer sparse-coding structure can outperform existing sparse-coding-based DRL in control, for example, completing Mountain Car with 140.81 steps, achieving near 10% reward increase from the single-layer sparse-coding DRL algorithm, and obtaining 286.08 scores in Catcher, which are over two times the rewards of the other algorithms. Moreover, the proposed algorithm can reduce over 80% parameters while keeping performance improvements from deep sparse representations.

Multiview dictionary learning (DL) is attracting attention in multiview clustering due to the efficient feature learning ability. However, most existing multiview DL algorithms are facing problems in fully utilizing consistent and complementary information simultaneously in the multiview data and learning the most precise representation for multiview clustering because of gaps between views. This article proposes an efficient multiview DL algorithm for multiview clustering, which uses the partially shared DL model with a flexible ratio of shared sparse coefficients to excavate both consistency and complementarity in the multiview data. In particular, a differentiable scale-invariant function is used as the sparsity regularizer, which considers the absolute sparsity of coefficients as the $\ell _{0}$ norm regularizer but is continuous and differentiable almost everywhere. The corresponding optimization problem is solved by the proximal splitting method with extrapolation technology; moreover, the proximal operator of the differentiable scale-invariant regularizer can be derived. The synthetic experiment results demonstrate that the proposed algorithm can recover the synthetic dictionary well with reasonable convergence time costs. Multiview clustering experiments include six real-world multiview datasets, and the performances show that the proposed algorithm is not sensitive to the regularizer parameter as the other algorithms. Furthermore, an appropriate coefficient sharing ratio can help to exploit consistent information while keeping complementary information from multiview data and thus enhance performances in multiview clustering. In addition, the convergence performances show that the proposed algorithm can obtain the best performances in multiview clustering among compared algorithms and can converge faster than compared multiview algorithms mostly.

Direct-optimization-based dictionary learning has attracted increasing attention for improving computational efficiency. However, the existing direct optimization scheme can only be applied to limited dictionary learning problems, and it remains an open problem to prove that the whole sequence obtained by the algorithm converges to a critical point of the objective function. In this article, we propose a novel direct-optimization-based dictionary learning algorithm using the minimax concave penalty (MCP) as a sparsity regularizer that can enforce strong sparsity and obtain accurate estimation. For solving the corresponding optimization problem, we first decompose the nonconvex MCP into two convex components. Then, we employ the difference of the convex functions algorithm and the nonconvex proximal-splitting algorithm to process the resulting subproblems. Thus, the direct optimization approach can be extended to a broader class of dictionary learning problems, even if the sparsity regularizer is nonconvex. In addition, the convergence guarantee for the proposed algorithm can be theoretically proven. Our numerical simulations demonstrate that the proposed algorithm has good convergence performances in different cases and robust dictionary-recovery capabilities. When applied to sparse approximations, the proposed approach can obtain sparser and less error estimation than the different sparsity regularizers in existing methods. In addition, the proposed algorithm has robustness in image denoising and key-frame extraction.

Convolutional transform learning (CTL), learning filters by minimizing the data fidelity loss function in an unsupervised way, is becoming very pervasive, resulting from keeping the best of both worlds: the benefit of unsupervised learning and the success of the convolutional neural network. There have been growing interests in developing efficient CTL algorithms. However, developing a convergent and accelerated CTL algorithm with accurate representations simultaneously with proper sparsity is an open problem. This article presents a new CTL framework with a $log$ regularizer that can not only obtain accurate representations but also yield strong sparsity. To efficiently address our nonconvex composite optimization, we propose to employ the proximal difference of the convex algorithm (PDCA) which relies on decomposing the nonconvex regularizer into the difference of two convex parts and then optimizes the convex subproblems. Furthermore, we introduce the extrapolation technology to accelerate the algorithm, leading to a fast and efficient CTL algorithm. In particular, we provide a rigorous convergence analysis for the proposed algorithm under the accelerated PDCA. The experimental results demonstrate that the proposed algorithm can converge more stably to desirable solutions with lower approximation error and simultaneously with stronger sparsity and, thus, learn filters efficiently. Meanwhile, the convergence speed is faster than the existing CTL algorithms.

Sparse coding, which aims at finding appropriate sparse representations of data with an overcomplete dictionary set, is a well-established signal processing methodology and has good efficiency in various areas. The varying sparse constraint can influence the performances of sparse coding algorithms greatly. However, commonly used sparse regularization may not be robust in high-coherence condition. In this paper, inspired from independently interpretable lasso (IILasso), which considers the coherence of sensing matrix columns in constraint to implement the strategy of selecting uncorrelated variables, we propose a new regularization by introducing $\ell _{p}$ norm $(0 < p < 1)$ into the regularization part of IILasso. The new regularization can efficiently enhance the performances in obtaining sparse and accurate coefficient. To solve the optimization problem with the new regularization, we propose to use the coordinate descent algorithm with weighted $\ell _{1}$ norm, named independently interpretable weighted lasso (IIWLasso), and the proximal operator, named independently interpretable iterative shrinkage thresholding algorithm (II-ISTA) and independently interpretable proximal operator for $\ell _{\frac {2}{3}}$ norm regularization (II2/3PO). We present synthetic data experiments and gene expression data experiments to validate the performance of our proposed algorithms. The experiment results show that all independently interpretable algorithms can perform better than their original ones in different coherence conditions. Among them, IIWLasso can obtain relatively best performance both in relative norm error and support error of synthetic data and misclassification error of tenfold cross-validating gene expression data.