In sixth-generation and beyond, space-air-ground integrated networks (SAGINs) extend network connectivity to space, thereby enabling broader service coverage. This paper proposes a topology-aware SAGIN framework to address the integrated sensing, communication, and wireless power transfer (ISCPT) problem, leveraging the distinctive visibility of satellite-terrestrial and satellite-satellite users as well as their constructing in-between channel strengths. By modeling the topology of the SAGIN as a bipartite graph, we formulate the ISCPT problem as a multi-objective joint optimization problem with specified topological structures to reflect connection relationships of satellite-terrestrial and satellite-satellite users. The ISCPT problem is then reformulated and carefully decomposed as several mixed-integer linear programs (MILPs) by leveraging the network topology to individually optimize sensing, communication, and power transfer. To reduce the computational complexity of the proposed method, a greedy algorithm deal with generalized multi-assignment problem (GMAP) is developed. Simulation results demonstrate superior performance in communication and sensing, with a tolerable trade-off in wireless power transfer.

Terahertz (THz) frequency band has been widely studied and is recognized as a promising candidate for centimeter-level localization. However, the limited coverage of THz networks may result in localization failures, while a heterogeneous deployment of millimeter-wave (mmWave) and THz radio units (RUs) offers a viable solution to mitigate this issue. This paper presents a theoretical framework for evaluating the performance limits of localization systems in mmWave and THz heterogeneous networks. In this architecture, the mmWave RUs serve as macro base stations (BSs), while the THz RUs function as micro BSs distributed around each mmWave RU. By leveraging the standard tools of stochastic geometry to model the spatial distributions of the RUs and ambient obstacles, the localizability of a target is computed to evaluate the probability of achieving sufficient signal-to-interference-plus-noise ratio for localization in both line-of-sight (LoS) and non-line-of-sight (NLoS) conditions. Furthermore, the Cram é r-Rao lower bounds in both LoS and NLoS scenarios are analytically derived to characterize the overall positioning performance. Numerical results demonstrate that the hybrid deployment strategy significantly improves both the network coverage and localization accuracy compared to mmWave-only and THz-only networks.

Fingerprint-based indoor localization is often labor-intensive due to the need for dense grids and repeated measurements across time and space. Maintaining high localization accuracy with extremely sparse fingerprints remains a persistent challenge. Existing benchmark methods primarily rely on the measured fingerprints, while neglecting valuable spatial and environmental characteristics. To address this issue, we propose a systematic integration of an Attentional Graph Neural Network (AGNN) model, capable of learning spatial adjacency relationships and aggregating information from neighboring fingerprints, and a meta-learning framework that utilizes datasets with similar environmental characteristics to enhance model training. To minimize the labor required for fingerprint collection, we introduce two novel data augmentation strategies: 1) unlabeled fingerprint augmentation using moving platforms, which enables the semi-supervised AGNN model to incorporate information from unlabeled fingerprints, and 2) synthetic labeled fingerprint augmentation through environmental digital twins, which enhances the meta-learning framework through a practical distribution alignment, which can minimize the feature discrepancy between synthetic and real-world fingerprints effectively. By integrating these novel modules, we propose the Attentional Graph Meta-Learning (AGML) model. This novel model combines the strengths of the AGNN model and the meta-learning framework to address the challenges posed by extremely sparse fingerprints. To validate our approach, we collected multiple datasets from both consumer-grade WiFi devices and professional equipment across diverse environments. These datasets can also serve as a valuable resource for benchmarking fingerprint-based indoor localization methods. Extensive experiments conducted on both synthetic and real-world datasets demonstrate that the AGML model-based localization method consistently outperforms all baseline methods using sparse fingerprints across all evaluated metrics.

The estimation of underwater sound velocity distribution serves as a critical basis for facilitating effective underwater communication and precise positioning, given that variations in sound velocity influence the path of signal transmission. Conventional techniques for the direct measurement of sound velocity, as well as methods that involve the inversion of sound velocity utilizing acoustic field data, necessitate on-site data collection. This requirement not only places high demands on device deployment but also presents challenges in achieving real-time estimation of sound velocity distribution. In order to construct a real-time sound velocity field and eliminate the need for underwater on-site data measurement operations, we propose a self-attention embedded multimodal data fusion convolutional neural network (SA-MDF-CNN) for real-time underwater sound speed profile (SSP) estimation. The proposed model seeks to elucidate the inherent relationship between remote sensing sea surface temperature (SST) data, the primary component characteristics of historical SSPs, and their spatial coordinates. This is achieved by employing CNNs and attention mechanisms to extract local and global correlations from the input data, respectively. The ultimate objective is to facilitate a rapid and precise estimation of sound velocity distribution within a specified task area. The comparative analysis demonstrates that the proposed approach achieves superior performance in terms of both accuracy and stability, exhibiting reduced error rates and enhanced resistance to disturbances when benchmarked against existing advanced techniques.

Received signal strength (RSS) has been extensively studied for localization purposes, and the distributed nature of cell-free massive multiple-input multiple-output (CF-mMIMO) systems offers a new synergistic avenue for achieving high-precision localization. In this work, Open RAN and software-defined radio are used to realize the central and distributed units of a CF-mMIMO system to acquire the RSS measurements. By analyzing the experimental data, it is revealed that the RSS measured from the first-order reflection path can yield a sufficiently high signal-to-noise ratio for localization, enabling localization even without line-of-sight (LoS) paths. Inspired by this finding, a hybrid localization scheme, that can attain the best accuracy benchmarked by Cramér-Rao lower bound, is proposed to estimate the target position using both LoS and first-order non-line-of-sight RSS measurements. Furthermore, a theoretical framework is established to assess the fundamental limits of RSS-based localization in CF-mMIMO systems, offering a principled guideline for system designers to deploy and design localization systems in real-world scenarios.

Radio map estimation (RME) is crucial for effective planning and optimization of wireless networks. Traditional approaches such as interpolation excel at capturing local smoothness in densely populated data but struggle with sparse or irregular data. Conversely, matrix completion (MC) approaches utilize global structures but require huge number of samples and may produce non-smooth estimates. To integrate these strengths, we propose a convex optimization approach for RME (IIMC-RME) that merges interpolation with MC. This approach formulates the RME task as a low-rank MC problem constrained by interpolated results. Additionally, we have developed a convergent algorithm utilizing the alternating direction method of multipliers (ADMM) to efficiently solve the IIMC-RME problem. Experimental evaluations on both synthetic and real-world datasets have shown that IIMC-RME surpasses existing approaches, thereby achieving superior accuracy in RME.

Representing variable-length and continuous-indexed signals through a linear combination of basis functions poses a fundamental challenge in science and engineering. Current approaches resort to preprocessing steps, such as interpolation and extrapolation, to handle irregular and off-grid measurements, which compromise the physical nature of signals and degrade the representation performance. To address this challenge, rather than utilizing discrete vectors, we introduce a Bayesian functional representation model that capitalizes on the continuous nature and rich expressiveness of Gaussian processes to facilitate interpretable and effective basis function learning. Moreover, an analytical and efficient algorithm based on the variational inference framework is developed. Experimental results using real-life datasets demonstrate the superior performance of our proposed method.

In modern communication systems, to conserve transmission energy, the collected data are often max-aggregated. This aggregation involves observing only the features with relatively larger values in each observed sample. Recovering the joint distribution from such systematically missing data is of great interest for numerous downstream applications, including network optimization and user localization. This paper is motivated by the aforementioned industrial problem and investigates the recoverability and estimation of the joint distribution in the bivariate case. We consider various parametric model assumptions (uniform, Gaussian, mixture of Gaussians) and derive corresponding loss functions based on the maximum likelihood (ML) principle. The effectiveness of the proposed method is demonstrated through simulations. We also discuss its potential application in future wireless communication networks.

The efficiency of Bayesian optimization (BO) relies heavily on the choice of the Gaussian process (GP) kernel, which plays a central role in balancing exploration and exploitation under limited evaluation budgets. Traditional BO methods often rely on fixed or heuristic kernel selection strategies, which can result in slow convergence or suboptimal solutions when the chosen kernel is poorly suited to the underlying objective function. To address this limitation, we propose a freshly-baked Context-Aware Kernel Evolution (CAKE) to enhance BO with large language models (LLMs). Concretely, CAKE leverages LLMs as the crossover and mutation operators to adaptively generate and refine GP kernels based on the observed data throughout the optimization process. To maximize the power of CAKE, we further propose BIC-Acquisition Kernel Ranking (BAKER) to select the most effective kernel through balancing the model fit measured by the Bayesian information criterion (BIC) with the expected improvement at each iteration of BO. Extensive experiments demonstrate that our fresh CAKE-based BO method consistently outperforms established baselines across a range of real-world tasks, including hyperparameter optimization, controller tuning, and photonic chip design. Our code is publicly available at https://github.com/richardcsuwandi/cake.

Accurate prediction of mmWave time-varying channels is essential for mitigating the issue ofchannel agingin highly dynamic scenarios. Existing channel prediction methods have limitations: classical model-based methods often struggle to track highly nonlinear channel dynamics due to limited expert knowledge, while emerging data-driven methods typically require substantial labeled data for effective training and often lack interpretability. To address these issues, this paper proposes a novel hybrid method that integrates a data-driven neural network into a conventional model-based workflow based on a state-space model (SSM), implicitly tracking complex channel dynamics from data without requiring precise expert knowledge. Additionally, a novel unsupervised learning strategy is developed to train the embedded neural network solely with unlabeled data. Theoretical analyses and ablation studies are conducted to interpret the enhanced benefits gained from the hybrid integration. Numerical simulations based on the 3GPP mmWave channel model corroborate the superior prediction accuracy of the proposed method, compared to state-of-the-art methods that are either purely model-based or data-driven. Furthermore, extensive experiments validate its robustness against various challenging factors, including among others severe channel variations.

Gaussian processes (GPs) stand as crucial tools in machine learning and signal processing, with their effectiveness hinging on kernel design and hyperparameter optimization. This article presents a novel GP linear multiple kernel (LMK) and a generic sparsity-aware distributed learning framework to optimize the hyperparameters. The newly proposed grid spectral mixture product (GSMP) kernel is tailored for multidimensional data, effectively reducing the number of hyperparameters while maintaining good approximation capability. We further demonstrate that the associated hyperparameter optimization of this kernel yields sparse solutions. To exploit the inherent sparsity of the solutions, we introduce the sparse linear multiple kernel learning (SLIM-KL) framework. The framework incorporates a quantized alternating direction method of multipliers (ADMMs) scheme for collaborative learning among multiple agents, where the local optimization problem is solved using a distributed successive convex approximation (DSCA) algorithm. SLIM-KL effectively manages large-scale hyperparameter optimization for the proposed kernel, simultaneously ensuring data privacy and minimizing communication costs. The theoretical analysis establishes convergence guarantees for the learning framework, while experiments on diverse datasets demonstrate the superior prediction performance and efficiency of our proposed methods.

Building agents with adaptive behavior in cooperative tasks stands as a paramount goal in the realm of multi-agent systems. Current approaches to developing cooperative agents rely primarily on learning-based methods, whose policy generalization depends heavily on the diversity of teammates they interact with during the training phase. Such reliance, however, constrains the agents' capacity for strategic adaptation when cooperating with unfamiliar teammates, which becomes a significant challenge in zero-shot coordination scenarios. To address this challenge, we propose ProAgent, a novel framework that harnesses large language models (LLMs) to create proactive agents capable of dynamically adapting their behavior to enhance cooperation with teammates. ProAgent can analyze the present state, and infer the intentions of teammates from observations. It then updates its beliefs in alignment with the teammates' subsequent actual behaviors. Moreover, ProAgent exhibits a high degree of modularity and interpretability, making it easily integrated into various of coordination scenarios. Experimental evaluations conducted within the Overcooked-AI environment unveil the remarkable performance superiority of ProAgent, outperforming five methods based on self-play and population-based training when cooperating with AI agents. Furthermore, in partnered with human proxy models, its performance exhibits an average improvement exceeding 10% compared to the current state-of-the-art method. For more information about our project, please visit https://pku-proagent.github.io.

Received signal strength (RSS) measurements can be converted to the distance estimates between the emission source and the sensors to construct a system of linear equations, thereby allowing for the use of the weighted linear least squares (WLLS) estimators for location estimation. However, estimating the squared distances from the RSS measurements governed by the log-normal shadowing effect presents a major challenge in such approaches. In this paper, we propose a linear minimum mean square error (LMMSE) estimator of the squared distance between the emission source and the sensor first. Then a LMMSE-aided WLLS (LMMSE-WLLS) location estimator and its unbiased counterpart are presented for source localization. Furthermore, their estimation performance are analyzed in terms of mean square error (MSE) and covariance. It is found that the proposed LMMSE-aided WLLS location estimators have better estimation performance than existing WLLS estimators. Numerical examples also demonstrate the performance superiority of the proposed location estimators for source localization.

Deep state-space models (DSSMs) have gained popularity in recent years due to their potent modeling capacity for dynamic systems. However, existing DSSM works are limited to single-task modeling, which requires retraining with historical task data upon revisiting a forepassed task. To address this limitation, we propose continual learning DSSMs (CLDSSMs), which are capable of adapting to evolving tasks without catastrophic forgetting. Our proposed CLDSSMs integrate mainstream regularization-based continual learning (CL) methods, ensuring efficient updates with constant computational and memory costs for modeling multiple dynamic systems. We also conduct a comprehensive cost analysis of each CL method applied to the respective CLDSSMs, and demonstrate the efficacy of CLDSSMs through experiments on real-world datasets. The results corroborate that while various competing CL methods exhibit different merits, the proposed CLDSSMs consistently outperform traditional DSSMs in terms of effectively addressing catastrophic forgetting, enabling swift and accurate parameter transfer to new tasks.

The problem of joint direction-of-arrival estimation and distorted sensor detection has received a lot of attention in recent decades. Most state-of-the-art work formulated such a problem via low-rank and row-sparse decomposition, where the low-rank and row-sparse components were treated in an isolated manner. Such a formulation results in a performance loss. Differently, in this paper, we entangle the low-rank and row-sparse components by exploring their inherent connection. Furthermore, we take into account the maximal distortion level of the sensors. An alternating optimization scheme is proposed to solve the low-rank component and the sparse component, where a closed-form solution is derived for the low-rank component and a quadratic programming is developed for the sparse component. Numerical results exhibit the effectiveness and superiority of the proposed method.

The Gaussian process state-space model (GPSSM) has attracted extensive attention for modeling complex nonlinear dynamical systems. However, the existing GPSSM employs separate Gaussian processes (GPs) for each latent state dimension, leading to escalating computational complexity and parameter proliferation, thus posing challenges for modeling dynamical systems with high-dimensional latent states. To surmount this obstacle, we propose to integrate the efficient transformed Gaussian process (ETGP) into the GPSSM, which involves pushing a shared GP through multiple normalizing flows to efficiently model the transition function in high-dimensional latent state space. Additionally, we develop a corresponding variational inference algorithm that surpasses existing methods in terms of parameter count and computational complexity. Experimental results on diverse synthetic and real-world datasets corroborate the efficiency of the proposed method, while also demonstrating its ability to achieve similar inference performance compared to existing methods. Code is available at https://github.com/zhidilin/gpssmProj.

Existing localization approaches utilizing environment-specific channel state information (CSI) excel under specific environment but struggle to generalize across varied environments. This challenge becomes even more pronounced when confronted with limited training data. To address these issues, we present the Bayes-Optimal Meta-Learning for Localization (BOML-Loc) framework, inspired by the PAC-Optimal Hyper-Posterior (PACOH) algorithm. Improving on our earlier MetaLoc [1], BOML-Loc employs a Bayesian approach, reducing the need for extensive training, lowering overfitting risk, and offering per-test-point uncertainty estimation. Even with very limited training tasks, BOML-Loc guarantees robust localization and impressive generalization. In both LOS and NLOS environments with site-surveyed data, BOML-Loc surpasses existing models, demonstrating enhanced localization accuracy, generalization abilities, and reduced overfitting in new and previously unseen environments.

Integrating information while recognizing dependence from multiple data sources and enhancing the predictive performance of the multi-output regression are challenging tasks. Multioutput Gaussian Process (MOGP) methods offer outstanding solutions with tractable predictions and uncertainty quantification. However, their practical applications are hindered by high computational complexity and storage demand. Additionally, there exist model mismatches in existing MOGP models when dealing with non-Gaussian data. To improve the model representation ability in terms of flexibility, optimality, and scalability, this paper introduces a novel multi-output regression framework, termed Graphical MOGP (GMOGP), which is empowered by: (i) Generating flexible Gaussian process priors consolidated from dentified parents, (ii) providing dependent processes with attention-based graphical representations, and (iii) achieving Pareto optimal solutions of kernel hyperparameters via a distributed learning framework. Numerical results confirm that the proposed GMOGP significantly outperforms state-of-the-art MOGP alternatives in predictive performance, as well as in time and memory efficiency, across various synthetic and real datasets.

Gaussian process latent variable models (GPLVMs) are a versatile family of unsupervised learning models commonly used for dimensionality reduction. However, common challenges in modeling data with GPLVMs include inadequate kernel flexibility and improper selection of the projection noise, leading to a type of model collapse characterized by vague latent representations that do not reflect the underlying data structure. This paper addresses these issues by, first, theoretically examining the impact of projection variance on model collapse through the lens of a linear GPLVM. Second, we tackle model collapse due to inadequate kernel flexibility by integrating the spectral mixture (SM) kernel and a differentiable random Fourier feature (RFF) kernel approximation, which ensures computational scalability and efficiency through off-the-shelf automatic differentiation tools for learning the kernel hyperparameters, projection variance, and latent representations within the variational inference framework. The proposed GPLVM, named advisedRFLVM, is evaluated across diverse datasets and consistently outperforms various salient competing models, including state-of-the-art variational autoencoders (VAEs) and other GPLVM variants, in terms of informative latent representations and missing data imputation.

The low Earth orbit (LEO) satellite edge computing paradigm provides remote sites with flexible, reliable, and scalable edge computing capabilities. Characterized by the orbital motion patterns and harsh space environments, the LEO satellite edge computing faces unique security challenges in terms of the secure collaboration of multiple satellites and the intellectual property protection of models. Under the unique space environment and security demands, we propose a secure satellite edge computing framework in this paper. By taking a remote electricity line outage identification use case as an example, our framework first achieves the secure delegation of the line outage identification task among multiple satellites, which is realized through a secure query$(\mathsf {SQuery})$scheme to check the availability of the target time slot. Meanwhile, we also design a SHE-enabled secure inner-product encryption ($\mathsf {SSIPE}$) protocol, to achieve the secure multinomial logistic regression (MLR) based line outage identification on-orbit. To reduce the complexity brought by the computationally intensive homomorphic multiplication between two ciphertexts, we further grasp the idea and design a “divide-and-conquer” based secure query ($\mathsf {DSQuery}$) scheme, which converts this homomorphic multiplication operation between ciphertexts into the homomorphic addition operation. As far as we know, this is the first scheme investigating the secure task delegation among different satellites on-orbit. Besides, detailed security analyses are performed to demonstrate the security properties of confidentiality and authentication. In performance evaluations, we test and compare the computational and communication overhead of our scheme and other straightforward schemes. Simulation results show that the$\mathsf {DSQuery}$scheme greatly reduces the computational cost, which saves the stringent on-orbit computation resources of LEO satellites.

Variational autoencoders (VAE) are one of the most prominent deep generative models for learning the underlying statistical distribution of high-dimensional data. However, training VAEs suffers from a severe issue called posterior collapse; that is, the learned posterior distribution collapses to the assumed/pre-selected prior distribution. This issue limits the capacity of the learned posterior distribution to convey data information. Previous work has proposed a heuristic training scheme to mitigate this issue, in which the core idea is to train the encoder and the decoder in an alternating fashion. However, there is still no theoretical interpretation of this scheme, and this paper, for the first time, fills in this gap by inspecting the previous scheme under the lens of the expectation maximization (EM) framework. Under this framework, we propose a novel EM-type training algorithm that gives a controllable optimization process and it allows for further extensions, e.g., employing implicit distribution models. Experimental results have corroborated the superior performance of the proposed EM-type VAE training algorithm in terms of various metrics.

Gaussian process state-space model (GPSSM) is a fully probabilistic state-space model that has attracted much attention over the past decade. However, the outputs of the transition function in the existing GPSSMs are assumed to be independent, meaning that the GPSSMs cannot exploit the inductive biases between different outputs and lose certain model capacities. To address this issue, this paper proposes an output-dependent and more realistic GPSSM by utilizing the well-known, simple yet practical linear model of coregionalization (LMC) framework to represent the output dependency. To jointly learn the output-dependent GPSSM and infer the latent states, we propose a variational sparse GP-based learning method that only gently increases the computational complexity. Experiments on both synthetic and real datasets demonstrate the superiority of the output-dependent GPSSM in terms of learning and inference performance.

Existing localization methods that intensively leverage the environment-specific received signal strength (RSS) or channel state information (CSI) of wireless signals are rather accurate in certain environments. However, these methods, whether based on pure statistical signal processing or data-driven approaches, often struggle to generalize to new environments, which results in considerable time and effort being wasted. To address this challenge, we propose MetaLoc, which is the first fingerprinting-based localization framework that leverages the Model-Agnostic Meta-Learning (MAML). Specifically, built on a deep neural network with strong representation capabilities, MetaLoc is trained on historical data sourced from well-calibrated environments, employing a two-loop optimization mechanism to obtain the meta-parameters. These meta-parameters act as the initialization for quick adaptation in new environments, reducing the need for much human effort. The framework introduces two paradigms for the optimization of meta-parameters: a centralized paradigm that simplifies the process by sharing data from all historical environments, and a distributed paradigm that maintains data privacy by training meta-parameters for each specific environment separately. Furthermore, the advanced distributed paradigm modifies the vanilla MAML loss function to ensure that the reduction of loss occurs in a consistent direction across various training domains, thus facilitating faster convergence during training. Our experiments on both synthetic and real datasets demonstrate that MetaLoc outperforms baseline methods in terms of localization accuracy, robustness, and cost-effectiveness. The code and datasets used in this study are publicly available at:https://github.com/WU-Dongze/MetaLoc.

Multitask Gaussian process (MTGP) is powerful for joint learning of multiple tasks with complicated correlation patterns. However, due to the assembling of additive independent latent functions (LFs), all current MTGPs including the salient linear model of coregionalization (LMC) and convolution frameworks cannot effectively represent and learn the hierarchical latent interactions between its LFs. In this paper, we further investigate the interactions in LMC of MTGP and then propose a novel kernel representation of the hierarchical interactions, which ameliorates both the expressiveness and the interpretability of MTGP. Specifically, we express the interaction as a product of function interaction (FI) and coefficient interaction. The FI is modeled by using cross convolution of LFs. The coefficient interaction between the LMCs is described as a free-form coupling coregionalization term. We validate that considering the interactions can promote knowledge transferring in MTGP and compare our approach with some state-of-the-art MTGPs on both synthetic-and real-world datasets.

Localizing the root cause of network faults is crucial to network operation and maintenance (O&M). Significant operational expenses will be saved if the root cause can be identified agilely and accurately. However, this is challenging for human beings due to the complicated wireless environments and network architectures. Resorting to data analysis and machine learning is promising but remains difficult due to various practical issues, such as the lack of well-labeled samples, hybrid fault behaviors, missing data, and so on. In this paper, we introduce a novel real-world dataset for wireless communication network fault diagnosis. The goal is to infer the root cause timely when we observe certain symptoms in a network. Several baseline methods are provided.

The existing indoor fingerprinting methods based on received signal strength (RSS) are rather accurate after intensive offline calibration for a specific scenario, but the well-calibrated localization model (can be a pure statistical one or a data-driven one) will present poor generalization ability in a new scenario, which results in big loss in knowledge and human effort. To break the scenario-specific localization bottleneck, we propose a new-fashioned data-driven fingerprinting method for localization based on meta-learning, named by MetaLoc, that can adapt itself rapidly to a new, possibly unseen, scenario with very little calibration work. Specifically, the underlying localization model is taken to be a deep neural network (NN), and we train an optimal set of group-specific meta-parameters by leveraging historical data collected from diverse well-calibrated indoor scenarios and the maximum mean discrepancy criterion. Simulation results confirm that the meta-parameters obtained for MetaLoc achieves very rapid adaptation to new scenarios, competitive localization accuracy, and high resistance to significantly reduced reference points (RPs), saving a lot of calibration effort.

Nowadays, the vehicle to grid (V2G) technology enables bi-directional energy interactions between the power grid and the battery of an electric car. However, there still exist some issues in terms of security and privacy preservation. In this paper, we propose an efficient and privacy-preserving scheme to achieve the two-way electricity trading between vehicles and the power grid, by exploiting a dynamic threshold public-key encryption algorithm. Our proposed scheme mainly consists of two phases: The first phase includes the secure aggregation of the vehicles’ electricity requests and the recovery of the aggregation result; if the power grid can satisfy the aggregated electricity request, the vehicles execute in the second phase electricity trading. Meanwhile, the proposed scheme can adapt to a varying threshold of participating users, which enables the dynamic fluctuation of users. Extensive security analysis demonstrate the security properties of the proposed scheme in terms of privacy preservation and authentication. Performance evaluations are conducted to show the efficiency of our proposed scheme, and simulation results show that our scheme greatly reduces the introduced communication and computation overheads.

Graph neural networks (GNNs) are popular to use for classifying structured data in the context of machine learning. But surprisingly, they are rarely applied to regression problems. In this work, we adopt GNN for a classic but challenging nonlinear regression problem, namely the network localization. Our main findings are in order. First, GNN is potentially the best solution to large-scale network localization in terms of accuracy, robustness and computational time. Second, proper thresholding of the communication range is essential to its superior performance. Simulation results corroborate that the proposed GNN based method outperforms all state-of-the-art benchmarks by far. Such inspiring results are theoretically justified in terms of data aggregation, non-line-of-sight (NLOS) noise removal and low-pass filtering effect, all affected by the threshold for neighbor selection. Code is available at https://github.com/Yanzongzi/GNN-For-localization.

Federated learning-based navigation has received much attention in vehicular IoT. The intention is to employ a big number of end-users for data collection along different trajectories and perform local training of a global learning model to substitute the global positioning system (GPS) in urban areas. The prerequisites for its commercialization, however, lie in the location-dependent input data trustworthiness and participants’ privacy preservation. In this paper, we propose a privacy-preserving proof-of-location mechanism using blockchain to meet these conditions. Specifically, the proposed scheme utilizes a Threshold Identity-Based Encryption (TIBE) system for the generation of secret shares, such that each anonymous location proof can only be verified with at least a threshold number of participants. In addition, the proposed scheme exploits a cuckoo filter for the secure and efficient maintenance and dissemination of location proofs. Systematic security analysis is conducted to demonstrate the fulfillment of harsh security requirements. Performance evaluations are carried out to validate the computation efficiency in comparison with an oblivious transfer (OT) protocol, which has been widely adopted for secure data acquisition.

Federated learning-based automotive navigation has recently received considerable attention, as it can potentially address the issue of weak global positioning system (GPS) signals under severe blockages, such as in downtowns and tunnels. Specifically, the data-driven navigation framework combines the position estimation offered by the high-sampling inertial measurement units and the position calibration provided by the low-sampling GPS signals. Despite its promise, the privacy preservation and flexibility of the participating users in the federated learning process are still problematic. To address these challenges, in this article, we propose an efficient, flexible, and privacy-preserving model aggregation scheme under a federated learning-based navigation framework named FedLoc. Specifically, our proposed scheme efficiently protects the locally trained model updates, flexibly supports the fluctuation of participants, and is robust against unregistered malicious users by exploiting a homomorphic threshold cryptosystem, together with the bounded Laplace mechanism and the skip list. We perform a detailed security analysis to demonstrate the security properties in terms of privacy preservation and dishonest user detection. In addition, we evaluate and compare the computational efficiency with two traditional schemes, and the simulation results show that our scheme greatly improves the computational efficiency during participant fluctuation. To validate the effectiveness of our scheme, we also show that only part of the model update is excluded from aggregation in the case of a dishonest user.

With the advances of Internet of Things (IoT) solutions in intelligent transportation systems, collected vehicle data can produce insights on emerging vehicular phenomenon, and further contribute to the further improvement of innovative and efficient vehicular systems. Particularly, by leveraging data collected from vehicle sensors and maintenance models constructed from operation and repair history, predictive maintenance aims to detect the anomalies of vehicles and provide early warnings before the occurrence of failure. However, privacy preservation still remains as one of the top concerns for vehicle owners in predictive maintenance, as the sensory data could potentially violate their location and identity privacy. To address this challenge, in this article, we propose a privacy-preserving and verifiable continuous data collection scheme with the intent of predictive maintenance in vehicular fog, which gathers and organizes the sensor data of each individual vehicle on a sliding window basis. Specifically, our proposed scheme exploits the homomorphic Paillier cryptosystem and truncated α-geometric technique to protect the content of each individual piece of sensory data. Meanwhile, our proposed scheme also aggregates and authenticates the collected sensory data reports on a time-series sliding window basis, which achieves the continuous observation of the recently collected vehicular sensory data. Detailed security analysis is carried out to demonstrate the security properties of our proposed scheme, including confidentiality, authentication and privacy preservation. In performance evaluations, we also compare our proposed scheme with a traditional scheme, and our scheme shows great improvement in terms of communication and computation overheads. Furthermore, to show the feasibility of our proposed scheme, we also compare and discuss the expected squared error introduced by the differential privacy mechanism.

We implement a wireless indoor tracking system based on the variational Gaussian process state-space model (GPSSM) with smartphone-collected WiFi received signal strength and inertial measurement unit readings. We adapt the existing variational GPSSM framework to wireless tracking scenarios, and provide a practical learning procedure for the variational GPSSM. The proposed system explores both the expressive power of the non-parametric Gaussian process model and its natural mechanism for integrating the state-of-the-art tracking techniques designed upon state-space model. Experimental results obtained from a real office environment validate the outstanding performance of the variational GPSSM in comparison with the traditional parametric state-space model in terms of tracking accuracy.

We address the problem of robust network localization in realistic mixed LOS/NLOS environments. We make use of the fact that the bias of range measurement errors is not only non-negative but also sparse when LOS dominates, which has been long overlooked in the existing literature. To exploit these two properties, we introduce a sparsity-promoting regularization term and relax the resulting optimization problem to a semi-definite programming (SDP) problem. The proposed method admits a neat mathematical formulation and is computationally cheap. Moreover, its global convergence is guaranteed and it achieves good robustness against NLOS measurements. In numerical results, the proposed method outperforms representative state-of-the-art SDP approaches, in terms of both localization accuracy and computational efficiency.

We propose a novel Gaussian process kernel that takes advantage of a deep neural network (DNN) structure but retains good interpretability. The resulting kernel is capable of addressing four major issues of the previous works of similar art, i.e., the optimality, explainability, model complexity, and sample efficiency. Our kernel design procedure comprises three steps: (1) Derivation of an optimal kernel with a non-stationary dot product structure that minimizes the prediction/test mean-squared-error (MSE); (2) Decomposition of this optimal kernel as a linear combination of shallow DNN subnetworks with the aid of multi-way feature interaction detection; (3) Updating the hyper-parameters of the subnetworks via an alternating rationale until convergence. The designed kernel does not sacrifice interpretability for optimality. On the contrary, each subnetwork explicitly demonstrates the interaction of a set of features in a transformation function, leading to a solid path toward explainable kernel learning. We test the proposed kernel with both synthesized and real-world data sets, and the proposed kernel is superior to its competitors in terms of prediction performance in most cases. Moreover, it tends to maintain the prediction performance and be robust to data over-fitting issue, when reducing the number of samples.

Gaussian process (GP) for machine learning has been well studied over the past two decades and is now widely used in many sectors. However, the design of low-complexity GP models still remains a challenging research problem. In this paper, we propose a novel scalable GP regression model for processing big datasets, using a large number of parallel computation units. In contrast to the existing methods, we solve the classic maximum likelihood based hyper-parameter optimization problem by a carefully designed distributed alternating direction method of multipliers (ADMM). The proposed method is parallelizable over a large number of computation units. Simulation results confirm the benefits of the proposed scalable GP model over the state-of-the-art distributed methods.

Distributed Gaussian Process (GP) is a scalable Bayesian method that is promising for handling big data. Our contribution in applying GP for traffic prediction is two-fold. First, to the best of our knowledge, this paper is the first to empower GP regression with the alternating direction method of multipliers (ADMM) for distributed hyper-parameter optimization in the training phase, where the ADMM training framework well balances local estimation and information consensus in a principled way. Second, in the prediction phase, we fuse local predictions obtained from distributed computing units via a cross-validation based optimal strategy, which demonstrates itself to be reliable and robust for general regression tasks. Moreover, the cross-validation based optimal fusion strategy is built upon a well acknowledged probabilistic model to retain the valuable closed-form GP prediction properties. Experimental results show that our proposed distributed GP model can outperform the state-of-the-art distributed GP models considerably, in terms of wireless traffic prediction performance.

Visual tracking is fragile in some difficult scenarios, for instance, appearance ambiguity and variation, occlusion can easily degrade most of visual trackers to some extent. In this paper, visual tracking is empowered with wireless positioning to achieve high accuracy while maintaining robustness. Fundamentally different from the previous works, this study does not involve any specific wireless positioning algorithms. Instead, we use the confidence region derived from the wireless positioning Cramér-Rao bound (CRB) as the search region of visual trackers. The proposed framework is low-cost and very simple to implement, yet readily leads to enhanced and robustified visual tracking performance in difficult scenarios as demonstrated by our experimental results. Most importantly, it is utmost valuable for the practioners to pre-evaluate how effectively can the wireless resources available at hand alleviate the visual tracking pains.

The cloud radio access network (C-RAN) is a promising paradigm to meet the stringent requirements of the fifth generation (5G) wireless systems. Meanwhile, the wireless traffic prediction is a key enabler for C-RANs to improve both the spectrum efficiency and energy efficiency through load-aware network managements. This paper proposes a scalable Gaussian process (GP) framework as a promising solution to achieve large-scale wireless traffic prediction in a cost-efficient manner. Our contribution is three-fold. First, to the best of our knowledge, this paper is the first to empower GP regression with the alternating direction method of multipliers (ADMM) for parallel hyper-parameter optimization in the training phase, where such a scalable training framework well balances the local estimation in baseband units (BBUs) and information consensus among BBUs in a principled way for large-scale executions. Second, in the prediction phase, we fuse local predictions obtained from the BBUs via a cross-validation-based optimal strategy, which demonstrates itself to be reliable and robust for general regression tasks. Moreover, such a cross-validation-based optimal fusion strategy is built upon a well acknowledged probabilistic model to retain the valuable closed-form GP inference properties. Third, we propose a C-RAN-based scalable wireless prediction architecture, where the prediction accuracy and the time consumption can be balanced by tuning the number of the BBUs according to the real-time system demands. The experimental results show that our proposed scalable GP model can outperform the state-of-the-art approaches considerably, in terms of wireless traffic prediction performance.

Hyperparameter optimization still remains the core issue in Gaussian processes (GPs) for machine learning. The classical hyperparameter optimization scheme based on maximum likelihood estimation is impractical for big data processing, as its computational complexity is cubic in terms of the number of data points. With the rapid development of efficient parallel data processing on ever cheaper and more powerful hardware, distributed models and algorithms will become ubiquitous. In this letter, we propose an alternative distributed GP hyperparameter optimization scheme using the efficient proximal alternating direction method of multipliers, proposed by Honget al.in 2016, and we derive the closed-form solution for the local sub-problems. In contrast to the existing schemes of similar kind, our proposed one well balances the computational load on each local machine and the communication overhead required for global consensus of the local hyperparameter estimates. The proposed scheme can work in either a synchronous or an asynchronous manner, thus very flexible to be adopted in different computing facilities. Experimental results with both synthetic and real datasets validate the outstanding performance of the proposed scheme.

We propose a modified spectral mixture (SM) kernel that serves as a universal stationary kernel for temporal Gaussian process regression (GPR). The kernel is named grid spectral mixture (GSM) kernel as we fix the frequency and variance parameters in the original SM kernel to a set of pre-selected grid points. The hyper-parameters are the non-negative weights of all sub-kernel functions and the resulting optimization task falls under the difference-of-convex programming. Due to the nice structure of the optimization problem, the hyper-parameters are solved by an efficient majorization-minimization method instead of the gradient descent methods. It turns out that the solution is sparse, which provides us with a principled guideline to identify the important frequency components of the data. Experimental results based on various classic time series data sets corroborate that the proposed GPR with GSM kernel significantly outperforms the GPR with SM kernel in terms of both the mean-squared-error (MSE) and the stability of the optimization algorithm.

Wireless traffic prediction can effectively reduce the uncertainty in network demand and supply, and thus is a key enabler of smart management in next-generation wireless networks. To the best of our knowledge, this paper is the first to establish a wireless traffic prediction model by applying the Gaussian Process (GP) method based on real 4G traffic data. Our work is two-fold: First, based on the observed wireless traffic patterns, the kernel in our proposed GP model is designed accordingly to capture both the periodic trend and dynamic deviations; second, by leveraging the Toeplitz structure in the covariance matrix, the computational complexity of hyperparameter learning is significantly reduced from O(n3) to O(n2) and that of inference is reduced from O(n3) to O(n \log n), without any loss of prediction accuracy. Experimental results show that the proposed GP model can attain up to 97% prediction accuracy, and outperform the state-of-the-art algorithms considerably.

Many industrial processes are complex nonlinear distributed parameter systems (DPSs) with time-varying spatiotemporal dynamics. However, time-varying spatiotemporal dynamics and the nonlinear relationships between spatial points are currently not given much consideration in the existing data-driven modeling methods. Thus, accurately modeling a nonlinear DPS with time-varying spatiotemporal dynamics using these current methods is challenging. Here, we propose a spatiotemporal extreme learning machine (ELM) to accurately model time-varying and nonlinear DPSs. First, we develop the nonlinear spatial activation function to describe the nonlinear relationships between spatial points. As a result, in contrast to the traditional ELM method which is only used to model the temporal dynamics, the spatiotemporal ELM inherently takes the spatial information into consideration. Next, an online time coefficient model is developed, which accounts for the time-varying temporal dynamics of the DPS. After the integration of the spatial activation function with the time coefficient model, this modeling method is able to adapt to real-time spatiotemporal variation. Unlike the existing data-driven DPS modeling approaches, the proposed method has the capability to accurately represent the nonlinear relationships between spatial points and has the adaptive ability for modeling time-varying dynamics. Finally, through application on practical curing experiments, the proposed method can improve the modeling precision for an unknown, time-varying, and nonlinear DPS due to smaller modeling error as compared to the several commonly used DPS modeling methods.

We study severely quantized received signal strength (RSS)-based cooperative localization in wireless sensor networks. We adopt the well-known ‘sum-product algorithm over a wireless network’ (SPAWN) framework in our study. To address the challenge brought by severely quantized measurements, we adopt the principle of importance sampling and design appropriate proposal distributions. Moreover, we propose a parametric SPAWN in order to reduce both the communication overhead and the computational complexity. Experiments with real data corroborate that the proposed algorithms can achieve satisfactory localization accuracy for severely quantized RSS measurements. In particular, the proposed parametric SPAWN outperforms its competitors by far in terms of communication cost. We further demonstrate that knowledge about non-connected sensors can further improve the localization accuracy of the proposed algorithms.

There is a big trend nowadays toward indoor proximity report based positioning. A binary valued proximity report can be obtained opportunistically through event-triggering, leading to significantly reduced signaling overhead for wireless communications. In this paper, we aim to derive two types of fundamental lower bound, namely the Cram'er-Rao bound and the Barankin bound, on the mean-square-error of any proximity report based position estimator. Using the maximum-likelihood estimator as a representative example, we show that the Barankin bound is potentially much tighter than the Cram 'er-Rao bound and conclude that the Barankin bound ought be better suited for benchmarking any proximity report based position estimator.

The commercial interest in proximity services is increasing. Application examples include location- based information and advertisements, logistics, social networking, file sharing, etc. In this paper, we consider network-based positioning based on times series of proximity reports from a mobile device, either only a proximity indicator, or a vector of RSS from observed nodes. Such positioning corresponds to a latent and nonlinear observation model. To address these problems, we combine two powerful tools, namely particle filtering and Gaussian process regression (GPR) for radio signal propagation modeling. The latter also provides some insights into the spatial correlation of the radio propagation in the considered area. Radio propagation modeling and positioning performance are evaluated in a typical office area with Bluetooth-Low-Energy (BLE) beacons deployed for proximity detection and reports. Results show that the positioning accuracy can be improved by using GPR.

We consider received-signal-strength-based robust geolocation in mixed line-of-sight/non-line-of-sight propagation environments. Herein, we assume a mode-dependent propagation model with unknown parameters. We propose to jointly estimate the geographical coordinates and propagation model parameters. In order to approximate the maximum-likelihood estimator (MLE), we develop an iterative algorithm based on the well-known expectation and maximization criterion. As compared to the standard ML implementation, the proposed algorithm is simpler to implement and capable of reproducing the MLE. Simulation results show that the proposed algorithm attains the best geolocation accuracy as the number of measurements increases.

The problem of locating a mobile station in a wireless network has been extensively investigated due to the growing need for reliable location-based services. In non-line-of-sight environments, the positioning accuracy of classical least-squares based solutions is inaccurate. In order to mitigate the effect induced by non-line-of-sight errors, we address a novel robust nonparametric approach. Herein, we first estimate the range error distribution using nonparametric adaptive kernel density estimation and then optimize the approximate log-likelihood function via a quasi-Newton method. In simulations, the proposed approach shows improved positioning accuracy when the non-line-of-sight contamination is high as compared to several other competitors. Furthermore, it requires only a small amount of computational time.