Channel state information (CSI) estimation is one of the most fundamental problems in wireless communication systems. Various methods, so far, have been developed to conduct CSI estimation, which usually requires high computational complexity. However, these methods are not suitable for 5G wireless communications due to many techniques (e.g., massive MIMO, OFDM, and millimeter-Wave (mmWave)) to be employed in 5G wireless communication systems. In this paper, we propose an efficient online CSI prediction scheme, called OCEAN, for predicting CSI from historical data in 5G wireless communication systems. Specifically, we first identify several important features affecting the CSI of a radio link and a data sample consists of the information of these features and the CSI. We then design a learning framework that is a combination of a CNN (convolutional neural network) and an LSTM (long short term with memory) network. We further develop an offline-online two-step training mechanism, enabling the prediction results to be more stable when applied in practical 5G wireless communication systems. To validate OCEAN's efficacy, we conduct extensive experiments by considering four typical case studies, i.e., two outdoor and two indoor scenarios. The experiment results show that OCEAN not only obtains the predicted CSI values very quickly but also achieves highly accurate CSI prediction with up to 2.650% - 3.457% average difference ratio (ADR) between the predicted and measured CSI.

The rise of deep learning and the Internet of Things (IoT) has driven a number of smart-world applications, which are mostly deployed in distributed environments. Federated learning, a privacy-preserving collaborative learning paradigm, has shown considerable potential to leverage the rich distributed data at network edges. Nonetheless, the heterogeneity of IoT devices and their connected network environment impedes federated learning applications in IoT systems. Particularly, the stale gradients updated by slower local learners impact the effectiveness of federated learning. Transmitting weight updates with a large number of users leads to network congestion at the edge of the network and incurs unaffordable communication costs. To overcome these challenges, we propose a transfer knowledge based federated learning framework under a resource-limited distributed system. We formulate a knowledge distillation based federated learning optimization problem with the consideration of dynamic local resource. The proposed approach carries out federated learning with the help of knowledge distillation to avoid occupying the expensive network bandwidth or bringing a heavy burden to the network. Theoretical analysis demonstrates convergence of the learning process. The experimental results on three public datasets illustrate that the proposed framework is capable of substantially improving the efficiency of federated learning and outperforming state-of-the-art schemes.

The growing adoption of unmanned aerial vehicles (UAV) in civilian applications has highlighted the necessity of securing these systems against potential cyber threats. This paper presents a distilled pruned network for unmanned aerial vehicles (UAV-DiPNID), a specialized intrusion detection system designed to monitor and detect unauthorized or malicious activities involving UAVs. The proposed method leverages a combination of knowledge distillation and model pruning to develop an effective and efficient intrusion detection system based on a deep neural network. Knowledge distillation transfers knowledge from a larger, more complex model (teacher) to a smaller, simpler model (student) to enhance inference time and model compactness. Additionally, model pruning is employed to reduce the model’s size by removing unnecessary connections and weights, further optimizing resource usage. The experimentation and evaluation of the proposed method demonstrate its effectiveness in terms of intrusion detection accuracy, inference time, and model size. Extensive experiments on the UAV-IDS 2020 dataset demonstrate that UAV-DiPNID achieves an average intrusion detection accuracy of 99.61%, outperforming the state-of-the-art benchmark of 99.37%. Furthermore, our method reduces the inference time by up to 80.70% and the model size by up to 90% compared to conventional convolutional neural network-based approaches. These results highlight the superiority of UAV-DiPNID compared to baseline intrusion detection systems for UAVs, making it well-suited for real-time applications in resource-constrained environments. Furthermore, the accuracy of intrusion detection remains high, showcasing the model’s robustness in capturing complex patterns and features.

Artificial Intelligence (AI) has been widely adopted in numerous fields and enabled various smart systems because of its strong ability to perform tasks, including prediction, event detection, and status estimation, among others. As one of the typical smart systems empowered by AI and Internet of Things (IoT) technologies, the smart transportation system has made dramatic progress for traditional transportation in numerous aspects, including autonomous driving, smart traffic lights, navigation, and traffic forecasting, among others. Deep learning is an essential component to enable such smart systems. Typically, specific deep learning models, e.g., Recurrent Neural Networks (RNN) and Convolutional Neural Networks (CNN), can be trained on collected transportation data for a particular task. However, traditional deep learning techniques rely on data sufficiency to build an effective model. Additionally, each trained model can only work on one single task. This has limited the efficacy of deep learning techniques in numerous application scenarios. To this end, multi-task learning (MTL) has been studied to train a single model that can work for multiple tasks. This technique effectively allows the learning model to expand to specific tasks with only a limited amount of data affiliated. In the meantime, MTL significantly reduces the training time of each task. The success of MTL requires that there are potential relationships among different tasks. Many tasks in the smart transportation system are related. For instance, traffic speed and vehicle volume estimations for each road are highly correlated. Based on this, research on applying MTL in smart transportation systems has been studied recently. This paper reviews the recent efforts to use MTL in smart transportation systems and conducts an extensive survey to provide insights. In particular, we categorize the MTL applications in smart transportation systems into traffic forecasting, traffic sign recognition, vehicle recognition, travel time estimation, road safety estimation, taxi demand prediction, and autonomous driving. Ultimately, we discuss challenges and future research directions in applying MTL in smart transportation systems.

The Internet of Things (IoT) integrates information collected from multisources and is able to support various intelligent smart city applications, such as industrial manufacturing, power systems, and mobile healthcare. In the big data era, multisourced data are collected on a daily basis, whereas a large part of the data may be irrelevant, redundant, noisy, or even malicious from a machine learning perspective. Feature selection has been a powerful data cleaning technique to reduce data redundancy and improve system performance in machine learning. Inspired by reinforcement learning that learns from its experience, in this article, we propose a novel efficient deep $Q$ -network (DQN)-based feature selection method for multisourced data cleaning. In particular, we model the feature selection problem as a competition between an agent and the environment in dynamic states, which is solved by a DQN. Traditional DQN suffers from high computational complexity and requires a significant amount of time in order to converge in the training process. To tackle these challenges, we develop a space searching algorithm called SS to speed up the training process of the DQN agent. To validate the efficacy and efficiency of the proposed method, we conduct extensive experiments on various types of IoT data. Simulation results show that the proposed DQN-based feature selection algorithms achieve much better performance compared with state-of-the-art methods, and are robust under data poisoning attacks.

Nonlinearly constrained nonlinear programming (NLC-NLP) problems arise in various optimal decision-making fields, such as financial engineering, urban planning, supply chain management, and power system control. These real-world NLC-NLP problems are usually large-scale because they need to take into account massive variables and constraints. Solving NLC-NLP problems with common algorithms, e.g., the gradient projection method (GPM), generally incurs very high time complexity. This results in the inevitable inability of solving large-scale NLC-NLP problems by using general-purpose computers within a feasible time. To address this challenge, a cost-effective solution is to adopt cloud computing, but this raises security concerns since real-world NLC-NLP problems always carry sensitive information. Previous secure outsourcing algorithms try to protect sensitive information, but still let cloud service tenants bear heavy computation workload. To fill the gap, we develop a practical secure outsourcing algorithm for solving large-scale NLC-NLP problems with the GPM. To be more prominent, we also parallelize the secure outsourcing algorithm for accelerating computations and avoiding possible memory overflowing. We implement the proposed algorithm on the Amazon Elastic Compute Cloud (EC2) and a laptop, and notice that it can significantly reduce the tenant's computing time even for large-scale NLC-NLP problems.