The accelerated development of the Industrial Internet of Things (IIoT) is catalyzing the digitalization of industrial production to achieve Industry 4.0. In this article, we propose a novel digital twin (DT) empowered IIoT (DTEI) architecture, in which DTs capture the properties of industrial devices for real-time processing and intelligent decision making. To alleviate data transmission burden and privacy leakage, we aim to optimize federated learning (FL) to construct the DTEI model. Specifically, to cope with the heterogeneity of IIoT devices, we develop the DTEI-assisted deep reinforcement learning method for the selection process of IIoT devices in FL, especially for selecting IIoT devices with high utility values. Furthermore, we propose an asynchronous FL scheme to address the discrete effects caused by heterogeneous IIoT devices. Experimental results show that our proposed scheme features faster convergence and higher training accuracy compared to the benchmark.

Recent years have seen rapid deployment of mobile computing and Internet of Things (IoT) networks, which can be mostly attributed to the increasing communication and sensing capabilities of wireless systems. Big data analysis, pervasive computing, and eventually artificial intelligence (AI) are envisaged to be deployed on top of the IoT and create a new world featured by data-driven AI. In this context, a novel paradigm of merging AI and wireless communications, called Wireless AI that pushes AI frontiers to the network edge, is widely regarded as a key enabler for future intelligent network evolution. To this end, we present a comprehensive survey of the latest studies in wireless AI from the data-driven perspective. Specifically, we first propose a novel Wireless AI architecture that covers five key data-driven AI themes in wireless networks, including Sensing AI, Network Device AI, Access AI, User Device AI and Data-provenance AI. Then, for each data-driven AI theme, we present an overview on the use of AI approaches to solve the emerging data-related problems and show how AI can empower wireless network functionalities. Particularly, compared to the other related survey papers, we provide an in-depth discussion on the Wireless AI applications in various data-driven domains wherein AI proves extremely useful for wireless network design and optimization. Finally, research challenges and future visions are also discussed to spur further research in this promising area.

Channel estimation for downlink millimeter wave (mmWave) multi-user multiple-input-multiple-output (MUMIMO) communications over a time-varying channel is demandingly challenging. This is due to a massive number of channel coefficients to be estimated and the severe propagation loss of mmWave transmissions. To tackle this problem, this paper proposes an effective channel estimation scheme for time-varying mmWave channels. Discovering the property that the variations of angles of arrival/departure (AoAs/AoDs) are much slower than that of path gains, we propose a novel transmission frame structure, where the channel estimation is decoupled into two separated stages including AoAs/AoDs estimation followed by path gains estimation. In the first stage, the sparse nature of mmWave channels is leveraged to formulate the AoAs/AoDs estimation into a block-sparse signal recovery problem, and we develop an adaptive angle estimation (AAE) algorithm to estimate the AoAs/AoDs. In the second stage, the pilot beamforming based on the estimated AoAs/AoDs is designed aiming for maximizing the desired pilot power so that the path gains can be accurately estimated. Simulation results demonstrate that the proposed scheme can effectively estimate the time-varying mmWave channels with a small number of pilots.

In this paper, we propose a codebook-based beam tracking strategy for mobile millimeter-wave (mmWave) systems, where the temporal variation of the angle of departure (AoD) is considered. A closed-form upper bound of the average tracking error probability (ATEP) is derived and further optimized. We first consider a slow-varying scenario where narrow training beams implemented by single radio-frequency (RF) chain are employed. We show that the ATEP can be reduced by optimizing the power allocation strategy over these training beams, which is formulated and transformed into a second-order cone programming. The fast-varying scenario is further considered where the use of narrow training beams becomes inefficient due to the rapid variations of AoD. In order to reduce the training time, multiple RF chains generating wide beams are employed to track the AoD’s variations, and the associated beam pattern design problem is shown to be a 0 − 1 nonlinear optimization problem (NLP). A sequential quadratic programming method is used to solve this binary NLP. To reduce the complexity, a progressive edge-growth algorithm is further introduced by associating the binary NLP with a bipartite graph. Numerical results demonstrate significant gains of the proposed beam tracking strategy over existing benchmarks for both scenarios.

Network slicing is a key enabler of 5G and beyond networks to satisfy the diverse quality of service (QoS) requirements of different services simultaneously. In network slicing, radio access network (RAN) slicing is essential to establish a functional network slice by connecting mobile devices and mapping virtualized resource units to different slices. This requires a highly efficient resource allocation scheme to maximize resource utilization efficiency and meet the diverse QoS requirements. In this paper, we propose a dynamic RAN slicing model that incorporates multiple distributions to accommodate different user request types and diverse priorities among traffic types in the same slice, where the total available resources are dynamically changing over time. We formulate resource allocation as a time-sequential dynamic optimization problem that takes into account system stability, resource limitation, different timescales, long-term system performance, and user priority. We propose a deep reinforcement learning-based (DRL-based) approach referred to as prediction-aided weighted DRL (PW-DRL) to online infer the power allocation and user acceptance decisions that can maximize a predefined reward function. Additionally, a prediction network is formulated to capture the correlation between current and future states. Simulation results validate that our proposed PW-DRL significantly outperforms state-of-the-art approaches by achieving the highest long-term reward and fastest convergence.

Due to the rapidly increasing number of base stations (BSs) in the operational cellular networks, their energy consumption is escalating. In this paper, we propose an intelligent data-driven BS sleeping mechanism relying on a wireless traffic prediction model that measures the BSs' capacity in different regions. Firstly, a spatio-temporal cellular traffic prediction model is proposed, where a multi-graph convolutional network (MGCN) is developed to capture the associated spatial features. Furthermore, a multi-channel long short-term memory (LSTM) solution involving hourly, daily and weekly periodic data is used to capture the relevant temporal features. Secondly, the capacities of macro-cell BSs (MBSs) and small-cell BSs (SBSs) having different environment characteristics are modeled, where both clustering and transfer learning algorithms are adopted for quantifying the traffic supported by the MBSs and SBSs. Finally, an optimal BS sleeping strategy is proposed for minimizing the network's power consumption. Experimental results show that the proposed MGCN-LSTM model outperforms the existing models in terms of its cellular traffic prediction accuracy, and the proposed BS sleeping strategy using an approximated non-linear model of the associated capacity function achieves near-maximal energy-saving at a modest complexity.