Constructing underwater acoustic sensor networks (UASNs) for data collection has gradually become an effective ocean exploration and exploitation method. However, the interference of the underwater environment and the limited capacity of underwater communication equipment increase the difficulty of information interaction, posing a challenge to secure data transmission strategies for UASNs. Therefore, this study proposes a safe and reliable data transmission scheme based on reinforcement learning-aided two-stage trust evaluation (RLTST) to overcome the problems mentioned above. This article proposes a distinct self-trust concept, different from traditional trust mechanisms. A node self-trust evaluation method based on Q-learning is designed in the first stage, which defects compromised nodes actively. In the second stage, the trustworthiness of data is calculated based on the real data received, followed by backtracking the transmission path of untrustworthy data to identify malicious nodes. Finally, the results show that our proposed scheme is more effective in malicious node detection and improves data collection reliability.

Wireless communication with the enormous demands of sensing ability have given rise to the integrated passive sensing and communication (IPSAC) technology. The main challenge of IPSAC is how to achieve high sensing and communication performance by integrating the passive sensing and communication demodulation. In this paper, we propose an integrated sensing and communication (ISAC) signal processing optimization scheme by jointly processing the pilot and data signals. To solve the optimization problem, we propose an ISAC signal processing algorithm based on iterative optimization, which alternates the passive sensing and channel reconstruction to realize target sensing. However, the hyper-parameter configuration of the iterative optimization algorithm influences the performance of target detection and communication demodulation. Recognizing this fact, we propose a model-driven ISAC network (ISAC-NET) that adopts the block-by-block signal processing method to improve the communication and sensing performance. The proposed ISAC-NET obtains suitable hyper-parameters by deep learning to guarantee the performance and convergence of communication and sensing signal processing. From the simulation results, ISAC-NET obtains better communication performance than the traditional signal demodulation algorithm, which is close to OAMP-Net2. Compared to the 2D-DFT algorithm, ISAC-NET demonstrates significantly enhanced sensing performance. In summary, ISAC-NET is a promising tool for the IPSAC systems.

Data-driven approaches have gained great success in the field of rotating machinery fault diagnosis for its powerful feature representation capability. However, in most of the current studies, model training process requires massive fault data which are costly to gather or even unavailable in some extreme operating conditions. At the same time, structural relationships between samples are not fully exploited to facilitate the model performance. In response to these problems, a novel GHOAN for rotating machinery fault diagnosis is proposed in this study. Specifically, the proposed approach incorporates the advantages of improved graph attention network model and the multiorder neighborhood feature perception to achieve richer feature representation by aggregating features from multiple neighborhood domains. In this way, effective fault diagnosis may be achieved by using fewer training samples based on vibration signal analysis. The results of experiments conducted on two benchmarking datasets and a practical experimental platform show that the proposed GHOAN achieve superior performance.

In industrial Internet of Things, the combination of specific emitter identification (SEI) and key authenticated technologies can effectively resist spoofing attacks and improve system security. However, most existing SEI approaches extract features based on real valued operations and only work in static scenario. This motivated us to develop a novel SEI method tasked with: exploiting the high potential model for SEI based on the inphase/quadrature (I/Q) signal that is represented by complex number, and realizing rapid reconstruction of the model in the face of dynamic scenarios. To this end, in this article, we introduce a two-stage cross-domain identification model. First, a complex-valued separate residual network (CVSRN) with novel separate residual modules is proposed as the pretrained model. The CVSRN can automatically extract effective inherent features directly from raw signals in an end-to-end manner, which favors complex-valued signals that are found in two distinct signal paths. Second, three transfer strategies are proposed to achieve rapid construction of the target SEI model. They leverage the knowledge learned from the pretrained CVSRN to facilitate the recognition of a new but similar emitters. We benchmark our proposed approach against four state-of-the-art SEI methods on real-world data and exhibit that it is not only competitive but also able to cope with complex dynamic scenarios.

In the process of data acquisition of underwater acoustic sensor networks (UASNs), the safety of the network is threatened by the disclosure of source node location information. So how to protect the security and privacy of source node location is the main challenge faced by UASN security. To realize this taeget, a hierarchical structure-based algorithm for protecting source location privacy (HSSLP) is proposed in this paper. Firstly, it is proposed to divide UASNs into dynamic and static layers based on Ekman drift model. Location privacy protection schemes suitable for source nodes located in different layers have been proposed separately. In the static layer, k-means clustering separates the nodes into groups, and the source node’s location privacy is protected using fake source node and phantom nodes, while auxiliary cluster head and sleep scheduling mechanism are used to save node energy. Nodes in the dynamic layer, whose positions are prone to change, are no longer clustered. The source node makes use of inducing nodes to take adversaries away from the source node, enhancing the privacy and security of the source node with minimal energy expenditure. Finally, autonomous underwater vehicles (AUV) need to support the cluster head in collecting data combined in the static layer and data uploaded in the dynamic layer. Based on the communication range of AUV, the network is segmented into areas, and when the AUV receives warning messages while traveling, it changes its route to lead the adversary to an area remote from the source node. Simulation results show that the proposed algorithm owns the capacity to balance the relationship between network security, transmission delay, and node energy consumption. To be more specific, the HSSLP algorithm improves the safety time by about 50$\%$, reduces the delay by about 20$\%$and saves the node energy by about 36$\%$as compared to the DIS-PLP algorithm.

Unauthorized underwater vehicles (UUVs) pose a serious threat to maritime security. To preserve maritime security, it is essential to pursue these UUVs. The majority of traditional pursuit methods are based on known environmental dynamics. However, the underwater environment is too complicated and unpredictable to describe these dynamics accurately. This study developed a novel online decision-making technique called multi-agent distributional soft actor-critic (MADA) to handle the issue of underwater cooperative pursuit. The method constructs a control-oriented framework based on multi-agent reinforcement learning that can map autonomous underwater vehicle (AUV) observations to pursuit actions. Multiple AUVs can combine to make prompt pursuit decisions. Then, the proposed method combines distributional soft actor-critic and curriculum learning to improve the success rates of multiple AUVs in pursuing UUVs. Experimental results show that the MADA can obtain a better cooperative pursuit strategy.

Remaining useful life (RUL) prediction of aero-engines is one of the important issues in research related to engine health management. Although deep learning has made great progress in fault diagnosis research, successful training of deep learning models is very time-consuming and difficult to meet the real-time requirements of online RUL prediction applications. Broad learning systems (BLS) provide an alternative to deep learning networks with low computational resource requirements, fast training time, and incremental scalability. Based on the typical BLS, we propose a new lightweight multiscale BLS (MSBLS). Considering that RUL is influenced by the working condition factor, the discrete wavelet transform is used to generate multiresolution components, and then feature nodes are extracted on top of the components. An elastic net regularization technique is used to constrain the output weights of the nodes, preserving the significant nodes, and finally obtaining a more sparse MSBLS. Experiments are conducted using the NASA publicly available commercial modular aero-propulsion system simulation (C-MAPSS) dataset and the N-CMAPSS dataset, and our proposed MSBLS not only improves the accuracy of RUL prediction but also has a very short training time compared with the latest research methods nowadays.

In existing research of cache-enabled UAV networks, cell association is content-centric regardless of UAV line-of-sight (LOS)/non-line-of-sight (NLOS) channel dynamics, and the fixed content distribution hinders system performance. This paper proposes user-centric cooperation for cache-enabled UAV networks to exploit both contents and transmission diversity, and optimizes the probabilistic caching distribution to minimize system outage probability considering UAV interference topology dynamics. Specifically, user-centric UAV group (UAVG) is formed for each user to increase both aggregated cache size and transmission resources, considering UAVs can’t realize inter-nodes cache file transform due to limited wireless backhaul. Users will search the required content within the UAVG and associate to the cache hit UAV that provides the strongest signal-to-interference ratio (SIR), thus the intrinsic gain of transmission diversity can be achieved. If the required content can’t be successfully transmitted by UAVG, the macro base stations (MBSs) will help transmit the target content. In addition, based on analytical results, the optimal content distribution (OCD) probability is optimized by Lagrange multiplier, thus the tradeoff between content diversity gain and limited cache size is exploited. Analytical results show that, compared with caching the most popular contents policy, our proposed OCD model can decrease system average outage probability by 44.8%.

Millimeter-wave Integrated Sensing and Communications (ISAC) with multi-beam design holds significant promise for vehicular networks, offering multi-target omnidirectional sensing and high-capacity communication services concurrently. Nonetheless, the considerable challenge of potential mutual interference arises due to the high mobility and density of transmitters in such networks. To address this challenge effectively, we propose leveraging inter-vehicle communication to schedule communication and sensing signals for vehicles, thereby enhancing networked sensing capabilities. We first introduce an analytical framework to characterize the mutual interference among multiple vehicles. Subsequently, we evaluate the effectiveness of our proposed interference mitigation method in terms of interference probability, duration, and the achievable detectable density. Additionally, recognizing the different performance requirements of communication and sensing functions, we investigate a joint resource allocation problem catering to both aspects. Simulation results demonstrate a notable enhancement in the proposed ISAC-based interference mitigation, with a 58% reduction in interference probability compared to benchmarking schemes.

To realize higher-level autonomous driving and advanced transportation applications, the introduction of the integrated sensing and communication (ISAC) technique in vehicular networks is indispensable. Different from the existing works, this paper investigates the power allocation problem for onboard ISAC systems of vehicles, during the vehicle-to-infrastructure communication, vehicle-to-vehicle communication and sensing progress, in case of the time-varying communication channel gains, the time-varying impulse responses of sensed targets, and the stochastic traffic. Note that both the inter-beam interference of a single vehicle and the inter-vehicle interference are important considerations. Specifically, we formulate a stochastic programming problem, which optimizes the sensing performance, subject to constraints on the network stability, power limits and quality-of-service requirements. Leveraging the Lyapunov optimization technique, this stochastic programming problem is transformed into a single-time slot non-convex problem. Taking advantages of genetic algorithm and particle swarm optimization (PSO), a hybrid meta-heuristic algorithm is designed to solve the non-convex problem. Typically, we improve the traditional PSO to balance the global search ability and local search ability of particles. Finally, a dynamic power allocation strategy is proposed. The theoretical analysis and simulation results show that this strategy achieves a communication performance-sensing performance tradeoff of [$ {\mathrm {O(}}1/V{\mathrm {)}} $,$ {\mathrm {O(}}V{\mathrm {)}} $] with$ V $being a control parameter.

Specific emitter identification (SEI) is a technology that extracts subtle features from signals sent by emitters to identify different individuals. It can effectively improve the security of the Industrial Internet of Things (IIoT) by acting on the physical layer of the internet. Recent research on SEI has focused on deep learning (DL) models that can automatically learn effective inherent emitter features from raw signals. Nevertheless, training popular DL models is computationally expensive because of the numerous hyperparameters and nonscalable structures. This limits the application of DL-based SEI models in certain practical IIoT scenarios. To address this concern, we propose an adaptive broad learning (ABL) method to build a lightweight SEI model. In the proposed model, the raw signal samples are mapped to feature nodes, and the emitters are denoted as the output nodes. The hidden nodes are directly connected to the output nodes by a broad network. Through this flat structure, the size and calculation amount of the model can be effectively reduced. To further economize the computational cost, we designed an adaptive node expansion strategy for rapidly obtaining the optimal hyperparameters of the models. The results of experiments on real-world data prove the superiority of ABL over popular state-of-the-art DL-based SEI models.

UAV swarms based on cooperative communication networks are widely used in many fields, which have the advantages of high mobility, high flexibility and low cost. However, UAVs face limited spectrum resources in a specific area and may interfere with primary users. Effective communication management between UAVs is a challenging problem. There-fore, this paper proposes a UAV clustering method based on the improved cluster head selection weight, which provides an effective management for the communication between UAVs and improves the efficiency of data collection. The proposed algorithm employs a new cluster head selection strategy based on the searched targets and available channel resources. Moreover, we analyze the weight factors of UAVs in flight and communication energy consumption. Considering the decreasing the member of the UAV clusters, we also design a maintenance strategy to improve the degree of data sharing in the cluster. The experimental results show that, compared with the traditional UAV clustering methods, the proposed method can effectively improve the network management for communication resources, reduce the collision and interference rate with the primary user by 25%, shorten the time required to fully acquire multi-target point data for the first time by 9%, and increase the amount of target point data collected by 26%.

Existing deep learning-based beamforming models’ input sizes are fixed, and they can only be used in multiple-input multiple-output (MIMO) networks containing a fixed number of users and base stations. Therefore, they cannot work and need retraining when the number of users is dynamic. In this letter, a fully convolutional beamforming model named FC-BFNet is proposed, where data of different dimensions can be addressed to adapt to the dynamic environment without retraining the model. Specifically, FC-BFNet has two computational parts which are mainly made up of convolution layers because the convolution operation is not sensitive to the size of the input data. Differently and importantly, the first part extra uses max-pooling to calculate the underlying features of the input and stores max-pooling indices, while the second part extra leverages up-pooling based on stored indices to obtain the expected results from the underlying features and ensure the size consistency of input and output. Extensive experiments demonstrate FC-BFNet has better dynamic environment adaptability and sum-rate performance than other baselines, while maintaining great computing efficiency.

Extended reality (XR) technologies, including augmented reality (AR) and virtual reality (VR), is now used in a wide spectrum of applications, e.g., telemedicine, remote education, and computer gaming. However, existing experiments are predominantly deployed in the indoor environment with the help of wired connectivity or Wi-Fi networks. Recently, the global commercialization of 5G networks provides new opportunities for XR applications. By using wireless cellular network, XR is able to offer seamless user experience from anywhere at any time. Despite remarkable advantages, current cellular-enhanced XR services are still facing certain challenges. For instance, edge users may not be able to achieve a reliable XR experience due to the limitation of transmission power and free-space path loss. In this paper, a novel flexible spectrum access (FSA) approach is proposed, which allows XR users to be flexibly handed over between multiple carriers. By such means, the reliability for the edge users can be significantly improved thanks to the use of lower frequency bands. It is also shown that the end-to-end latency has been reduced due to more evenly distributed time frequency domain resources.

This paper investigates the multi-channel access and rendezvous problem in unmanned aerial vehicle (UAV) relaying system in the absence of pre-allocated control channel. Both the geographical sensing range and the spectrum sensing bandwidth of each UAV are limited due to onboard payload constraints, hence it becomes challenging to design effective and efficient channel rendezvous mechanisms. To address the challenge, this paper first observes and analyzes the effects of UAV relaying network topology and geographical sensing range on the rendezvous, and it is found that a joint exploitation of motion and frequency control is essential to achieve efficient rendezvous. Based on the important insight, this paper formulates the UAV relaying rendezvous problem and proposes a novel joint space-frequency rendezvous (JSFR) method for multi-UAV networks, incorporating with distributed reinforcement learning techniques. The simulation results show that the JSFR method may significantly improve the effectiveness and efficiency of rendezvous in UAV relaying networks, in terms of rendezvous probabilities and convergence rates.

The last ten years have witnessed explosive growth in mobile data traffic, which leads to rapid increases in energy consumption of cellular networks. One potential solution to this issue is to seek out a green deployment strategy. In this paper, we investigate the energy-efficient deployment strategy under coverage performance constraints for both homogeneous and heterogeneous cellular networks. Unlike just considering the base station (BS) density in previous work, we jointly optimize the BS density and the BS transmission power. First, we derive the relation between the average coverage probability and deployment strategy (i.e., BS density and BS transmission power) with stochastic geometry tools. Then, based on the expression results, we formulate a network energy consumption minimization framework considering coverage performance constraints and jointly determine the optimal macro BS (MaBS) density, MaBS transmission power, and micro BS (MiBS) density. With practical data sets, numerical simulation results show the following: 1) compared with homogeneous network deployment, heterogeneous network deployment has the advantage in energy efficiency performance, and 2) our joint BS density and BS transmission power optimization strategy exceeds the existing strategy, which just considers the BS density optimization in terms of energy efficiency.

Coded cooperation (CC) is an excellent scheme for cooperative communication. However, the throughput of CC is not high enough. In this paper, we propose a novel cooperative scheme called hierarchically modulated coded cooperation (HMCC). This scheme takes the frame group as the input instead of a single frame in CC and adopts hierarchical modulation in source node according to the asymmetry between the links of source-relay and source-destination, expecting that some frames can be received by destination node directly and others are forwarded by relay node. In HMCC, a higher order modulation can be used by source node. Theoretical analysis and simulation show that HMCC achieves a better performance on throughput compared to other schemes. On the other hand, HMCC distributes burst errors to the frame group owing to the hierarchical modulation, which suggests that HMCC is able to struggle against the burst errors.

The Femtocell is an attractive solution to improve the indoor coverage of cellular systems. By applying spectrum reuse, femtocells can improve the efficiency of spectrum usage. However, since femtocells are deployed by end-users, insufficient coordination between base stations leads to the reuse of the same frequency resources and results in CCI (co-channel interference), which seriously complicates the synchronization problem, particularly in OFDM based systems, such as WiMAX and LTE. In this paper, we investigate the timing synchronization problem in the presence of CCI of an OFDMA macro/femto uplink scenario. We consider the cross-correlation based timing synchronization and propose a pre-filter to mitigate the CCI effect. The simulation results show that this method effectively eliminates the effect of CCI on the cross-correlation based timing synchronization and provides a performance close to that of no CCI case.

Nowadays, energy saving in wireless communications has become a hot topic as energy consumption increasingly becomes a global environment problem. In this paper, we formulate an Energy Consumption Rating (ECR) minimization problem in a multi-layer network and provide analysis of its property and complexity. To solve this problem, we propose the Enhanced Mobility Load Balancing (EMLB) firstly. A heuristic and practical algorithm is then introduced, which transfers traffic using EMLB in vertical and horizontal directions and adaptively switches off/on some cells based on traffic load condition. The performance of the proposed algorithm is evaluated by comparing with existing static and dynamic schemes through system level simulations. The simulation results demonstrate that our proposed scheme not only has good system throughput performance, but also achieves significant power saving in typical traffic scenario.

In order to provide high data rate for indoor services, femtocells are proposed in LTE-Advanced system. Under this architecture, the main problem is how to reduce the interference between macro and femto cells and that among femtocells. In this paper, an interference graph is constructed in which the vertexes are Macro User (MUE) and femtocell. Besides, Regional Average Channel State (RACS) metric is proposed to estimate the weight of interference. Therefore, a dynamic spectrum assignment algorithm called hybrid clustering based on interference graph (HCIG) is proposed to reduce the interference, in which the optimal clustering problem is constructed as a MAX-K cut problem and a heuristic algorithm is given. Based on the cluster results, a resource allocation scheme is given to reduce the interference and improve the spectrum efficiency. System level simulation results show that compared to other three schemes, the SINR of both MUE and femto user are improved by HCIG.

Time synchronization is an important challenge for femtocell design. An inaccurate clock can introduce interference, degrade spectrum accuracy and disrupt handovers. Among the potential solutions, wireless network-based time synchronization turns out to be cost effective because no extra provisions are required to ensure synchronization even on a regular basis. In wireless networks, receiver-receiver synchronization schemes offer very good performance. In this paper, we design an MS-assisted receiver-receiver synchronization strategy for femtocells. Two schemes are proposed for different scenarios to ensure a more comprehensive availability. Analysis and simulation results demonstrate that our proposed schemes provide sufficient synchronization accuracy for practical scenarios.