With the rapid increase of industrial systems, industrial spectrum is stepping into the era of big data, and at the same time spectrum resources are facing serious shortage. Cognitive industrial system (CIS) based on cognitive radio can improve spectrum utilization by accessing the idle spectrum licensed to primary user (PU). However, the CIS must find enough idle channels by performing spectrum sensing. In this paper, a reinforcement learning-based multislot double-threshold spectrum sensing with Bayesian fusion is proposed to sense industrial big spectrum data, which can find required idle channels faster while guaranteeing spectrum sensing performance. Double thresholds are set to guarantee both high detection probability and spectrum access probability, and weighed energy detection is proposed to maximize detection probability when the energy statistic falls into the confusion area between the double thresholds. Bayesian fusion is proposed get a final decision on the channel availability by combining the local sensing decisions of all the time slots. An idle channel prediction and selection algorithm is proposed to predict the idle probability of each channel and find required idle channels from the sorted channel set. From simulation results, the proposed spectrum sensing scheme outperforms cooperative spectrum sensing and energy detection, which can predict idle channels accurately and get needed idle channels with fewer sensing operations.

This paper considers worst-case downlink secure communication from base station (BS) to mobile legitimate vehicle at the presence of an eavesdropper, where reconfigurable intelligent surface mounted unmanned ariel vehicle (RIS-UAV) acts as an ariel passive relay to assist the communication. Our aim is to achieve the maximum worst-case downlink secrecy rate via optimizing power allocation, RIS passive beamforming and UAV trajectory. The formulated optimization problem is non-convex and non-smoothness, which is quite hard to solve directly. Therefore, we divide the original optimization problem into three subproblems, power allocation optimization, passive beamforming optimization and UAV trajectory optimization, which can be solved by applying successive convex approximation (SCA), $\mathcal {S}$ -Procedure and semidefinite relaxtion (SDR). Finally, we propose an alternating iterative optimization algorithm to get suboptimal solutions by alternately optimizing the three subproblems. Numerical simulations show the effectiveness of the proposed scheme against the other benchmarks.

Unmanned aerial vehicle (UAV) enabled Internet of Things (IoT) can keep network connectivity when the ground infrastructures are paralyzed. However, its transmission perform will be restricted due to the limited energy of the UAV. In this paper, a multi-UAV enabled IoT is proposed, where the UAVs as base stations send information to the ground IoT nodes via downlink within the flight time. And a fair energy-efficient resource optimization scheme for the IoT is studied to ensure fair energy consumption of multiple UAVs. The optimization problem seeks to maximize the minimum energy efficiency of each UAV by jointly optimizing communication scheduling, power allocations and trajectories of the UAVs. We decompose the non-convex optimization problem into three sub-optimization problems and solve them by Dlinkelbach method and successive convex approximation (SCA). Then a joint optimization algorithm is put forward to obtain the global optimal solutions by iteratively optimizing the three sub-optimization problems. The simulations results show that the multi-UAV enabled IoT can achieve significant performance improvement, and the energy efficiency between UAVs can achieve relative fairness by the fair resource optimization.

The key technologies for applying 5G in high-altitude platform station (HAPS) communication systems have many outstanding advantages, especially large-scale antenna array technology. HAPS communication systems have inherent advantages that can perfectly complement large-scale array antenna technology and solve the problems of many existing ground communication systems. The chances of long-packet data users and short-packet data users being serviced are different due to the differing data length of each user's packet. According to the user's channel environment and data length, the user with the smallest delay is selected for transmission. Dynamic and interdependent characteristics between systems and the evaluation of the efficiency of wireless resources are studied in this article. Game theory is applied to the evaluation of the wireless resource efficiency of the system, and an efficiency evaluation model is constructed. In addition, the efficiency of this preliminary work is verified by simulation.

There are some urgent problems in the construction of the Pearl River Delta smart water conservancy project. In terms of operation, this project still needs improvement. Particularly, localization and tracking of personnel and equipment are the technical issues that urgently need to be solved. In order to improve the accuracy and the environmental adaptability of personnel and equipment localization algorithms in the construction and operation of pipeline network, this paper proposes a fingerprint localization algorithm (KF-KNN) based on frequency modulation (FM) signals. Firstly, the FM data collection device is used to obtain the received signal strength indicator (RSSI) fingerprint information within the coverage area, and train them to build a fingerprint database. Secondly, the k-nearest neighbors (KNN) technology is used to complete the rough localization calculation based on the RSSI data received by the module, which includes the RSSI fingerprint database and the environmental noise parameters. Finally, the Kalman filter model is used to predict and optimize the rough position information in order to have better environmental adaptability and effectively improve the accuracy of localization. The analysis results show that the proposed KF-KNN algorithm has better performance in localization with average localization error as low as 1.9 meters compared with the original KNN and the weighted k–nearest neighbors (WKNN) fingerprint localization models.