24 December 2019 in IEEE Transactions on Industrial Informatics
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.
21 October 2020 in IEEE Access
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.
28 May 2019 in EURASIP Journal on Wireless Communications and Networking
Channel allocation is the prerequisite for the HAPS (high-altitude platform station) 5G communication network to transmit information. An intelligent wireless channel allocation algorithm for HAPS 5G massive MIMO (multiple-input multiple-output) system based on reinforcement learning was proposed. Q-learning reinforcement learning algorithm and the back-propagation neural network were combined, which enabled HAPS 5G massive MIMO systems autonomous learn according to the environment and allocate channel resources of the system efficiently. Agents perceive the state information in the channel environment through continuous interaction with the channel environment, learn from the environment state to the action mapping, and use the back-propagation neural network for learning training, use the neural network instead of the Q value table, and train the network with each Q update as a training example to update the evaluation function. It effectively improves the overall performance of the system.
11 October 2023 in Journal of Materials Chemistry C
The intriguing optical phenomena of photochromism and photoluminescence, when integrated into a solitary material, can offer innovative applications in optical memory devices and multimode anti-counterfeiting. However, the exploration and utilization of such materials with dynamic anti-counterfeiting still presents a substantial challenge. This study puts forth an innovative strategy to attain a convergence between the photochromic absorption peak and the photoluminescent emission peak, fundamental to the creation of high-efficiency dynamic anti-counterfeiting materials. By sintering in a reducing atmosphere, a large number of oxygen vacancies are introduced into Ba3MgSi2O8:Er3+ ceramics, which exhibits a reversible white-pink color change with excellent fatigue resistance upon alternate 330 nm illumination and thermal stimulus. Furthermore, Ba3MgSi2O8:Er3+ ceramics demonstrate a distinct fluorescence emission, transitioning from yellow-green to orange to red under 330 nm continuous irradiation. This fascinating behavior can be attributed to the fluorescence resonance energy transfer process, facilitating efficient energy transfer between the luminescent centers and coloring centers. The exceptional optical properties exhibited by Ba3MgSi2O8:Er3+ ceramics have enabled the successful implementation of dynamic anti-counterfeiting applications in this study. This advancement is anticipated to fuel endeavours towards luminescence modulation instigated by photochromism.
3 March 2023 in Science and Technology of Nuclear Installations
A deep-learning model was proposed for predicting the remaining time to automatic scram during abnormal conditions of nuclear power plants (NPPs) based on long short-term memory (LSTM) and dropout. The proposed model was trained by simulated condition data of abnormal conditions; the input of the model was the deviation of the monitoring parameters from the normal operating state, and the output was the remaining time from the current moment to the upcoming reactor trip. The predicted remaining time to the reactor trip decreases with the development of abnormal conditions; thus, the output of the proposed model generates a predicted countdown to the reactor trip. The proposed prediction model showed better prediction performance than the Elman neural network model in the experiments but encountered an overfitting problem for testing data containing noise. Therefore, dropout was applied to further improve the generalization ability of the prediction model based on LSTM. The proposed automatic scram prediction model can provide NPP operators with an alert to the automatic scram during abnormal conditions.