- Underwater Vehicles and Communication Systems
- Indoor and Outdoor Localization Technologies
- Underwater Acoustics Research
- Energy Harvesting in Wireless Networks
- Tea Polyphenols and Effects
- Water Quality Monitoring Technologies
- Structural Health Monitoring Techniques
- Structural Engineering and Vibration Analysis
- Food Quality and Safety Studies
- Blind Source Separation Techniques
- Power Line Inspection Robots
- Energy Efficient Wireless Sensor Networks
- Structural Load-Bearing Analysis
- UAV Applications and Optimization
- Aluminum Alloys Composites Properties
- Wireless Communication Security Techniques
- Sparse and Compressive Sensing Techniques
- Arsenic contamination and mitigation
- Security in Wireless Sensor Networks
- Marine animal studies overview
- Plant nutrient uptake and metabolism
- Advanced Fiber Optic Sensors
- Advanced MEMS and NEMS Technologies
- Geophysics and Sensor Technology
- Fault Detection and Control Systems
Xiamen University
2014-2025
Key Laboratory of Chemistry for Natural Products of Guizhou Province and Chinese Academy of Sciences
2024-2025
Guizhou University
2023-2025
University of Science and Technology Beijing
2024
North China University of Science and Technology
2022-2024
South China Botanical Garden
2023
China Railway Design Corporation (China)
2022-2023
Anhui Jianzhu University
2023
Kunming University
2023
Chinese Academy of Sciences
2023
Underwater Wireless Sensor Networks (UWSNs) are expected to support a variety of civilian and military applications. In UWSNs, Medium Access Control (MAC) protocol has attracted strong attention due its potentially large impact the overall network performance. Unlike terrestrial networks, which mainly rely on radio waves for communications, UWSNs utilize acoustic waves, pose new research challenge in design MAC protocols. To present development protocols this paper surveys current...
In this letter, we present an anti-jamming underwater transmission framework that applies reinforcement learning to control the transmit power and uses transducer mobility address jamming in acoustic networks. The deep Q-networks-based scheme can achieve optimal node without knowing model channel dynamic game. Experiments performed with transducers a non-anechoic pool show our proposed reduce bit error rate of against reactive compared Q-learning based scheme.
The natural cave located on the bank of Chishui River in Guizhou is currently only base country for brewing sauce-flavor Baijiu cave. Utilizing not enriches microbial source Baijiu, but also helps conserving land resources. This article studied communities and flavor compounds fermented grains environmental microorganisms during first second rounds heap fermentation cave-brewed (CBSB). results indicated that Thermomyces, Torulaspora, Thermoascus were dominant fungal genera fermentation,...
To improve the quality of multimedia services and stimulate secure sensing in Internet Things applications, such as healthcare traffic monitoring, mobile crowdsensing (MCS) systems must address security threats jamming, spoofing faked attacks during both information exchange processes large-scale dynamic heterogeneous networks. In this article, we investigate present ways to use deep learning (DL) methods, stacked autoencoder, neural networks, convolutional reinforcement learning, approaches...
In this letter, we propose a physical (PHY)-layer authentication framework to detect spoofing attacks in underwater sensor networks. This scheme exploits the power delay profile of acoustic channel discriminate sensors and applies reinforcement learning (RL) choose parameter without being aware network model. We an RL-based provide light-weight detection deep further improve accuracy for sinks that support learning. Experiment results show improves increases utility compared with benchmark...
Underwater sensor networks (UWSNs) are vulnerable to jamming attacks due the narrow frequency bandwidth and fast fading channels. In this paper, we propose a reinforcement learning (RL)-based antijamming relay scheme for UWSNs that enables an underwater decide whether leave heavily jammed location choose power based on state consists of bit error rate previous transmission, power, current transmit sensor, measured by node. We also deep-RL-based further improve performance node supports deep...
In this paper, we propose a reinforcement learning-based adaptive modulation and coding scheme for underwater communications; more specifically, based on the network states such as quality of service requirement sensing message, previous transmission quality, energy consumption. This applies learning to choose policy in dynamic communication system. We provide performance bound perform experiments both pool sea environments. The experimental data were collected post-processed. Compared with...
At the same time, mechanical characteristic data and vibration during action process of circuit breaker are collected in article, time reference point is set through fracture to establish threshold components. By determining position operating mechanism occurrence abnormal signals, cause malfunction can be diagnosed. The research results detect faults before leaves factory, avoiding safety hazards that may arise use breaker.
Beacon-aided autonomous underwater vehicle (AUV) localization supporting maritime surveillance applications in acoustic sensor networks selects a fixed number of beacons with constant transmit power, and thus has degradation accuracy severe channel fading position fluctuation beacons. In this paper, we propose reinforcement learning based AUV scheme to choose the their power improve energy efficiency on depth, received signal strength, selected beacon consumption. According least squares...
Abstract The article takes 13X molecular sieve as the object, studies effects of three modification methods, including calcination temperature, NaOH solution concentration, and CuO MnO 2 loading, on adsorption performance SO sieve, preliminarily explores kinetics characteristics 13X‐xwt % . results show that method loading 13X‐6wt %MnO is better, penetration time about 130 minutes, maximum penetrable capacity 57.8 mg⋅g ‐1 .The study shows R quasi‐second‐order model close to 0.99, indicating...
Energy-efficient data aggregation is important for underwater acoustic sensor networks due to its energy constrained character. In this paper, we propose a kind of energy-efficient scheme reduce communication cost and prolong network lifetime based on distributed compressed sensing theory. First, introduce model cluster-based in which spatial temporal correlations are both considered. Second, two schemes, namely, BUTM-DCS (block upper triangular matrix DCS) BDM-DCS diagonal DCS), proposed...
UAVs are widely used in transmission line inspections because of their simple operation, safety and reliability, over-the-horizon flight characteristics, while video surveillance equipment can monitor fixed sections around the clock. The purpose this paper is to study UAV inspection technology its application. existing application status analyzed researched, operation process summarized. According content scope, select appropriate model for inspection. characteristics image information...
Litopenaeus vannamei, as the main target of shrimp farming in China, has a great market prospect and high economic value. However, domestic vannamei feeding efficiency are low, mainly through use hand tools by farmers to complete feed throwing work, cost is high. Therefore, view density low visibility culture ponds aquaculture industry, it impossible judge quantity time visual way. In this study, based on analysis acoustic signal characteristics accurate intelligent control amount proposed,...
In this letter, we proposed an efficient underwater acoustic (UWA) image communication algorithm based on reinforcement learning which can improve the quality while reduce energy consumption and time delay in fast variant UWA channels. algorithm, received other performance parameters are estimated at sink continuously then feedback to sensor by independent channel order avoid bandwidth loss caused large delay. At sensor, most suitable modulation coding method is chosen maximize a special...
At present, in the power industry, there has always been a demand for intelligent computing and real-time feedback on edge side using embedded devices. Due to number of parameters, calculations, memory usage deep learning model, its deployment devices is severely affected. Based this, this paper proposes lightweight object detection network based coordinate attention. The YOLOv5, decouples large convolution kernels channel space, reduces parameters kernel calculation amount operations,...
Spoofing detection is essential in underwater sensor networks (UWSNs), especially for safety-critical applications. Because of the high bit error rates, low bandwidth and large delay acoustic channels, it challenging to detect spoofing attacks UWSNs. In this paper, we propose strategies based on reinforcement learning. The interactions between a surface station an spoofer are formulated as zero-sum game. Nash equilibrium (NE) game UWSNs derived existing condition unique NE provided. A method...
Underwater localization is an important and fundamental part of the Acoustic Networks (UANs). The problem we must face that radio waves optical are heavily attenuated underwater, so acoustic signals become most common form communication. However, speed sound wave not constant will be affected by environmental factors. inaccurate velocity have a serious impact on traditional positioning results. Therefore, symmetry correction based least square estimation (SC-LSE) proposed in this paper....
This paper proposes the use of genetic algorithm to select an optimal feature set for distinguishing computer graphics from digital photographic images. Our previously developed approach has derived a 234-D vector each test image in HSV color space. The statistical moments characteristic functions and its wavelet subbands were selected as features. Since it is possible that only certain features contain significant information with respect classification, insignificant contributions...
Localization plays a more and important role in underwater wireless sensor networks (UWSNs). But the large-scale UWSNs, localization algorithm can't be realized for continuous packet collision. Therefore, we need to consider impact of MAC protocol on positioning algorithm. First, this paper proposes multi-layer model based network architecture. Then according non-synchronous scheme, analyze reason collisions propose variable interval ALOHA (VI-ALOHA) Poisson distribution. The VI-ALOHA...
Adaptive modulation techniques which use different methods to match time-varying channel-state information (CSI) are widely used improve the robustness of underwater acoustic communication links. We propose a novel Q-learning-based adaptive modulation-switching strategy. During communication, CSI is estimated from handshake signals or training sequences, etc., in each block at receiver, and then sent back transmitter feedback channel. The adaptively selects an appropriate method via...
Inorganic arsenic is a well-known environmental toxicant and carcinogen, there overwhelming evidence for an association between this metalloid poisoning hepatic diseases. However, the biological mechanism involved not well characterized. In present study, we probed how inorganic modulates polarization of macrophages, as roles PTEN-induced kinase 1 (PINK1)/Parkin-mediated mitophagy participates in regulating metalloid-mediated macrophage polarization. Our results indicate that acute exposure...
A high purity Al2O3 ceramic (HPAOC) metallizing strategy was developed via gradient coating process of pastes with different ratios Mo to manganese glass (MnG) contents, improve the wettability and reactivity metallized layer (ML) substrate secondary or sealed metals. Self-made HPAOC samples firstly coated by a paste lower proportion Mo:MnG superposed higher were fired at 1450°C in hydrogen atmosphere. The crystal phase structure, microstructure element distribution wer characterized XRD,...