Fang Miao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Fang Miao has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
Wenjie Fan
3 shared publications- Zhao Li
College of Computer Science, Chengdu University, Chengdu 610106, China
3 shared publications - Quanhai Wang
College of Computer Science and Cyber Security, Chengdu University of Technology, Chengdu 610059, China
2 shared publications - Ahmad Yahya Dawod
International College of Digital Innovation, Chiang Mai University, Chiang Mai 50200, Thailand
1 shared publication - An Wang
College of Information Science and Technology, Chengdu University of Technology, Chengdu, China
1 shared publication - Ansheng Ye
Key Lab of Earth Exploration & Information Techniques of Ministry Education, Chengdu University of Technology, Chengdu 610059, China
1 shared publication
Latest Publications
- Open Access
- 26 August 2025
- Published by Innovative Research Publishing in International Journal of Innovative Research and Scientific Studies
- Vol. 8 (6) , 1920-1929
Research on sybil attack detection method for industrial wireless sensor networks based on CNN BiLSTM attention and K-means clustering
In Industrial Wireless Sensor Networks (IWSNs), Sybil attacks compromise network topology and reduce data reliability by forging virtual nodes, leading to degraded network performance and significantly diminished monitoring accuracy. To address these issues, this study aims to propose a high-accuracy and highly robust Sybil attack detection method to overcome the limitations of traditional detection approaches, such as low precision and difficulty in handling ambiguous probability boundaries. The research designs a collaborative detection mechanism that integrates a CNN-BiLSTM-Attention (CBSA) deep learning module with the K-means clustering algorithm. By combining "multidimensional feature extraction via deep learning + clustering-based classification boundary optimization," an end-to-end Sybil attack detection model (CBSA-Kmeans) is constructed.The specific implementation includes four parts: 1. A Convolutional Neural Network (CNN) processes the raw sensor data matrix to extract spatial local patterns and capture abnormal correlation features among nodes. 2. A Bidirectional Long Short-Term Memory network (BiLSTM) processes the feature sequences output by the CNN. The forward LSTM learns the "past-present" temporal dependencies to identify the cumulative effects of attacks, while the backward LSTM models the "present-past" temporal correlations to trace attack origins. 3. An Attention mechanism is introduced to dynamically focus on key time steps corresponding to critical attack features, generating a weighted context vector and outputting attack probability predictions. 4. The K-means clustering algorithm is employed to perform secondary partitioning on the prediction probability space output by the CBSA module. By measuring Euclidean distances, high-density attack clusters and normal data clusters are constructed to form decision regions, thereby optimizing classification boundaries.Through a progressive approach of "spatial feature extraction → temporal dependency modeling and key feature enhancement → probability space clustering optimization," the model achieves attack detection: CNN first performs preliminary spatial feature screening, BiLSTM and Attention collaboratively mine temporal attack features and highlight critical information, and finally, K-means clusters the prediction probabilities to clarify the boundaries between attack and normal data. Experimental results demonstrate that the CBSA-Kmeans model excels in IWSN Sybil attack detection tasks: it achieves a detection accuracy of 98.2% and a recall rate of 96.7%, representing an improvement of over 12% compared to traditional detection methods. Additionally, the model has minimal negative impact on network performance, increasing IWSN network throughput by 23.5% and reducing data transmission latency by 31.8%, while effectively addressing the ambiguous probability boundary issue present in traditional methods. In conclusion, the CBSA-Kmeans model achieves high-precision and highly robust detection of Sybil attacks in IWSNs through the synergistic integration of deep learning and clustering algorithms, validating the effectiveness and superiority of this collaborative detection mechanism. This method provides a practical technical solution for IWSN security protection, ensuring network topology integrity and data transmission reliability while enhancing operational efficiency and monitoring accuracy. It holds significant practical application value for ensuring the secure and stable operation of wireless sensor networks in industrial settings.
- 26 August 2025
- Published by Innovative Research Publishing in International Journal of Innovative Research and Scientific Studies
- Vol. 8 (6) , 1920-1929
Open AccessResearch on sybil attack detection method for industrial wireless sensor networks based on CNN BiLSTM attention and K-means clustering
In Industrial Wireless Sensor Networks (IWSNs), Sybil attacks compromise network topology and reduce data reliability by forging virtual nodes, leading to degraded network performance and significantly diminished monitoring accuracy. To address these issues, this study aims to propose a high-accuracy and highly robust Sybil attack detection method to overcome the limitations of traditional detection approaches, such as low precision and difficulty in handling ambiguous probability boundaries. The research designs a collaborative detection mechanism that integrates a CNN-BiLSTM-Attention (CBSA) deep learning module with the K-means clustering algorithm. By combining "multidimensional feature extraction via deep learning + clustering-based classification boundary optimization," an end-to-end Sybil attack detection model (CBSA-Kmeans) is constructed.The specific implementation includes four parts: 1. A Convolutional Neural Network (CNN) processes the raw sensor data matrix to extract spatial local patterns and capture abnormal correlation features among nodes. 2. A Bidirectional Long Short-Term Memory network (BiLSTM) processes the feature sequences output by the CNN. The forward LSTM learns the "past-present" temporal dependencies to identify the cumulative effects of attacks, while the backward LSTM models the "present-past" temporal correlations to trace attack origins. 3. An Attention mechanism is introduced to dynamically focus on key time steps corresponding to critical attack features, generating a weighted context vector and outputting attack probability predictions. 4. The K-means clustering algorithm is employed to perform secondary partitioning on the prediction probability space output by the CBSA module. By measuring Euclidean distances, high-density attack clusters and normal data clusters are constructed to form decision regions, thereby optimizing classification boundaries.Through a progressive approach of "spatial feature extraction → temporal dependency modeling and key feature enhancement → probability space clustering optimization," the model achieves attack detection: CNN first performs preliminary spatial feature screening, BiLSTM and Attention collaboratively mine temporal attack features and highlight critical information, and finally, K-means clusters the prediction probabilities to clarify the boundaries between attack and normal data. Experimental results demonstrate that the CBSA-Kmeans model excels in IWSN Sybil attack detection tasks: it achieves a detection accuracy of 98.2% and a recall rate of 96.7%, representing an improvement of over 12% compared to traditional detection methods. Additionally, the model has minimal negative impact on network performance, increasing IWSN network throughput by 23.5% and reducing data transmission latency by 31.8%, while effectively addressing the ambiguous probability boundary issue present in traditional methods. In conclusion, the CBSA-Kmeans model achieves high-precision and highly robust detection of Sybil attacks in IWSNs through the synergistic integration of deep learning and clustering algorithms, validating the effectiveness and superiority of this collaborative detection mechanism. This method provides a practical technical solution for IWSN security protection, ensuring network topology integrity and data transmission reliability while enhancing operational efficiency and monitoring accuracy. It holds significant practical application value for ensuring the secure and stable operation of wireless sensor networks in industrial settings.
- Open Access
- 10 September 2023
- Published by MDPI AG in Remote Sensing
- Vol. 15 (18) , 4451
Spatial Downscaling of Soil Moisture Based on Fusion Methods in Complex Terrains
Large-area soil moisture (SM) data with high resolution and precision are the foundation for the research and application of hydrological and meteorological models, water resource evaluation, agricultural management, and warning of geological disasters. It is still challenging to downscale SM products in complex terrains that require fine spatial details. In this study, SM data from the Soil Moisture Active and Passive (SMAP) satellite were downscaled from 36 to 1 km in the summer and autumn of 2017 in Sichuan Province, China. Genetic-algorithm-optimized backpropagation (GABP) neural network, random forest, and convolutional neural network were applied. A fusion model between SM and longitude, latitude, elevation, slope, aspect, land-cover type, land surface temperature, normalized difference vegetation index, enhanced vegetation index, evapotranspiration, day sequence, and AM/PM was established. After downscaling, the in situ information was fused through a geographical analysis combined with a spatial interpolation to improve the quality of the downscaled SM. The comparative results show that in complex terrains, the GABP neural network better captures the soil moisture variations in both time and space domains. The GDA_Kriging method is able to merge in situ information in the downscaled SM while simultaneously maintaining the dynamic range and spatial details.
- 10 September 2023
- Published by MDPI AG in Remote Sensing
- Vol. 15 (18) , 4451
Open AccessSpatial Downscaling of Soil Moisture Based on Fusion Methods in Complex Terrains
Large-area soil moisture (SM) data with high resolution and precision are the foundation for the research and application of hydrological and meteorological models, water resource evaluation, agricultural management, and warning of geological disasters. It is still challenging to downscale SM products in complex terrains that require fine spatial details. In this study, SM data from the Soil Moisture Active and Passive (SMAP) satellite were downscaled from 36 to 1 km in the summer and autumn of 2017 in Sichuan Province, China. Genetic-algorithm-optimized backpropagation (GABP) neural network, random forest, and convolutional neural network were applied. A fusion model between SM and longitude, latitude, elevation, slope, aspect, land-cover type, land surface temperature, normalized difference vegetation index, enhanced vegetation index, evapotranspiration, day sequence, and AM/PM was established. After downscaling, the in situ information was fused through a geographical analysis combined with a spatial interpolation to improve the quality of the downscaled SM. The comparative results show that in complex terrains, the GABP neural network better captures the soil moisture variations in both time and space domains. The GDA_Kriging method is able to merge in situ information in the downscaled SM while simultaneously maintaining the dynamic range and spatial details.
- Open Access
- 22 July 2023
- Published by Elsevier in Results in Physics
- Vol. 52
Embedded solitons, chaotic behavior, bifurcation of cubic-quartic and nonlinear susceptibilities with multiplicative white noise- 22 July 2023
- Published by Elsevier in Results in Physics
- Vol. 52
Open AccessEmbedded solitons, chaotic behavior, bifurcation of cubic-quartic and nonlinear susceptibilities with multiplicative white noise - Open Access
- 12 October 2022
- Published by Elsevier in Results in Physics
- Vol. 43
Phase portraits and optical soliton solutions of coupled Sasa–Satsuma model in birefringent fibers
- 12 October 2022
- Published by Elsevier in Results in Physics
- Vol. 43
Open AccessPhase portraits and optical soliton solutions of coupled Sasa–Satsuma model in birefringent fibers
- Open Access
- 19 January 2021
- Published by MDPI AG in ISPRS International Journal of Geo-Information
- Vol. 10 (1) , 39
FuNet: A Novel Road Extraction Network with Fusion of Location Data and Remote Sensing Imagery
Road semantic segmentation is unique and difficult. Road extraction from remote sensing imagery often produce fragmented road segments leading to road network disconnection due to the occlusion of trees, buildings, shadows, cloud, etc. In this paper, we propose a novel fusion network (FuNet) with fusion of remote sensing imagery and location data, which plays an important role of location data in road connectivity reasoning. A universal iteration reinforcement (IteR) module is embedded into FuNet to enhance the ability of network learning. We designed the IteR formula to repeatedly integrate original information and prediction information and designed the reinforcement loss function to control the accuracy of road prediction output. Another contribution of this paper is the use of histogram equalization data pre-processing to enhance image contrast and improve the accuracy by nearly 1%. We take the excellent D-LinkNet as the backbone network, designing experiments based on the open dataset. The experiment result shows that our method improves over the compared advanced road extraction methods, which not only increases the accuracy of road extraction, but also improves the road topological connectivity.
- 19 January 2021
- Published by MDPI AG in ISPRS International Journal of Geo-Information
- Vol. 10 (1) , 39
Open AccessFuNet: A Novel Road Extraction Network with Fusion of Location Data and Remote Sensing Imagery
Road semantic segmentation is unique and difficult. Road extraction from remote sensing imagery often produce fragmented road segments leading to road network disconnection due to the occlusion of trees, buildings, shadows, cloud, etc. In this paper, we propose a novel fusion network (FuNet) with fusion of remote sensing imagery and location data, which plays an important role of location data in road connectivity reasoning. A universal iteration reinforcement (IteR) module is embedded into FuNet to enhance the ability of network learning. We designed the IteR formula to repeatedly integrate original information and prediction information and designed the reinforcement loss function to control the accuracy of road prediction output. Another contribution of this paper is the use of histogram equalization data pre-processing to enhance image contrast and improve the accuracy by nearly 1%. We take the excellent D-LinkNet as the backbone network, designing experiments based on the open dataset. The experiment result shows that our method improves over the compared advanced road extraction methods, which not only increases the accuracy of road extraction, but also improves the road topological connectivity.
- Open Access
- 24 February 2019
- Published by MDPI AG in Mathematics
- Vol. 7 (2) , 210
More Results on the Domination Number of Cartesian Product of Two Directed Cycles
Let denote the domination number of a digraph D and let denote the Cartesian product of and , the directed cycles of length . Liu et al. obtained the exact values of for m up to 6 [Domination number of Cartesian products of directed cycles, Inform. Process. Lett. 111 (2010) 36–39]. Shao et al. determined the exact values of for [On the domination number of Cartesian product of two directed cycles, Journal of Applied Mathematics, Volume 2013, Article ID 619695]. Mollard obtained the exact values of for [M. Mollard, On domination of Cartesian product of directed cycles: Results for certain equivalence classes of lengths, Discuss. Math. Graph Theory 33(2) (2013) 387–394.]. In this paper, we extend the current known results on with m up to 21. Moreover, the exact values of with n up to 31 are determined.- 24 February 2019
- Published by MDPI AG in Mathematics
- Vol. 7 (2) , 210
Open AccessMore Results on the Domination Number of Cartesian Product of Two Directed Cycles
Let denote the domination number of a digraph D and let denote the Cartesian product of and , the directed cycles of length . Liu et al. obtained the exact values of for m up to 6 [Domination number of Cartesian products of directed cycles, Inform. Process. Lett. 111 (2010) 36–39]. Shao et al. determined the exact values of for [On the domination number of Cartesian product of two directed cycles, Journal of Applied Mathematics, Volume 2013, Article ID 619695]. Mollard obtained the exact values of for [M. Mollard, On domination of Cartesian product of directed cycles: Results for certain equivalence classes of lengths, Discuss. Math. Graph Theory 33(2) (2013) 387–394.]. In this paper, we extend the current known results on with m up to 21. Moreover, the exact values of with n up to 31 are determined. - Open Access
- 1 January 2019
- Published by Wiley in Shock and Vibration
- Vol. 2019 (1)
A Fault Diagnosis Model for Rotating Machinery Using VWC and MSFLA‐SVM Based on Vibration Signal Analysis
Fault diagnosis of rotating machinery mainly includes fault feature extraction and fault classification. Vibration signal from the operation of machinery usually could help diagnosing the operational state of equipment. Different types of fault usually have different vibrational features, which are actually the basis of fault diagnosis. This paper proposes a novel fault diagnosis model, which extracts features by combining vibration severity, dyadic wavelet energy time‐spectrum, and coefficient power spectrum of the maximum wavelet energy level (VWC) at the feature extraction stage. At the stage of fault classification, we design a support vector machine (SVM) based on the modified shuffled frog‐leaping algorithm (MSFLA) for the accurate classifying machinery fault method. Specifically, we use the MSFLA method to optimize SVM parameters. MSFLA can avoid getting trapped into local optimum, speeding up convergence, and improving classification accuracy. Finally, we evaluate our model on real rotating machinery platform, which has four different states, i.e., normal state, eccentric axle fault (EAF), bearing pedestal fault (BPF), and sealing ring wear fault (SRWF). As demonstrated by the results, the VWC method is efficient in extracting vibration signal features of rotating machinery. Based on the extracted features, we further compare our classification method with other three fault classification methods, i.e., backpropagation neural network (BPNN), artificial chemical reaction optimization algorithm (ACROA‐SVM), and SFLA‐SVM. The experiment results show that MSFLA‐SVM achieves a much higher fault classification rate than BPNN, ACROA‐SVM, and SFLA‐SVM.
- 1 January 2019
- Published by Wiley in Shock and Vibration
- Vol. 2019 (1)
Open AccessA Fault Diagnosis Model for Rotating Machinery Using VWC and MSFLA‐SVM Based on Vibration Signal Analysis
Fault diagnosis of rotating machinery mainly includes fault feature extraction and fault classification. Vibration signal from the operation of machinery usually could help diagnosing the operational state of equipment. Different types of fault usually have different vibrational features, which are actually the basis of fault diagnosis. This paper proposes a novel fault diagnosis model, which extracts features by combining vibration severity, dyadic wavelet energy time‐spectrum, and coefficient power spectrum of the maximum wavelet energy level (VWC) at the feature extraction stage. At the stage of fault classification, we design a support vector machine (SVM) based on the modified shuffled frog‐leaping algorithm (MSFLA) for the accurate classifying machinery fault method. Specifically, we use the MSFLA method to optimize SVM parameters. MSFLA can avoid getting trapped into local optimum, speeding up convergence, and improving classification accuracy. Finally, we evaluate our model on real rotating machinery platform, which has four different states, i.e., normal state, eccentric axle fault (EAF), bearing pedestal fault (BPF), and sealing ring wear fault (SRWF). As demonstrated by the results, the VWC method is efficient in extracting vibration signal features of rotating machinery. Based on the extracted features, we further compare our classification method with other three fault classification methods, i.e., backpropagation neural network (BPNN), artificial chemical reaction optimization algorithm (ACROA‐SVM), and SFLA‐SVM. The experiment results show that MSFLA‐SVM achieves a much higher fault classification rate than BPNN, ACROA‐SVM, and SFLA‐SVM.
- 2 September 2015
- Published by Springer Nature in Bulletin of Engineering Geology and the Environment
- Vol. 75 (3) , 923-944
GIS-based landslide susceptibility mapping with logistic regression, analytical hierarchy process, and combined fuzzy and support vector machine methods: a case study from Wolong Giant Panda Natural Reserve, China
- 2 September 2015
- Published by Springer Nature in Bulletin of Engineering Geology and the Environment
- Vol. 75 (3) , 923-944
GIS-based landslide susceptibility mapping with logistic regression, analytical hierarchy process, and combined fuzzy and support vector machine methods: a case study from Wolong Giant Panda Natural Reserve, China
- Open Access
- 1 January 2013
- Published by Atlantis Press SARL
Research on Multiple Spatial Data Gathering Mechanism Based on Geo-information Browser
- 1 January 2013
- Published by Atlantis Press SARL
Open AccessResearch on Multiple Spatial Data Gathering Mechanism Based on Geo-information Browser