Most recent issue published online in the International Journal of Ad Hoc and Ubiquitous Computing.
International Journal of Ad Hoc and Ubiquitous Computing
http://www.inderscience.com/browse/index.php?journalID=145&year=2024&vol=45&issue=2
Inderscience Publishers Ltd
en-uk
support@inderscience.com
International Journal of Ad Hoc and Ubiquitous Computing
1743-8225
1743-8233
© 2024 Inderscience Enterprises Ltd.
© 2024 Inderscience Publishers Ltd
editor@inderscience.com
International Journal of Ad Hoc and Ubiquitous Computing
https://www.inderscience.com/images/files/coverImgs/ijahuc_scoverijahuc.jpg
http://www.inderscience.com/browse/index.php?journalID=145&year=2024&vol=45&issue=2
-
Consumer IoT device deployment optimisation through deep learning: a CNN-LSTM solution for traffic classification and service identification
http://www.inderscience.com/link.php?id=136819
The internet of things (IoT) has revolutionised our world, connecting devices and creating a more intelligent and interconnected environment. However, managing and utilising the vast amount of data generated by these devices is a major challenge. To address this, we propose a novel approach in this article that combines convolutional neural networks (CNNs) with long short-term memory (LSTM) networks to optimise IoT device deployment. The process involves data preparation, defining and training a deep learning model on preprocessed data, and using the trained model to categorise network traffic from IoT devices. Our experimental results demonstrate exceptional accuracy of over 99.99%. We evaluate the model's performance using classification metrics and compare it with commonly used traffic predictive models. Additionally, our approach provides valuable insights into the services offered by IoT devices by analysing their traffic patterns, distinguishing between monitoring, home automation, and appliance usage.
Consumer IoT device deployment optimisation through deep learning: a CNN-LSTM solution for traffic classification and service identification
Imane Chakour; Sajida Mhammedi; Cherki Daoui; Mohamed Baslam
International Journal of Ad Hoc and Ubiquitous Computing, Vol. 45, No. 2 (2024) pp. 65 - 81
The internet of things (IoT) has revolutionised our world, connecting devices and creating a more intelligent and interconnected environment. However, managing and utilising the vast amount of data generated by these devices is a major challenge. To address this, we propose a novel approach in this article that combines convolutional neural networks (CNNs) with long short-term memory (LSTM) networks to optimise IoT device deployment. The process involves data preparation, defining and training a deep learning model on preprocessed data, and using the trained model to categorise network traffic from IoT devices. Our experimental results demonstrate exceptional accuracy of over 99.99%. We evaluate the model's performance using classification metrics and compare it with commonly used traffic predictive models. Additionally, our approach provides valuable insights into the services offered by IoT devices by analysing their traffic patterns, distinguishing between monitoring, home automation, and appliance usage.]]>
10.1504/IJAHUC.2024.136819
International Journal of Ad Hoc and Ubiquitous Computing, Vol. 45, No. 2 (2024) pp. 65 - 81
Imane Chakour
Sajida Mhammedi
Cherki Daoui
Mohamed Baslam
Faculty of Science and Technology, Sultan Moulay Slimane University, B.P. 523, Beni Mellal, Morocco ' National School of Applied Sciences, Sultan Moulay Slimane University, B.P. 523, Beni Mellal, Morocco ' Faculty of Science and Technology, Sultan Moulay Slimane University, B.P. 523, Beni Mellal, Morocco ' Faculty of Science and Technology, Sultan Moulay Slimane University, B.P. 523, Beni Mellal, Morocco
internet of things
IoT
consumer IoT devices
convolutional neural network
CNN
long short-term memory
LSTM
IoT traffic classification
traffic analysis
2024-02-22T23:20:50-05:00
Copyright © 2024 Inderscience Enterprises Ltd.
45
2
65
81
2024-02-22T23:20:50-05:00
-
Lightweight and personalised e-commerce recommendation based on collaborative filtering and LSH
http://www.inderscience.com/link.php?id=136826
Nowadays, e-commerce has become one of the most popular shopping ways for worldwide customers especially after the outbreak of COVID-19 worldwide. To aid the scientific shopping decision-makings of customers, collaborative filtering is often used to discover similar customers as well as their common shopping preferences. However, traditional collaborative filtering methods often need to read massive shopping records of customers, which usually consumes much time for discovering the customer preferences and consequently, leads to a slow response and decreases customers' shopping quality of experiences. Moreover, traditional collaborative filtering methods cannot always guarantee to discover similar customers as well as their common shopping preferences especially when different customers share few commonly-bought commodities. Motivated by the above two limitations, locality-sensitive hashing used widely in information retrieval domain is recruited in this paper to aid e-commerce platforms to make accurate and scientific shopping decisions for the customers of the platforms. The advantage of our solution is that it can help to improve the response efficiency of e-commerce platforms and provide lightweight and personalised e-commerce recommendation strategies especially when the shopping records of customers are both massive and sparse. We prove the innovations of our algorithm with multiple sets of experiments.
Lightweight and personalised e-commerce recommendation based on collaborative filtering and LSH
Dejuan Li; James A. Esquivel
International Journal of Ad Hoc and Ubiquitous Computing, Vol. 45, No. 2 (2024) pp. 82 - 91
Nowadays, e-commerce has become one of the most popular shopping ways for worldwide customers especially after the outbreak of COVID-19 worldwide. To aid the scientific shopping decision-makings of customers, collaborative filtering is often used to discover similar customers as well as their common shopping preferences. However, traditional collaborative filtering methods often need to read massive shopping records of customers, which usually consumes much time for discovering the customer preferences and consequently, leads to a slow response and decreases customers' shopping quality of experiences. Moreover, traditional collaborative filtering methods cannot always guarantee to discover similar customers as well as their common shopping preferences especially when different customers share few commonly-bought commodities. Motivated by the above two limitations, locality-sensitive hashing used widely in information retrieval domain is recruited in this paper to aid e-commerce platforms to make accurate and scientific shopping decisions for the customers of the platforms. The advantage of our solution is that it can help to improve the response efficiency of e-commerce platforms and provide lightweight and personalised e-commerce recommendation strategies especially when the shopping records of customers are both massive and sparse. We prove the innovations of our algorithm with multiple sets of experiments.]]>
10.1504/IJAHUC.2024.136826
International Journal of Ad Hoc and Ubiquitous Computing, Vol. 45, No. 2 (2024) pp. 82 - 91
Dejuan Li
James A. Esquivel
Graduate School, Angeles University Foundation, Angeles City, Philippines; Shandong Provincial University Laboratory for Protected Horticulture, Weifang University of Science and Technology, Weifang 262700, China ' Graduate School, Angeles University Foundation, Angeles City, Philippines
e-commerce recommender system
lightweight
personalisation
collaborative filtering
locality-sensitive hashing
2024-02-22T23:20:50-05:00
Copyright © 2024 Inderscience Enterprises Ltd.
45
2
82
91
2024-02-22T23:20:50-05:00
-
Performance evaluation of strapdown inertial navigation and Beidou satellite navigation system based on intelligent image processing technology
http://www.inderscience.com/link.php?id=136824
This paper proposes image processing technology and Beidou/SINS compact integrated navigation system, and analyses its performance. The experimental results show that in terms of attitude angle accuracy, the attitude data output by the integrated navigation system in the static state is stable, and the root mean square error is less than 0.6°. In terms of navigation performance, the static positioning error of the integrated navigation system does not exceed 3.5 m, the rotation speed does not exceed 0.12 m/s, and the accuracy is high. Compared with the Beidou satellite, the navigation data consistency of the integrated navigation system is better, the positioning accuracy is less than 5.5 m, and the speed deviation is less than 0.3 m/s. It can be seen that the Beidou/SINS compact integrated navigation system developed in this paper can achieve high precision under dynamic conditions.
Performance evaluation of strapdown inertial navigation and Beidou satellite navigation system based on intelligent image processing technology
Jianzhong Wang; Lijun Huang
International Journal of Ad Hoc and Ubiquitous Computing, Vol. 45, No. 2 (2024) pp. 92 - 99
This paper proposes image processing technology and Beidou/SINS compact integrated navigation system, and analyses its performance. The experimental results show that in terms of attitude angle accuracy, the attitude data output by the integrated navigation system in the static state is stable, and the root mean square error is less than 0.6°. In terms of navigation performance, the static positioning error of the integrated navigation system does not exceed 3.5 m, the rotation speed does not exceed 0.12 m/s, and the accuracy is high. Compared with the Beidou satellite, the navigation data consistency of the integrated navigation system is better, the positioning accuracy is less than 5.5 m, and the speed deviation is less than 0.3 m/s. It can be seen that the Beidou/SINS compact integrated navigation system developed in this paper can achieve high precision under dynamic conditions.]]>
10.1504/IJAHUC.2024.136824
International Journal of Ad Hoc and Ubiquitous Computing, Vol. 45, No. 2 (2024) pp. 92 - 99
Jianzhong Wang
Lijun Huang
School of Artificial Intelligence, Chongqing Youth Vocational and Technical College, Chongqing, 400712, Chongqing, China ' School of Artificial Intelligence, Chongqing Youth Vocational and Technical College, Chongqing, 400712, Chongqing, China
Beidou satellite navigation
image processing technology
strapdown inertial navigation
compact integrated navigation
digital image processing
2024-02-22T23:20:50-05:00
Copyright © 2024 Inderscience Enterprises Ltd.
45
2
92
99
2024-02-22T23:20:50-05:00
-
Cryptographic analysis and construction of complete permutations using a recursive approach
http://www.inderscience.com/link.php?id=136833
This paper proposes a novel recursive approach for constructing generalised complete permutations over finite fields and analyses their cryptographic properties. The method can generate complete, strong complete, <i>K</i>-strong complete, and <i>S</i>-complete permutations by recursively combining component permutations and arbitrary mappings. Compared to prior recursive techniques, the approach provides larger classes of permutations with superior cryptographic strengths like higher algebraic degree, etc. Algebraic degree, nonlinearity, differential/boomerang uniformity, and other properties are investigated. For instance, constructed complete permutations can achieve optimal algebraic degrees to resist structural attacks. The analysis also derives tight lower and upper bounds on (<i>c</i>-)differential uniformity, (<i>c</i>-)boomerang uniformity and nonlinearity. Results demonstrate improved algebraic degree, differential and boomerang uniformities over previous recursive methods. Overall, this work makes significant contributions around complete permutation generation and analysis in cryptography.
Cryptographic analysis and construction of complete permutations using a recursive approach
Shuang Xiang; Yingqi Tang; Yan Tong; Jinzhou Huang
International Journal of Ad Hoc and Ubiquitous Computing, Vol. 45, No. 2 (2024) pp. 100 - 122
This paper proposes a novel recursive approach for constructing generalised complete permutations over finite fields and analyses their cryptographic properties. The method can generate complete, strong complete, <i>K</i>-strong complete, and <i>S</i>-complete permutations by recursively combining component permutations and arbitrary mappings. Compared to prior recursive techniques, the approach provides larger classes of permutations with superior cryptographic strengths like higher algebraic degree, etc. Algebraic degree, nonlinearity, differential/boomerang uniformity, and other properties are investigated. For instance, constructed complete permutations can achieve optimal algebraic degrees to resist structural attacks. The analysis also derives tight lower and upper bounds on (<i>c</i>-)differential uniformity, (<i>c</i>-)boomerang uniformity and nonlinearity. Results demonstrate improved algebraic degree, differential and boomerang uniformities over previous recursive methods. Overall, this work makes significant contributions around complete permutation generation and analysis in cryptography.]]>
10.1504/IJAHUC.2024.136833
International Journal of Ad Hoc and Ubiquitous Computing, Vol. 45, No. 2 (2024) pp. 100 - 122
Shuang Xiang
Yingqi Tang
Yan Tong
Jinzhou Huang
Data and Information Technology Engineering Center, ChangJiang Water Resources and Hydropower Development Group Co., Ltd., Wuhan, 430062, China; College of Informatics, Huazhong Agricultural University, Wuhan, 430070, China; School of Computer and Information Engineering, Hubei Normal University, Huangshi, 435002, China ' Data and Information Technology Engineering Center, ChangJiang Water Resources and Hydropower Development Group Co., Ltd., Wuhan, 430062, China ' College of Informatics, Huazhong Agricultural University, Wuhan, 430070, China; Hubei Key Laboratory of Applied Mathematics, Hubei University, Wuhan, 430062, China ' School of Computer Engineering, Hubei University of Arts and Science, Xiangyang, 441053, China
cryptography
complete permutation
algebraic degree
(c-)differential uniformity
(c-)boomerang uniformity
nonlinearity
2024-02-22T23:20:50-05:00
Copyright © 2024 Inderscience Enterprises Ltd.
45
2
100
122
2024-02-22T23:20:50-05:00
-
Detection of image recognition forgery technology under machine vision
http://www.inderscience.com/link.php?id=136852
Traditional image forgery detection methods have difficulty keeping up with the development of forgery technology and cannot effectively detect complex forged images. With the comprehensive use of modern computer vision and deep learning technology, this paper provides a new solution for the complex image forgery problem on the internet. By improving the Xception model, the accuracy of forged image detection can be improved, and the spread of forged images can be effectively identified and prevented. First, image forgery was carried out via forms such as the generative adversarial network (GAN), the cascaded refinement network (CRN), and implicit maximum likelihood estimation (IMLE). Second, this paper preprocessed the collected forged image dataset, extracted texture features, edge features, colour features and local discontinuities of the image and performed feature-level fusion of different types of features. Last, an improved Xception model was utilised to detect forged images. The experimental results showed that homologous data detection accuracy and single heterosource data detection accuracy were 97.8% and 86.9%, respectively, in the ProGAN dataset. The improved Xception model can effectively improve the accuracy of forged image detection and provide an effective detection method for complex forged images on the internet.
Detection of image recognition forgery technology under machine vision
Yong Liu; Yinjie Zhang; Zonghui Wang; Ruosi Cheng; Xu Zhao; Baolan Shi
International Journal of Ad Hoc and Ubiquitous Computing, Vol. 45, No. 2 (2024) pp. 123 - 134
Traditional image forgery detection methods have difficulty keeping up with the development of forgery technology and cannot effectively detect complex forged images. With the comprehensive use of modern computer vision and deep learning technology, this paper provides a new solution for the complex image forgery problem on the internet. By improving the Xception model, the accuracy of forged image detection can be improved, and the spread of forged images can be effectively identified and prevented. First, image forgery was carried out via forms such as the generative adversarial network (GAN), the cascaded refinement network (CRN), and implicit maximum likelihood estimation (IMLE). Second, this paper preprocessed the collected forged image dataset, extracted texture features, edge features, colour features and local discontinuities of the image and performed feature-level fusion of different types of features. Last, an improved Xception model was utilised to detect forged images. The experimental results showed that homologous data detection accuracy and single heterosource data detection accuracy were 97.8% and 86.9%, respectively, in the ProGAN dataset. The improved Xception model can effectively improve the accuracy of forged image detection and provide an effective detection method for complex forged images on the internet.]]>
10.1504/IJAHUC.2024.136852
International Journal of Ad Hoc and Ubiquitous Computing, Vol. 45, No. 2 (2024) pp. 123 - 134
Yong Liu
Yinjie Zhang
Zonghui Wang
Ruosi Cheng
Xu Zhao
Baolan Shi
College of Cyberspace Security, PLA Strategic Support Force Information Engineering University, ZhengZhou 450001, Henan, China ' Research Institute of Intelligent Networks, Zhejiang Lab, Hangzhou, Zhejiang 311121, China ' College of Computer Science and Technology, Zhejiang University, Hangzhou 310027, Zhejiang, China ' College of Cyberspace Security, PLA Strategic Support Force Information Engineering University, ZhengZhou 450001, Henan, China ' College of Cyberspace Security, PLA Strategic Support Force Information Engineering University, ZhengZhou 450001, Henan, China ' College of Engineering and Applied Science, University of Colorado Boulder, Boulder 80309, Colorado, USA
image forgery
forgery detection
internet images
improved Xception model
machine vision
deep learning
2024-02-22T23:20:50-05:00
Copyright © 2024 Inderscience Enterprises Ltd.
45
2
123
134
2024-02-22T23:20:50-05:00
-
Detection of deepfake technology in images and videos
http://www.inderscience.com/link.php?id=136851
In response to the low accuracy, weak generalisation, and insufficient consideration of cross-dataset detection in deepfake images and videos, this article adopted the miniXception and long short-term memory (LSTM) combination model to analyse deepfake images and videos. First, the miniXception model was adopted as the backbone network to fully extract spatial features. Secondly, by using LSTM to extract temporal features between two frames, this paper introduces temporal and spatial attention mechanisms after the convolutional layer to better capture long-distance dependencies in the sequence and improve the detection accuracy of the model. Last, cross-dataset training and testing were conducted using the same database and transfer learning method. Focal loss was employed as the loss function in the training model stage to balance the samples and improve the generalisation of the model. The experimental results showed that the detection accuracy on the FaceSwap dataset reached 99.05%, which was 0.39% higher than the convolutional neural network-gated recurrent unit (CNN-GRU) and that the model parameter quantity only needed 10.01 MB, improving the generalisation ability and detection accuracy of the model.
Detection of deepfake technology in images and videos
Yong Liu; Tianning Sun; Zonghui Wang; Xu Zhao; Ruosi Cheng; Baolan Shi
International Journal of Ad Hoc and Ubiquitous Computing, Vol. 45, No. 2 (2024) pp. 135 - 148
In response to the low accuracy, weak generalisation, and insufficient consideration of cross-dataset detection in deepfake images and videos, this article adopted the miniXception and long short-term memory (LSTM) combination model to analyse deepfake images and videos. First, the miniXception model was adopted as the backbone network to fully extract spatial features. Secondly, by using LSTM to extract temporal features between two frames, this paper introduces temporal and spatial attention mechanisms after the convolutional layer to better capture long-distance dependencies in the sequence and improve the detection accuracy of the model. Last, cross-dataset training and testing were conducted using the same database and transfer learning method. Focal loss was employed as the loss function in the training model stage to balance the samples and improve the generalisation of the model. The experimental results showed that the detection accuracy on the FaceSwap dataset reached 99.05%, which was 0.39% higher than the convolutional neural network-gated recurrent unit (CNN-GRU) and that the model parameter quantity only needed 10.01 MB, improving the generalisation ability and detection accuracy of the model.]]>
10.1504/IJAHUC.2024.136851
International Journal of Ad Hoc and Ubiquitous Computing, Vol. 45, No. 2 (2024) pp. 135 - 148
Yong Liu
Tianning Sun
Zonghui Wang
Xu Zhao
Ruosi Cheng
Baolan Shi
College of Cyberspace Security, PLA Strategic Support Force Information Engineering University, ZhengZhou 450001, Henan, China ' Research Institute of Intelligent Networks, Zhejiang Lab, Hangzhou, Zhejiang 311121, China ' College of Computer Science and Technology, Zhejiang University, Hangzhou 310027, Zhejiang, China ' College of Cyberspace Security, PLA Strategic Support Force Information Engineering University, ZhengZhou 450001, Henan, China ' College of Cyberspace Security, PLA Strategic Support Force Information Engineering University, ZhengZhou 450001, Henan, China ' College of Engineering and Applied Science, University of Colorado Boulder, Boulder 80309, Colorado, USA
deepfake technology
fake image and video detection
transfer learning
parameter quantity
detection across datasets
2024-02-22T23:20:50-05:00
Copyright © 2024 Inderscience Enterprises Ltd.
45
2
135
148
2024-02-22T23:20:50-05:00