Search

The software-defined wireless sensor network (SDWSN) has the potential to improve flexibility, scalability, and network performance, but security and quality of service (QoS) are major challenges due to attackers, poor network management, and inefficient route selection. Several existing works for intrusion detection had drawbacks like poor security, inefficient network management, higher energy consumption and latency, and lesser throughput. A modified honeycomb structure-based intrusion detection system for SDWSN is proposed to address these challenges, which includes secure authentication using the 3D cube algorithm, modified honeycomb-based network partitioning, clustering, reinforcement learning-based intelligent routing with a transfer learning-based deep Q network (TLDQN), and a hybrid intrusion detection system. The latter detects malicious nodes using a driver training-based optimization (DTO) algorithm and intrusions with a bidirectional generative adversarial network (Bi-GAN). The results show that the proposed system outperforms existing solutions in terms of security, network performance, and efficiency. The simulation of this research is conducted by NS-3.26 Network Simulator, and the performances are evaluated based on various performance metrics (with respect to the total number of nodes) like energy consumption, latency, throughput, packet delivery ratio, network lifetime, computation overhead, detection accuracy, packet drop ratio, and control overhead, which proved that the proposed work achieves superior performance compared to existing works. The evaluation also includes a total simulation period during which the system’s real-time performance was conducted. Time-based metrics such as precision, recall, and F1-score, as well as confusion matrices, are utilized to analyze the system’s effectiveness in real-time in response to dynamic network threats.

The convolution process in the Capsule Network method can result in a loss of spatial data from the Electroencephalogram signal, despite its ability to characterize spatial information from Electroencephalogram signals. Therefore, this study applied the Continuous Capsule Network method to overcome problems associated with emotion recognition based on Electroencephalogram signals using the optimal architecture of the (1) 1st, 2nd, 3rd, and 4th Continuous Convolution layers with values of 64, 128, 256, and 64, respectively, and (2) kernel sizes of 2×2×4, 2×2×64, and 2×2×128 for the 1st, 2nd, and 3rd Continuous Convolution layers, and 1×1×256 for the 4th. Several methods were also used to support the Continuous Capsule Network process, such as the Differential Entropy and 3D Cube methods for the feature extraction and representation processes. These methods were chosen based on their ability to characterize spatial and low-frequency information from Electroencephalogram signals. By testing the DEAP dataset, these proposed methods achieved accuracies of 91.35, 93.67, and 92.82% for the four categories of emotions, two categories of arousal, and valence, respectively. Furthermore, on the DREAMER dataset, these proposed methods achieved accuracies of 94.23, 96.66, and 96.05% for the four categories of emotions, the two categories of arousal, and valence, respectively. Finally, on the AMIGOS dataset, these proposed methods achieved accuracies of 96.20, 97.96, and 97.32% for the four categories of emotions, the two categories of arousal, and valence, respectively. Doi: 10.28991/ESJ-2023-07-01-09 Full Text: PDF

Participants' emotional reactions are strongly influenced by several factors such as personality traits, intellectual abilities, and gender. Several studies have examined the baseline reduction approach for emotion recognition using electroencephalogram signal patterns containing external and internal interferences, which prevented it from representing participants’ neutral state. Therefore, this study proposes two solutions to overcome this problem. Firstly, it offers a modified weighted mean filter method to eliminate the interference of the electroencephalogram baseline signal. Secondly, it determines an appropriate baseline reduction method to characterize emotional reactions after the smoothing process. Data collected from four scenarios conducted on three datasets was used to reduce the interference and amplitude of the electroencephalogram signals. The result showed that the smoothing process can eliminate interference and lower the signal's amplitude. Based on the three baseline reduction methods, the Relative Difference method is appropriate for characterizing emotional reactions in different electroencephalogram signal patterns and has higher accuracy. Based on testing on the DEAP dataset, these proposed methods achieved accuracies of 97.14, 99.70, and 96.70% for the four categories of emotions, the two categories of arousal, and the two categories of valence, respectively. Furthermore, on the DREAMER dataset, these proposed methods achieved accuracies of 89.71, 97.63, and 96.58% for the four categories of emotions, the two categories of arousal, and the two categories of valence, respectively. Finally, on the AMIGOS dataset, these proposed methods achieved accuracies of 99.59, 98.20, and 99.96% for the four categories of emotions, the two categories of arousal, and the two categories of valence, respectively. Doi: 10.28991/ESJ-2022-06-06-03 Full Text: PDF

In accordance with the principle that the cube graph can be obtained by superimposing planes, a multiple-image encryption scheme based on hyperchaotic map and 3D cube is designed in this paper. Multiple images are segmented by columns, and a regular cube is obtained by stacking multiple fixed-size planes. Each stacked plane is considered as a stacked plane of the cube along the z-axis direction. Firstly, each tangent plane is rotated by different angles in the counterclockwise direction and then DNA encoded, followed by face swap, row swap, and column swap in turn. Then, the DNA-encoded image cube is subjected to DNA addition with the chaotic cube. Finally, two DNA sequences are randomly selected for DNA mutation. The mutated cube is decoded into a digital cube by DNA and the multiple-image encryption algorithm is completed. Theoretically, the cube used in the encryption process can be infinitely large, which means that the number and the size of 2D images that can be encrypted can be unrestricted. Experimental simulations and performance tests prove that the proposed image encryption scheme is effective and secure.

In accordance with the principle that the cube graph can be obtained by superimposing planes, a multiple-image encryption scheme based on hyperchaotic map and 3D cube is designed in this paper. Multiple images are segmented by columns, and a regular cube is obtained by stacking multiple fixed-size planes. Each stacked plane is considered as a stacked plane of the cube along the z-axis direction. Firstly, each tangent plane is rotated by different angles in the counterclockwise direction and then DNA encoded, followed by face swap, row swap, and column swap in turn. Then, the DNA-encoded image cube is subjected to DNA addition with the chaotic cube. Finally, two DNA sequences are randomly selected for DNA mutation. The mutated cube is decoded into a digital cube by DNA and the multiple-image encryption algorithm is completed. Theoretically, the cube used in the encryption process can be infinitely large, which means that the number and the size of 2D images that can be encrypted can be unrestricted. Experimental simulations and performance tests prove that the proposed image encryption scheme is effective and secure. [ABSTRACT FROM AUTHOR]

Objective: This study aimed to compare the accuracy of rectal cancer restaging after neoadjuvant therapy with 3D CUBE sequence with 2D T2-weighted fast spin-echo (FSE) sequence., Methods: This retrospective study comprised 72 patients with rectal cancer confirmed by colonoscopy and biopsy. After neoadjuvant therapy, all patients underwent pelvic magnetic resonance imaging (MRI) examination at 1.5T MRI sequences including a single coronal 3D CUBE T2-weighted FSE sequence with 1.4 mm thickness and a 2D T2-weighted FSE sequence in the sagittal, coronal and axial planes with 5 mm thickness. The total acquisition time of the two sequences was recorded. Results were compared with postsurgical pathology (gold standard). The diagnostic accuracy was evaluated; and receiver operating characteristic (ROC) curves and the area under the curves (AUC) were calculated., Results: The T category staging accuracy of 3D T2WI and 2D T2WI was 81.9% and 72.2%, respectively, for reviewer 1 and 86.1% and 75.0% for reviewer 2. The AUC of 3D was higher than that of 2D (0.878 vs. 0.783 for reader 1 and 0.905 vs. 0.796 for reader 2; both P < 0.05) when judging whether the tumor broke through the muscle layer. There was no significant difference between 3D and 2D in judging whether lymph nodes were malignant (AUC 0.719 vs. 0.698 for reader 1 and 0.740 vs. 0.698 for reader 2; both P > 0.05). There were no significant differences in the visibility of the rectal wall layer, tumor lesion and the overall image quality (all P > 0.05). Compared with 2D sequences, the 3D sequence had shorter acquisition time and higher signal intensity ratio (both P < 0.05)., Conclusion: 3D CUBE T2-weighted sequences offer better diagnostic accuracy in rectal cancer restaging after neoadjuvant therapy when compared with 2D T2-weighted FSE sequences; it has a shorter scanning time and more versatility of orientation reconstruction., (Published by Oxford University Press and Sixth Affiliated Hospital of Sun Yat-Sen University.)