Journal of High-Frequency Communication Technologies

Deployment of Network Intrusion Detection System on a Local Area Network

Bello Alhaji Buhari — 2025-04-01

Strong security solutions are becoming crucial, especially in academic and institutional networks, due to the increasing frequency and sophistication of cyberattacks. This research is the implementation of the Snort Network Intrusion Detection System (NIDS) on the Local Area Network (LAN) segment. Its main goal is to create a system that can monitor, identify and notify network managers of any security risks within the LAN. The study entails setting up Snort on a server connected to the LAN network for the identification of malicious activity and other intrusion attempts. The use of the Snort tool to improve network security is demonstrated when properly configured. The results showed the Snort-ids system recording 106 TCP, 0 UDP, and 271 ICMP alerts. So, Snort can assist safeguarding the LAN’s network architecture from both internal and external threats through the offer of real-time monitoring and alarms. The results of the system evaluation showed a false positive rate of 13.23% and a false negative rate of 86.7%.

Elimination of Noise in Speech Signals Utilizing METS Deep Convolutional Neural Networks

Jung-Hua Wang — 2025-04-16

In the realm of digital signal processing, speech enhancement plays a crucial role in applications such as teleconferencing, voice recognition, and biometric systems. Noise and distortions significantly affect speech quality, necessitating advanced enhancement techniques. This paper proposes an optimized deep convolutional neural network (CNN)-based speech enhancement method, integrating signal subspace searching and the minimum error and time-spectral estimator (METS). The model is trained and evaluated using the LJ Speech Dataset, augmented with various noise conditions. Experimental results demonstrate that the proposed method achieves a PESQ of 3.7, STOI of 0.92, and SNR improvement of 12.3 dB, outperforming traditional and deep learning-based methods such as Spectral Subtraction, Wiener Filtering, MMSE, SEGAN, and DCRN. The integration of METS refines the spectral estimation, while CNN effectively reconstructs speech features, leading to better intelligibility and reduced spectral distortion. Future research will focus on real-time processing and adaptive noise handling, ensuring robust speech enhancement for diverse applications.

Integrating Data Mining with Transcranial Focused Ultrasound to Refine Neuralgia Treatment Strategies

Mascella Raffaele — 2025-04-22

The heterogeneity in the causes and responses to pain in patients makes neuralgia, a condition defined by persistent severe nerve pain, a challenging treatment problem. However, inconsistent therapeutic results and long patient suffering are common results of traditional therapy procedures that depend on generic methodologies. This research presents a technological framework that combines data mining and transcranial focused ultrasound (tFUS) to improve strategies for the treatment of neuralgia, with the aim of overcoming these limitations. The first step of the proposed system is to gather multimodal datasets that have been preprocessed using normalization, noise reduction, and feature extraction methods. These data sets include patient-reported pain ratings, clinical history, and brain imaging (fMRI, EEG). Next, data mining algorithms such as clustering and classification are used to find patterns of brain activity and pain attributes. Dimensionality reduction methods such as variational autoencoders (VAEs) make complex associations easier to observe and understand. Optimal tFUS parameters frequency, intensity, and focal depth are predicted for individual patients using machine learning models (MLM), such as gradient-boosted decision trees (GBDT) and Random Forests (RF). Based on the biomarkers detected, these predictions direct the deployment of tFUS procedures to a specific area of the brain. During treatment, real-time neural feedback systems track patients’ reactions, allowing adaptive alterations to boost effectiveness. Incorporating post-treatment results into an iterative feedback loop allows the continued improvement of prediction models for future sessions. An increase in pain reduction measures was observed compared to traditional techniques, greater neuroplasticity and fewer side effects when the framework was evaluated on data sets from patients with neuralgia. The proposed method achieves neuroplasticity by 97.86% and 97.14%, side effects of 34.61% and 37.83%, pain reduction of 98.64% and 96.36%, effectiveness and patient safety of 97.04% and 98.67%.