The Internet of Things (IoT) has become increasingly practical in applications such as smart homes, autonomous vehicles, and environmental monitoring. However, this rapid expansion has led to significant cybersecurity threats. Detecting these threats is critical, and while machine learning techniques are valuable, they struggle with high-dimensional data. Feature selection helps by reducing computational costs while maintaining model generalization. Selecting the most effective feature selection method is a crucial task. This research addresses this gap by testing five feature selection methods: Random Forest (RF), Recursive Feature Elimination (RFE), Logistic Regression (LR), XGBoost Regression (XGBoost), and Information Gain (IG) using the CIC-IoT 2023 dataset. It evaluates these methods when being used with five machine learning models: Decision Tree (DT), Random Forest (RF), k-Nearest Neighbors (k-NN), Gradient Boosting (GB), and Multi-layer Perceptron (MLP) using metrics like accuracy, precision, recall, and F1-score across three datasets. The results show that RFE, especially with the RF model, achieves the highest accuracy (99.57%) with 30 features. RF is the most stable, with accuracy from 83% to 99.56%. Additionally, the 5-feature scheme is best for implementing IDS on resource-limited IoT devices, with RFE paired with the k-NN model being the optimal combination.

This paper introduces a novel and simple filtering topology for band-reject (notch) and inverse band-reject applications, utilizing two voltage-adjustable operational transconductance amplifiers. The utilization of these active devices enables the implementation of identical topologies for both band-reject and inverse band-reject transfer functions. The resulting responses are harnessed for the cascade synthesis of a specialized comb filter, capable of amplification or attenuating/rejecting specific bands shown on three sub-bands. Both the design of individual sections and the entire cascade have been experimentally verified using active devices manufactured in the TSMC 180 nm CMOS process. Measurements conducted over a range from 10 Hz to 100 kHz demonstrate the importance of the selective filtering, which is significant for various applications, particularly in acoustic, vibration, and magnetic sensing readouts. The example of peaking suppression in magnitude response of emulated environment (acoustic coupling of piezo and microphone) is shown.

Cybersecurity is a vital part of digital economies and digital governing but the discipline is suffering from a pronounced skills shortage. Nevertheless, the reasons for the inability of academia to produce enough graduates with the skills that reflect the needs of the cybersecurity industry are not well understood.

This paper introduces a new electronically controllable first-order bilinear immittance based on a single CMOS active element (voltage differencing current conveyor) and its implementation in frequency response equalization including constant phase emulation (i.e. fractional-order approximation). A cascade of current feedback operational amplifiers and grounded bilinear immittances is required for these operations. The concept brings electronic reconfigurability of parameters and simplifies also circuitry. Several application examples (equalization, fractional-order integrator/differentiator) are shown and verified experimentally with fabricated CMOS IC (C07 I2T100 0.7 μm ON Semiconductor process).

In this paper, a new non-integer-order (fractional-order) low-pass filter with independent electronic control of the order, quality factor and pole frequency is presented. The filter operates in the current mode and uses Operational Transconductance Amplifiers (OTAs) and Adjustable Current Amplifiers (ACAs) as active elements. The values of non-integer order, quality factor and pole frequency of the filter are controlled by the internal and electronically controllable parameters of the used active elements (transconductance, current gain in our case). The proposed filter allows tunability of the pole frequency in range from 43.0 to 364.0 kHz. The filter order is controlled from 1.17 to 1.75 and finally it provides possibility to control quality factor from 0.64 to 1.9. Features of the proposed circuit are verified by PSpice simulations using transistor-level simulation models of the active elements.

The paper presents a proposal of modified 2nd-order fully-differential universal frequency filter. Both the pole frequency and quality factor (controlled without disturbing each other) of the filter can be controlled by two independent parameters (i.e. dual-parameter control) in order to increase the range of their control. It is the main improvement in comparison to recent state-of-the-art. The function of the proposed filter has been verified by PSpice simulations and also by experimental measurements.

In this paper, a new realization of a fully-differential (F-D) first-order all-pass filter (APF) operating in current mode (CM) is presented. In the proposed F-D CM APF a single adjustable current amplifier (ACA) and two current followers (CFs) with non-unity gain are used as active building blocks. Considering the input intrinsic resistance of CFs as useful active filter parameter, the proposed filter employs only a floating capacitor as external passive component. The pole frequency of the proposed resistorless circuit can be tuned via mutual change of input intrinsic resistance of CFs while the filter pass-band gain by means of gain of ACA. The theoretical results are verified by SPICE simulations, where TSMC 0.18 μm level-7 SCN018 CMOS process parameters were used.

This paper discusses new quadrature oscillator consisting of two all-pass sections and inverting amplifier that is capable of generating two signals with arbitrary phase shift between them. Constrains and limits of this simple idea are discussed. Verifications of theoretical expectations were provided by Spice simulations based on active devices with TSMC 0.18 μm NMOS and PMOS technological models.