This demo shows how to use the You Only Look Once (YOLO) version 8 framework to identify radio-frequency (RF) signals for spectrum sensing in Software-Defined-Radio (SDR) and Cognitive-Radio (CR) systems. To this purpose, a trained YOLOv8 nano framework was embedded in an IoT device based on a Raspberry Pi 5 and connected to an ADALM-PLUTO SDR board to detect and classify the activity of RF frames around the industrial, scientific, and medical (ISM) frequency band. This framework detects wireless standards such as Bluetooth and WiFi. This hardware demonstrator operates in real-time and can detect RF frames showing its potential application in SDR/CR terminals1.

This paper presents a Radio-Frequency (RF) spectrum prediction system based on a Convolutional Neural Network (CNN) architecture for image-based forecasting intended for Cognitive Radio (CR) terminals. Compared to previous approaches based on the use of time series, we propose a more efficient way for occupancy spectrum prediction based on channel availability tables (derived from spectrograms) to train a deep learning model. As a result, the neural engine is able to predict the best band available for Secondary User (SU) transmission in edge devices. As a proof of concept, a simple CR demonstrator combining a computational model for the neural engine – previously modeled and trained in Python using the Keras API – in MATLAB/SIMULINK with two Software-Defined Radio (SDR) boards has been developed. Three different table sizes were used as input for the predictor and a comparison is presented. The system performance is assessed using real over-the-air signals captured in the 2.412 GHz central frequency with 60 MHz bandwidth to validate the presented approach1.

In this paper, we analyze the use of ChatGPT in teaching random signals and processes and image processing, and we identify pros and cons compared to a classical approach. We focus on these two important engineering topics, which can be challenging for beginners. Regarding random signals and processes, we selected two examples from our learning platform dedicated to better understanding random signals and processes. The first example demonstrates why a linear transformation of normal and uniform random variables does not change the type of variables. The second example explains why the sum of two normal random variables is normal while the sum of two uniform random variables is not. Next, we consider a demo example for teaching image processing from the Image Enhancement Module: Notch Filter for the Moiré Pattern. We conclude the paper by summarizing the findings from the classical, and ChatGPT approaches in the learning process.

This work aims to correct the wide passband in a novel multistage CIC-based decimation filter, recently proposed in the literature, preserving its high aliasing rejection and low complexity. Accordingly, a low complexity compensation filter is proposed, which corrects a high passband droop introduced in the multistage decimation filter due to the increased aliasing rejection. The magnitude characteristic of the compensation filter has a sinusoidal form, and the design parameters are obtained using particle swarm optimization (PSO) and MATLAB. Next, the obtained parameters are presented in an SPT (signed power-of-two) form to get a multiplierless design. The compensated multistage filter is compared with similar filters from the literature.

This paper presents a novel multiplierless two-stage decimator for even decimation factor M. In comparison with a modified cosine-based decimator previously proposed in the literature, this decimator has an improved alias rejection in all folding bands. This improvement is achieved by introducing an expanded modified cosine filter in the first stage of the decimator. The choice of the expanded factor is crucial to achieve alias improvement. The two cases of the parity of the decimation factor of the first stage, being even or odd are discussed separately. Both cases are illustrated with examples. The efficient structures are also presented for both cases. Finally, the comparison with the cosine-based method is presented.

This paper discusses design considerations of LC-based bandpass continuous-time ΣΔ modulators intended for the analog-to-digital conversion of radio-frequency signals in software-defined-radio receivers. A top-down high-level synthesis methodology - based on the combination of undersampling techniques, programmable notch frequency and adequate selection of loop-filter coefficients - is presented to design efficient modulator architectures, which can be adapted to fulfill the requirements of diverse wireless communication standards, while keeping stability and robustness to circuit errors. Main architectureand circuit-level design issues are analyzed, considering diverse case studies based on Gm-LC resonators and different kinds of digital-to-analog converters. Time-domain simulations validate the presented approach1.

Classical decimation structures usually use comb filters at first stage due to the simplicity of comb filters. However, comb filters cannot satisfy high performance demands of state-of-the-art Sigma-Delta (ΣΔ) analog-digital converters (ADCs). Some possible solutions are comb based structures which are power and area efficient and posses an improved magnitude characteristic. The principal issues in the comb-based filter design are: power and area efficiency, high alias rejection and approximately flat passband characteristic, considering also high values of the decimation factors. In this paper we first review the new trends in ΣΔ ADCs and demands for the decimation block design. Next we review power and area efficient structures. The methods to improve the alias rejection of comb filters, as well as the methods for the compensation for the comb passband droop, and those strategies which simultaneously improve the alias rejection and the passband droop are reviewed in the following. Finally we review the multirate decimation structures.

This paper introduces a power and area efficient comb-based decimation structure, particularly suited for high values of decimation factors which are a power of two. The proposed topology has two stages, where the first stage is in a non-recursive form and the second one is in a recursive form (CIC filter). Moreover, a slight modification of the proposed decimator structure is presented in order to obtain an improved alias rejection. Simulation results are shown to validate the proposed approach.

Demo programs that illustrate concepts with visual aids, become the important tools in the field of engineering education. They help students to learn better, remember longer, and better identify the appropriate concepts. Another benefit of the demo programs is, that they can be used as a complementary tool in the traditional classes in the classroom, to visualize and improve understanding of the presented subject. To this end here is presented the demo program for teaching the principal characteristics of digital IIR (Infinite impulse response) filters: Butterwort, Chebyshev I, Chebyshev II, and Elliptic filters. The program was implemented in MATLAB using MATLAB tool makeshow which has different favorable characteristics. First it is demonstrated how the analog filters can be converted into digital IIR filters using impulse invariance and bilinear transforms. The goal is to visualize the advantages of the bilinear transform. Next goal was to visualize the comparison of different above mentioned IIR filters in terms of complexity, magnitude and phase responses. The program can be also used as a self-study tool for the classes on the fundamentals of digital signal processing at the graduate level. The student validation of the program is also included.

The method for a design of a multiplierless CIC compensation filter is presented. The proposed filter compensates the CIC passband droop in the passband region defined by the passband frequency ωp=π/2M. Using a multirate identity this filter can be moved to a low rate thus becoming a second order filter. In the proposed filter design, the maximum value of the passband deviation is limited to be less than 0.4 dB.

The paper presents the design of complex allpass filters that satisfy a desired degree of flatness and a desired phase at any specified frequency point. The set of linear equations are derived based on the specifications of the flatness and the values of the phase at the given frequency points. The filter coefficients are obtained by solving this set of equations.

A new structure for the realization of a comb decimation filter with a sharpened magnitude response is advanced. The proposed structure consists of two main sections: a comb section and a sharpening comb section with the latter section operating at a lower rate than the high input rate. Using a polyphase decomposition, the sub-filters of the first section can also be operated at this lower rate. The improved magnitude response has been obtained by using the filter sharpening approach of J.F. Kaiser and R.W. Hamming (see IEEE Trans. on Acoustics, Speech, and Sig. Process., vol.ASSP25, p.415-22, 1977). The proposed filter has much less passband droop and better attenuation than the equivalent comb filter.

We propose an efficient sharpening of a CIC decimation filter for an even decimation factor. The proposed structure consists of two main sections: a section composed of a cascade of the first-order moving average filters, and a sharpening filter section. The proposed decomposition scheme allows a sharpening section to operate at half of the input rate. In addition, the sharpened CIC filter is of length that is half of that of the original CIC filter. With the aid of the polyphase decomposition, the polyphase subfilters of the first section can also be operated at the half of the input rate.