Using ANNs to predict the evolution of spectrum occupancy in cognitive-radio systems P.I. Enwere, E. Cervantes-Requena, L.A. Camuñas-Mesa and J.M. de la Rosa Journal Paper · Integration, vol. 93, 2023 abstractdoi
This paper analyzes the use of Artificial Neural Networks (ANNs) to identify and predict the evolution of vacant portions or frequency holes of the radio spectrum in Cognitive Radio (CR) systems. The operating frequency of CR transceivers can be modified over the air according to the information provided by the ANN in order to establish the communication in the least occupied band. To this end, ANNs are trained with time-series datasets sensed from the electromagnetic environment. Several network architectures are considered in the study, including Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks and hybrid combinations of them. These ANNs are modeled and compared in terms of their complexity, speed and accuracy of the prediction. Both simulations and experimental results are shown to validate the approach presented in this work.
Reducing the Nonlinearity and Harmonic Distortion in FD-SOI CMOS Current-Starved Inverters and VCROs P.I. Okorie, J. Ahmadi-Farsani and J.M.de la Rosa Journal Paper · AEU - International Journal of Electronics and Communications, vol. 142, article 153992, 2021 abstractdoipdf
This paper demonstrates experimentally how to reduce the nonlinearity of some analog and mixed-signal circuits by using the enhanced body effect provided by Fully-Depleted Silicon on Insulator (FD-SOI) CMOS technology. A current-starved CMOS inverter and a Voltage-Controlled Ring Oscillator (VCRO) are considered as case studies. The inverter is configured as a simple amplifier stage in which the harmonic distortion can be reduced and even removed by the combined action of the control voltages applied at the gate and bulk terminals of the current-source transistors. This current-starved inverter is used as the basic building block of a VCRO, where a more linear voltage-to-frequency characteristic can be achieved if the bulk terminal is used as the control voltage of the oscillator. The circuits under study have been designed and fabricated in a 28-nm FD-SOI technology and experimental results are shown to validate the presented approach.
Using ANNs to Predict Frequency Spectrum Occupancy in Cognitive-Radio Receivers P.I. Okorie, L.A. Camuñas-Mesa and J.M. de la Rosa Conference · Conference on Design of Circuits and Integrated Systems DCIS 2022 abstract
This paper analyses the use of Artificial Neural Networks (ANNs) to identify vacant portions of the electromagnetic spectrum or frequency holes in Cognitive Radio (CR) systems. Several ANN topologies are considered, including Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks or hybrid combinations of them. These ANNs are modeled and compared in terms of their complexity, speed and accuracy of the prediction. As an application, a CR-based receiver is simulated, where Radio-Frequency (RF) signals are digitized by a Band-Pass Sigma-Delta Modulator (BP-ΣΔM) with a tunable notch frequency, which is modified according to the less occupied band predicted by the ANNs.