Enrique Stevens-Navarro

This study presents a deep learning-based framework for beam pattern synthesis in optimized uniform linear antenna arrays, combining Differential Evolution–based pre-optimization with recurrent neural network (RNN) modeling. Radiation patterns are first generated to satisfy sidelobe suppression and directivity constraints and are then used to train recurrent models that learn the mapping between radiation patterns and complex excitation parameters. A formal mathematical formulation of the Simple RNN, Gated Recurrent Unit (GRU), and Long Short-Term Memory (LSTM) architectures is provided, together with a per–time-step computational cost analysis based on dominant matrix–vector multiplications. A comparative evaluation under identical training conditions shows that gated architectures significantly outperform the standard RNN. Although the LSTM achieves the lowest prediction errors, the GRU attains comparable performance with reduced structural complexity. Beam pattern synthesis experiments for unseen steering directions demonstrate accurate reconstruction of main lobe alignment, sidelobe levels (approximately −12 to −13 dB), and directivity values close to 8 dB. The floating-point operations (FLOPs) analysis indicates that the GRU requires fewer dominant operations per time step than the LSTM, potentially reducing computational cost and energy consumption in resource-constrained beamforming applications.

This article describes the design, implementation, and evaluation of an indoor localization system based on Received Signal Strength Indicator (RSSI) measurements in wireless sensor networks. While the majority of the literature uses the IEEE 802.15 standard for this type of system, all of the measurements in this study were performed using a test bench operating under the IEEE 802.11ax standard in the 2.4 GHz band. RSSI is widely used due to its simplicity and availability; however, its accuracy is limited by signal attenuation, electromagnetic interference, and environmental variability. To mitigate these limitations, the present work proposes the implementation of advanced techniques, including weighted averages and positioning algorithms such as Min–Max, Maximum Likelihood, and trilateration, aiming to achieve an accuracy of 2 m in controlled conditions. The design also included a specialized test bench to calculate the coordinates and estimate the location of unknown nodes using anchor node positioning. This approach combines the simplicity of RSSI with optimized algorithms, providing a robust and practical solution for indoor localization. The results validate the system’s effectiveness and highlight its potential for future applications in real-world environments, opening new possibilities for optimizing wireless sensor networks and addressing the current challenges in localization systems.

This work proposes and describes a deep learning-based approach utilizing recurrent neural networks (RNNs) for beam pattern synthesis considering uniform linear arrays. In this particular case, the deep neural network (DNN) learns from previously optimized radiation patterns as inputs and generates complex excitations as output. Beam patterns are optimized using a genetic algorithm during the training phase in order to reduce sidelobes and achieve high directivity. Idealized and test beam patterns are employed as inputs for the DNN, demonstrating their effectiveness in scenarios with high prediction complexity and closely spaced elements. Additionally, a comparative analysis is conducted among the three DNN architectures. Numerical experiments reveal improvements in performance when using the long short-term memory network (LSTM) compared to fully connected and convolutional neural networks.

A dual-beam coherent feeding system design approach with a non-uniform layout on a concentric ring array is described and synthesized. In this case, the feeding system is based on a reconfigurable topology composed of a set of alternated power dividers and combiners, providing coherent in-phase outputs. Thus, in this paper, a two-beam architecture based on a coherent feeding system formed by a set of intercalated input signals feeding each circular ring in a non-uniform antenna array with multi-beam shaping and steering features is analyzed. The task of optimizing the aperiodic layout on the shared aperture based on the radii of the circular rings is realized by the differential evolution method. Numerical experiments grounded in antenna synthesis validate the capabilities and improved performance of the proposed dual-beam configuration with a non-uniform layout in contrast with its uniform counterpart, with enhanced performance on average by up to −6.1 dB for sidelobe level and 3.5 dB for directivity. Additionally, the results show a significantly less complex two-beam feeding network in contrast with the case of a typical electronically scanned array—in this proposal, each direction of maximum radiation is conformed and scanned with approximately half of the control inputs.

The efficient utilization of the radio-electric spectrum (or simply spectrum) is essential to satisfy the ever- increasing amount of bandwidth required by future wireless communications networks. Cognitive radio (CR) networks aim to improve this efficiency by dynamically exploiting the underutilized spectrum (also called spectrum opportunities). To identify these transmission opportunities, cognitive users might draw on spectrum sensing, although this task increases the energy consumption. For battery-powered terminals, this increment might represent a challenge, also considering that spectrum sensing must be recurrently performed. For a scenario in which the CR user first senses the spectrum and then, if allowed, transmit data, the average energy consumption depends on the time used for spectrum sensing and for data transmission, which also impacts the spectrum-efficiency. Thus, improving the energy-efficiency might implicate a reduction on the spectrum-efficiency. This paper analyses the energy-efficiency in the context of spectrum sensing of superimposed training-based transmissions, showing the advantages of using an enhanced spectrum sensing method in terms of the relationship between the spectrum and energy- efficiency.

En este artículo se presenta un protocolo de difusión de mensajes de seguridad para emergencia que utiliza una adaptación dinámica del tiempo de retraso en el entorno de una red vehicular cognitiva. El escenario que se estudia considera la compartición del espectro entre la banda para comunicaciones vehiculares y la banda de la televisión digital con el objetivo de mitigar los efectos de la tormenta de difusión en una red vehicular. Dos de las técnicas de mitigación de la tormenta de difusión más utilizadas en la literatura son comparadas contra el protocolo propuesto en términos de las probabilidades de accesibilidad de los mensajes de emergencia y de la ocupación del canal. Los resultados numéricos muestran que el protocolo con la adaptación dinámica del tiempo presenta un mejor desempeño en condiciones de alta densidad de vehículos, lo cual es característico de un escenario de tormenta de difusión.

This paper presents an approach to exploit the superimposed training (ST)-based primary users’ (PUs) transmissions in the context of spectrum sensing for cognitive radio. In the low signal-to-noise ratio (SNR), the proposed scheme splits the spectrum sensing phase into two sample processing periods, allowing a secondary user (SU) to carry out a training sequence synchronization (with a small probability of error) before the implementation of a robust spectrum sensing algorithm that enhances the detection, based on the deterministic signal components embedded in the ST PU’s signals along with the unknown data signal. The overall sensing performance is improved using a reasonable number of samples to achieve a high probability of detection, resulting in a reduced spectrum sensing duration. Furthermore, a low computational complexity version of the proposed ST combined approach for a reduced phase (SCAR-Phase) of spectrum sensing is presented, which attains the same detection performance with a smaller number of real operations in the low SNR. In the practical consideration of imperfect training sequence synchronizations, the results show the advantages of exploiting the ST sequence to perform spectrum sensing, thus quantifying the significant improvement in detection performance and the maximum SU’s achievable throughput.

In this work, a flexible and reconfigurable feeding network design for a nonuniform aperture on circular concentric ring arrays is proposed and analyzed. The network subsystem delivers coherent in-phase outputs with a Gaussian-like amplitude distribution, in a modular and basic topology based on sets of alternated power combiners and dividers. A complete antenna system in a monobeam configuration with a coherent network based on grouped inputs (blocks) per ring for an aperiodic concentric ring array with beam scanning and beam shaping properties is synthesized and analyzed. Additionally, a comparative analysis based on nonuniform and uniform concentric ring arrays fed by the proposed coherent network configuration is conducted and assessed. The optimization of the aperiodic layout on the antenna aperture (radii and interelement antenna spacings) is done by the differential evolution algorithm. Numeric experimentation demonstrates the performance advantages and capabilities of the proposed coherent network configuration with a nonuniform aperture over its uniform counterpart, with an improvement in average equal to −8.7 dB of side lobe level and 3.9 dB of directivity. Furthermore, the numeric examples show a complexity reduction on the coherent feeding network configuration based on the number of control signal inputs compared with a conventional phased antenna array; in the proposed configuration, the main beam is steered and shaped with N-1 control feeding ports per ring in this antenna system.

Multiple-input Multiple-output (MIMO) systems are an attractive choice to enhance the communication performance by increasing the number of antennas at transmitter and receiver. However, the physical implementation of MIMO systems has several issues such as high cost and high computational complexity for large antenna arrays. Additionally, due to space limitations, huge MIMO systems cannot be implemented in small devices. Hence, antenna selection techniques emerge as a solution to maintain the benefits of MIMO with an affordable trade-off between complexity and implementation cost. Besides, cooperative wireless sensor networks (WSN) may consider virtual MIMO systems. Thus, in this paper, we propose an antenna selection algorithm called Antenna Adaptive Selection of Information (ASSI) for being implemented on WSN. AASI adapts the transmission information according to channel characteristics. Our algorithm achieves affordable results in performance and capacity close to the optimal antenna selection for a specific signal-to-noise ratio with a reduced computational complexity.