In this article, they focus on the Automatic Modulation Classification (AMC). AMC is essential to carry out multiple CR techniques, such as dynamic spectrum access, link adaptation and interference detection, aimed at improving communications throughput and reliability and, in turn, spectral efficiency. In recent years, multiple Deep Learning (DL) techniques have been proposed to address the AMC problem. These DL techniques have demonstrated better generalization, scalability and robustness capabilities compared to previous solutions. However, most of these techniques require high processing and storage capabilities that limit their applicability to energy- and computation-constrained end-devices.
In this work, they propose a new gated recurrent unit neural network solution for AMC that has been specifically designed for resource-constrained IoT devices.
They trained and tested their solution with over-the-air measurements of real radio signals, which were acquired with the MIGOU platform.
Their results show that the proposed solution has a memory footprint of 73.5 kBytes, 51.74% less than the reference model, and achieves a classification accuracy of 92.4%.
Synopsis: Due to the spectrum scarcity problem, mostly in license-free ISM bands, and the forecasts regarding the increasing adoption of wireless communications, especially in scenarios like cities, it is essential to optimize the use of the spectrum to ensure the proper functioning of services and devices in the near future.
As the characteristics of the spectrum, by their own physical nature and its use, are very dynamic and vary constantly, devices must be able to intelligently adapt to these changes, as the Cognitive Radio paradigm proposes. Moreover, this adaptation should be done quickly in order to be effective and it should minimize the impact on the use of the spectrum.
Because of that, this work is going to be mainly focused on the development and evaluation of cognitive strategies with zero or minimum communication overhead. In other words, the aim of the research is to evaluate the degree of optimization of resources that can be achieved in a Cognitive Wireless Sensor Network (CWSN) by doing the cognitive cycle (spectrum sensing, learning and adaptation) mostly at node-level. To better exploit the cognitive radio capabilities of these networks, and thanks to the current development of wireless and processing technology, Software-Defined Radio (SDR) techniques are going to be used in sensor nodes for that purpose. This approach supposes a new paradigm in CWSNs which implies new challenges to be faced.
At this point, it appears to be necessary to evaluate some issues about the future of wireless communications. Will someday the need for cognition to use the spectrum outweigh the current energy constraints? In other words, will it be possible to achieve efficient and reliable wireless communication without cognitive capabilities in the near future? Answering this question will reveal whether it still make sense to compare the power consumption of SDR solutions with other platforms based on COTS radio transceivers or, conversely, the addition of cognitive capabilities will cease to pose a challenge to maximize systems’ efficiency and become a key point for their proper operation.
El objetivo de este Proyecto Fin de Carrera es el despliegue de un banco de pruebas para una red de sensores cognitiva (CWSN). Esta red contará con varios nodos cognitivos que permitirán la prueba de estrategias de optimización en este tipo de redes. Este banco de pruebas se realizará contando con una serie de nodos cognitivos previamente desarrollados en el laboratorio (cNGD) sobre el que se han hecho varios desarrollos software para adaptar tanto el protocolo de comunicación radio como la arquitectura cognitiva.
El despliegue del banco de pruebas cubrirá todas las salas permitidas del laboratorio B105 y el Departamento de Ingeniería Electrónica. Este proyecto abarca tanto la planificación del montaje físico de los nodos como el desarrollo de una interfaz para la gestión y recolección de información del banco de pruebas. Algunos parámetros a tener en cuenta serán el alcance de los nodos, su accesibilidad o la fuente de alimentación.
Title: Cognitive Wireless Sensor Network Platform for Cooperative Communications
Authors: Agustín Tena, Guillermo Jara, Juan Domingo, Elena Romero, Alvaro Araujo
Published in: International Journal of Distributed Sensor Networks
Date of Publication: January 2014
Digital Object Identifier : 10.1155/2014/473905
Nowadays, Wireless Ad-Hoc Sensor Networks (WAHSNs), specially limited in energy and resources, are subject to development constraints and difficulties such as the increasing Radio Frequency (RF) spectrum saturation at the unlicensed bands. Cognitive Wireless Sensor Networks (CWSNs), leaning on a cooperative communication model, develop new strategies to mitigate the inefficient use of the spectrum that WAHSNs face. However, few and poorly featured platforms allow their study due to their early research stage.
This paper presents a versatile platform that brings together cognitive properties into WAHSNs. It combines hardware and software modules as an entire instrument to investigate CWSNs. The hardware fits WAHSN requirements in terms of size, cost, features, and energy. It allows communication over three different RF bands, becoming the first cognitive platform for WAHSNs with this capability. In addition, its modular and scalable design is widely adaptable to almost any WAHSN application.
Significant features such as Radio Interface (RI) agility or energy consumption have been proved throughout different performance tests.
Title: PUE Attack Detection in CWSN Using Collaboration and Learning Behavior
Authors: Javier Blesa, Elena Romero, Alba Rozas, Alvaro Araujo and Octavio Nieto-Taladriz
Published in: International Journal of Distributed Sensor Networks
Date of Publication: June 2013
Digital Object Identifier : 10.1155/2013/815959
Cognitive Wireless Sensor Network (CWSN) is a new paradigm which integrates cognitive features in traditional Wireless Sensor Networks (WSNs) to mitigate important problems such as spectrum occupancy. Security in Cognitive Wireless Sensor Networks is an important problem because these kinds of networks manage critical applications and data. Moreover, the specific constraints of WSN make the problem even more critical. However, effective solutions have not been implemented yet. Among the specific attacks derived from new cognitive features, the one most studied is the Primary User Emulation (PUE) attack. This paper discusses a new approach, based on anomaly behavior detection and collaboration, to detect the PUE attack in CWSN scenarios. A nonparametric CUSUM algorithm, suitable for low resource networks like CWSN, has been used in this work. The algorithm has been tested using a cognitive simulator that brings important results in this area. For example, the result shows that the number of collaborative nodes is the most important parameter in order to improve the PUE attack detection rates. If the 20% of the nodes collaborates, the PUE detection reaches the 98% with less than 1% of false positives.