Cognitive Wireless Sensor Networks Framework for Green Communications Design Alvaro Araujo, Elena Romero, Javier Blesa, Octavio Nieto-Taladriz E.T.S.I. Telecomunicacion Electronic Engineering Department Universidad Politécnica de Madrid Madrid, Spain {araujo,elena,jblesa,nieto}@die.upm.es Abstract— Cognitive Wireless Sensor Networks is an emerging technology with great potential to avoid traditional wireless problems such as reliability, interferences and spectrum scarcity. Because of the Wireless Sensor Network fast growth and the use of batteries, current rate of power consumption per unit of data cannot be sustained. Therefore, one of the major challenges face today is low power consumption in Wireless Sensor Networks.Cognitive Wireless Sensor Networks framework is a key issue in green communications because of many protocols, strategies and optimization algorithms could be tested. In this paper a framework composed of a network simulator with cognitive capabilities and low power Cognitive Wireless Sensor Networks real devices with a feedback relation is presented. The benefits of the proposed framework are demonstrated with three different scenarios and simple cognitive green communications strategies. Results show how new concepts have been integrated in the framework with good results and as simple cognitive radio strategies can reduce large amount of power. Keywords-cognitive; framework; low power design; wireless sensor networks I. INTRODUCTION Wireless network power consumption has not been an important research issue because it has been insignificant in comparison with wired network consumption. Over the recent years, wireless and mobile communications have increasingly become popular with consumer. According to [1], global mobile data traffic will increase 26-fold between 2010 and 2015 (in 2010 global mobile data traffic grew 2.6- fold). Mobile data traffic will grow at a compound annual growth rate of 92 percent from 2010 to 2015, reaching 6.3 exabytes per month by 2015. Taking into account this prediction, the current rate of power consumption per unit of data cannot be sustained. In regards to wireless networks, one of the fastest growing sectors in recent years was undoubtedly Wireless Sensor Networks (WSN). According to the report [2], WSN market will grow rapidly from $0.45 billion in 2011 to $2 billion in 2021. WSN are increasingly introduced into our daily lives. Potential fields of applications are from home control to military scenarios or critical information infrastructure protection. In this kind of scenarios, lifetime of the nodes typically ranges from 2 to 5 years, making power consumption a dramatic requirement to establish. Thus, reducing energy consumption is one of the most important challenges to face when designing WSN. Recently, to increase lifetime (as well as other very important problems like spectrum scarcity, interferences or reliable connections), most WSN rely on the new cognitive paradigm. Cognitive Network is an intelligent wireless communication system that is aware of its surrounding environment, and with the possibility to adapt its internal parameters to achieve reliable and efficient communications (in terms of power consumption too) [3]. This solution benefits from “free” environmental energy according to the “green” philosophy, which is to reduce the carbon footprint and to improve reliability of power supply automations. In order to enable design and development of new green protocols and power reduction techniques for Cognitive Wireless Sensor Networks (CWSN) and evaluate their performance, simulation and emulation environments are necessary. The challenge in simulators is to determine if these simulations provide us a good enough correspondence with real deployments. In this paper, a complete simulation and emulation framework for CWSN using regular standards is presented. The simulator is based on the Castalia simulator including all the cognitive modules. The simulations are fed with real CWSN devices to provide a more realistic approach. The organization of this paper is as follows. In Section 2, works in CWSN simulator and emulator frameworks are reviewed. In Section 3, new CWSN framework is described. In Section 4, a proof of concept is shown. Finally, the conclusions are drawn in Section 5. II. RELATED WORK Because of the novel research field, there are not many specific frameworks for green communications design over CWSN. It is natural that most of works are based on WSN simulators. There are several WSN simulators used by researchers to develop their works. For example, NS-2 [4] is one of the most well-known simulators. Most of the WSN research society uses this simulator, although the latest release was in 2008. NS-3 will be its substitute, but it is still in the early stages. OMNET++ [5] is another framework very well- known among researchers. It proposes a modular library which could be used to develop network simulators. Only by composing different modules, the developer can create its own simulator or scenario. 34 Copyright (c) IARIA, 2012. ISBN: 978-1-61208-197-7 COCORA 2012 : The Second International Conference on Advances in Cognitive Radio