Decentralized Average Consensus in Wireless Sensor Networks with Unreliable Communication Channels

Decentralized Average Consensus in Wireless Sensor Networks with Unreliable Communication Channels

Steve Saed (Indiana University-Purdue University Indianapolis, USA), Lingxi Li (Indiana University-Purdue University Indianapolis, USA) and Dongsoo S. Kim (Indiana University-Purdue University Indianapolis, USA)
Copyright: © 2012 |Pages: 18
DOI: 10.4018/jhcr.2012070103
OnDemand PDF Download:
No Current Special Offers


This study proposes and evaluates an average consensus scheme for wireless sensor networks. For this, two communication error models, the fading signal error model and approximated fading signal error model, are introduced and incorporated into the proposed decentralized average consensus scheme, especially adapted to the constraints of wireless sensor networks. A mathematical analysis is introduced to derive the approximated fading signal model from the fading signal model and different simulation scenarios are introduced and their results analyzed to evaluate the performance of the proposed scheme and its effectiveness in meeting the needs of wireless sensor networks.
Article Preview


In recent years, major advances in wireless communication and digital electronics led to the development of tiny sensors that can broadcast sensed data over short communication ranges. A wireless sensor network (WSN) is composed of a number of sensor nodes that have sensing, processing and communication components that enable all the network nodes to collaborate in order to achieve a particular task (Akyildiz, Su, Sankarasubramaniam, & Cayirci, 2002). Each node in a WSN is designed to have limited computational and power capacities and use basic broadcast communication protocols in order to exchange information with its neighbors (Perkins, 2000). Moreover, sensor nodes are designed to be densely and randomly deployed very close to the phenomenon that the WSN is expected to monitor. This allows WSNs to be used in a wide range of applications where the exact position of each sensor cannot be pre-determined and the nature of the topology makes more sensors prone to failure such as the ones used in military applications (Intanagonwiwat, Govindan, & Estrin, 2000).

Wireless sensor networks gained the attention of many research communities due to the wide range of applications where they can be employed. Military applications of WSNs include battlefield surveillance where WSNs deployed in enemy territory can track and report the movement of enemy troops. Another important military application is detecting chemical and biological attacks by monitoring the atmosphere of the battlefield (Akyildiz et al., 2002). Furthermore, WSNs can be used in many different environmental applications that intend to track the movement of certain species in remote ecosystems, detect forest fire, monitor the pollution at the bottom of a certain river (Jaikaeo, Srisathapornphat, & Shen, 2001). Furthermore, WSNs can be used in health-care, home-applications (Rabaey, Ammer, da Silva, Patel, & Roundy, 2000) (Warneke, Last, Liebowitz, & Pister, 2001), etc.

Several applications of WSNs can be incorporated in transportation systems. First, WSNs can be integrated into vehicular ad hoc networks (VANETs), which are a type of ad hoc wireless mobile networks used to exchange information among vehicles on the road (Xi & Li, 2008). Sensors in each vehicle are used to collect information about road congestion, speed of the different vehicles on the road, etc, which is then transmitted from one vehicle to another. This information can be used to alert the drivers to potential hazards which would help improve road safety (Batool & Khan, 2005). Also, such information can be used to reduce traffic congestions by giving drivers information about congestions ahead of them so that they find a less congested route to their destination (Biswas, Tatchikou, & Dion, 2006). Second, WSNs can be used in intelligent transportation systems (ITS), which are composed of a bundle of sensing, communication and decision-making components added to vehicles that are designed to help the driver avoid any potential hazards. For instance, an ITS mounted in a certain vehicle can monitor the distance from other vehicles on the road and automatically control vehicle-to-vehicle distance when necessary (Somda, Cormerais, & Buisson, 2009).

Many problems often arise in WSNs due to their network architecture, environments where they are deployed or the computational and power limitations of their nodes. The network nodes are often exposed to harsh conditions, like the ones experienced in a battlefield, or may run out of power after a short period of time, which means that the network is going to be losing nodes over time. Also, the modest computational power of the nodes will impose certain constraints on the algorithms that are going to be implemented in the network in terms of computational requirements, memory storage capabilities and computational efficiency. Consequently, the algorithms implemented in the network need to work in a decentralized manner in order to minimize the effect of losing nodes on the network functionality. Other problems experienced in WSNs are similar to those found in all ad hoc mobile wireless networks like communication errors, hidden terminal problem, exposed terminal problem, etc.

Complete Article List

Search this Journal:
Open Access Articles: Forthcoming
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing