I- INTRODUCTION

II- BIG DATA CONCEPT, DIMENSIONS AND ANALYTIC TOOLS IN WIRELESS SENSOR NETWORKS

III- BIG DATA CHALLENGES IN WIRELESS SENSOR NETWORKS

IV- BIG DATA AGGREGATION PROPOSED STRATEGIES IN WIRELESS SENSOR NETWORKS

V- CONCLUSION

REFERENCES

ABSTRACT

The emergence of new data handling technologies and analytics enabled the organization of big data in processes as an innovative aspect in wireless sensor networks (WSNs). Big data paradigm, combined with WSN technology, involves new challenges that are necessary to resolve in parallel. Data aggregation is a rapidly emerging research area. It represents one of the processing challenges of big sensor networks. This paper introduces the big data paradigm, its main dimensions that represent one of the most challenging concepts, and its principle analytic tools which are more and more introduced in the WSNs technology. The paper also presents the big data challenges that must be overcome to efficiently manipulate the voluminous data, and proposes a new classification of these challenges based on the necessities and the challenges of WSNs. As the big data aggregation challenge represents the center of our interest, this paper surveys its proposed strategies in WSNs.

INTRODUCTION

The technological advancement in several research areas, including wireless communications, led the researchers to focus on the wireless sensor networks field [1], [2] which represents an innovating technology occupying a crucial place in the data processing, combining wireless communication, detection functions and embedded technology. This emerging technology is gaining ground and is becoming increasingly ubiquitous in all the aspects of the environmental monitoring and processing. The main feature of these systems is the possibility of their deployment in remote and hostile locations, providing users with flexible organization options and facilitating the access to data. A wireless sensor network (WSN) is formed by collections of sensor nodes widely deployed in generally inaccessible areas and forming data propagation networks. Their main role is to observe a process, collect data and transmit them to a base station for handling. The sensors have introduced the idea of sensor networks based on the collaborative effort between large sets of sensors. Sensor networks have developed rapidly in recent years and their deployment represents an advantage for new applications. The large utilization of WSN applications and the diversity of the involved fields contributed to increase the volume of data collected and processed. Indeed, when the WSN networks grow and gain in volume and the deployment space, the data collected and processed grow exponentially requiring thus efficient processing, and making consequently traditional data processing methods difficult to use. Big data technology [3], [4] can represent an effective solution for collecting, analyzing, storing and transmitting data in voluminous wireless sensor networks. Indeed, since the applications of WSN are increasing massively, the sensors deployed are responsible of producing the data in large volume making wireless sensor networks key contributors of big data. The big data paradigm in wireless sensor networks [5] is young and emerging. It was initially adapted to wired networks, but is gaining momentum in wireless sensor networks, increasing thus the need for new technologies and architectures to handle data. The term big data is used to characterize large data sets that can be complex and therefore difficult to manage by conventional data processing methods. In a wireless sensor network, these huge masses of data are generated every minute and must be collected by the sensor nodes before being transmitted to the base station.