Elsevier

Ad Hoc Networks

Volume 8, Issue 4, June 2010, Pages 378-390
Ad Hoc Networks

Placement of multiple mobile data collectors in wireless sensor networks

https://doi.org/10.1016/j.adhoc.2009.08.007Get rights and content

Abstract

A major challenge affecting the lifetime of Wireless Sensor Networks (WSNs) comes from the unbalanced energy consumption over different parts of the network. This unbalanced energy consumption is a direct result of having a stationary sink: nodes near the sink are intensively used to relay data for other nodes to the sink. A natural solution to such a problem is to have multiple mobile sink nodes (which we call data collectors), and to change their locations periodically so that the load is distributed evenly among all sensor nodes. In this paper we propose a mobile data collector placement scheme for extending the lifetime of the network. In our scheme the lifetime of the network is divided into rounds and data collectors are moved to new locations at the beginning of each round. While previous work has focused on placing data collectors at predefined spots (e.g., the work in Gandham et al. (2003) [1]) or at the boundary of the network (e.g., the work in Azad and Chockalingam (2006) [2]), we define and solve two problems which are more general: the on-track placement where data collectors can be placed only along predefined tracks (roads) spanning the sensing field, and the general placement where data collectors may be placed at any point in the sensing field.

We formulate the problems as Mixed Integer Linear Programs (MILPs) and use a MILP solver (with a constant time limit) to find near-optimal placements of the data collectors and to find routing paths to deliver data to data collectors. Our experiments show that our schemes make significant extension to the lifetime of the network.

Introduction

The recently developed tiny sensor nodes have enabled a new generation of large-scale networks of untethered, unattended sensor devices suitable for a wide range of commercial, scientific, health, surveillance, and military applications. This rapidly evolving technology promises to revolutionize the way we interact with the physical environment and to facilitate collecting data which has never been available before [3]. However, as a result of the limited energy supply for sensor nodes, extending the lifetime of Wireless Sensor Networks (WSNs) has been a primary target for a significant amount of research during the last decade.

A sensor node has a wireless communication interface through which it can communicate with other devices in its vicinity. Due to the scarcity of the energy reservoir and due to the fact that communication is the dominant power consumer in a sensor node, the transmission range of these nodes is limited for energy-efficiency purposes. Sensor nodes which are spatially distant from the sink node use multi-hop relaying to deliver data to the sink. Multi-hop communication results in an unbalanced energy expenditure over different parts of the network; nodes around the sink deplete their energy reserve much faster than distant nodes [4], [1], [5]. Not only does this stop those nodes around the sink from functioning, but it also renders the sink unreachable by other nodes. While existing energy-aware protocols, at physical, Medium Access Control (MAC), and network layers, achieve significant energy savings for individual sensor nodes, they fail to solve this topology-related problem. One way to balance the load over the network is to deploy more nodes in areas closer to the sink. The authors in [6] proposed a nonuniform node distribution strategy that divides the sensing field into a number of coronas and gives the ratio in node densities between two consecutive coronas. One problem of this approach is the exponential growth of the total number of sensor nodes in the network. Moreover, since the area around the sink will have too many nodes, there will be a need for a complicated MAC protocol to control the access of these nodes to the wireless channel and/or to manage their duty cycles.

In this paper, we argue for using multiple mobile data collectors and propose a scheme for placing these data collectors in a way that balances the energy expenditure and increases the lifetime of the network. Our scheme divides the lifetime of the network into fixed length rounds (e.g., hours, days, or weeks) and moves the data collectors, which can be mounted on Autonomous Unmanned Vehicles (AUVs), to new locations at the beginning of each round. Some recently proposed schemes have addressed the issue of mobile sinks. However, they are either limited to a given set of predefined spots for sink locations and/or provide results that can be arbitrarily far from the optimal ones. To this end, the novel contribution of this paper is twofold:

  • 1.

    We define and solve two placement and routing problems. The first one assumes the existence of predefined tracks (e.g., a road network) spanning the sensing field, and data collectors can be moved over and placed at any point along these tracks. This would be practical in a situation where data collectors are carried on AUVs or robots that move along paved roads only. In the second one, a data collector can be placed anywhere in the sensing field. Some previous work assumes the existence of a set of predefined spots (i.e., points) where data collectors may be placed [1], and some is limited to placing data collectors at the boundaries of the field [2].

  • 2.

    We discretize the search space without affecting the quality of the derived solutions: we devise an algorithm that finds a finite set of relatively small number of points, and we prove the existence of an optimal placement in which each data collector is placed at a point in that set. Since the problem is modeled as a Mixed Integer Linear Program (MILP), making the cardinality of this set as small as possible would significantly improve the efficiency and the solution quality. Moreover, our approach involves solving one linear program per round; this makes our approach more efficient than earlier schemes, viz [7], which is based on solving a number of linear programming instances which is exponential in the number of data collectors.

By formulating the problems as MILPs, optimal solutions can be found. However, that may require exponential run-time in the worst case [8]. Therefore, we impose a time limit on a branch-and-bound solver in order to find near-optimal solutions in reasonable time.

The rest of this paper is organized as follows. Section 2 describes the model of the system and gives a formal problem definition. In Section 3, we present our placement schemes. Section 4 shows the experimental results. Finally, in Section 5, we conclude by summarizing the contributions and pointing out some related future research directions.

Section snippets

System model and problem definition

We consider a WSN consisting of N sensor nodes and R data collectors. Each sensor node collects data from the surrounding environment and sends the collected data to one of the data collectors. The transmission range, which is modeled as a disk, of all sensor nodes is fixed to r (m). The topology of the network is modeled as a graph G=(V,E), where V={n0,n1,,nN-1} is the set of N sensor nodes, and (i,j)E, if sensor nodes ni and nj are within the transmission range of each other. Each sensor

Data collector placement scheme

A major challenge in this placement problem is the infinite search space for data collector locations. The first step in our approach is to make the search space finite without affecting the quality of the solution. To explain our method of finding such a finite search space, we make the following definitions.

Definition 1

A finite set of points K is complete if and only if it satisfies the following property: there exists an optimal

Simulation results

We compare our proposed schemes with two other schemes: a static scheme where data collectors are stationary and a mobile scheme that ignores the residual energy of different sensor nodes. In the static scheme, data collectors are placed randomly in the sensing field, and we use a similar MILP to find near-optimal flow paths. While our schemes have the objective of Maximizing the minimum Residual energy (MR), the mobile scheme we compare with Minimizes the Maximum energy (MM) consumed by a

Conclusion

In this paper, we address the problem of unbalanced energy expenditure in WSNs resulting from using a stationary sink and multi-hop relaying. To alleviate the effect of this problem, we argue for using multiple, mobile data collectors, and propose a scheme for finding near-optimal placement of mobile data collectors together with the routing patterns to deliver data to data collectors. The novelty of our approach stems from:

  • 1.

    Solving a general problem in which a data collector can be placed

Acknowledgements

We thank Robert Benkoczi for his helpful discussions and insightful comments. We also thank the Natural Sciences and Engineering Research Council of Canada and King Saud University in Saudi Arabia for their support.

Waleed M. Alsalih received his Ph.D. degree from the School of Computing at Queen’s University, Ontario, Canada in 2009. He is currently an assistant professor in the Computer Science Department at King Saud University. Waleed’s research interests include wireless sensor networks, ad hoc networks, and energy-aware design.

References (12)

  • S. Gandham, M. Dawande, R. Prakash, S. Venkatesan, Energy efficient schemes for wireless sensor networks with multiple...
  • A. Azad, A. Chockalingam, Mobile base stations placement and energy aware routing in wireless sensor networks, in:...
  • I. Akyildiz et al.

    A survey on sensor networks

    IEEE Personal Communications Magazine

    (2002)
  • W. Heinzelman, A. Chandrakasan, H. Balakrishnan, Energy-efficient communication protocol for wireless microsensor...
  • S. Olariu, I. Stojmenovic, Design guidelines for maximizing lifetime and avoiding energy holes in sensor networks with...
  • X. Wu, G. Chen, S. Das, On the energy hole problem of nonuniform node distribution in wireless sensor networks, in:...
There are more references available in the full text version of this article.

Cited by (0)

Waleed M. Alsalih received his Ph.D. degree from the School of Computing at Queen’s University, Ontario, Canada in 2009. He is currently an assistant professor in the Computer Science Department at King Saud University. Waleed’s research interests include wireless sensor networks, ad hoc networks, and energy-aware design.

Hossam S. Hassanein is a leading researcher in the School of Computing at Queen’s University in the areas of broadband, wireless and variable topology networks architecture, protocols, control and performance evaluation. Before joining Queen’s University in 1999, he worked at the department of Mathematics and Computer Science at Kuwait University (1993–1999) and the department of Electrical and Computer Engineering at the University of Waterloo (1991–1993). Dr. Hassanein obtained his Ph.D. in Computing Science from the University of Alberta in 1990. He is the founder and director of the Telecommunication Research (TR) Lab http://www.cs.queensu.ca/~trl in the School of Computing at Queen’s. Dr. Hassanein has more than 300 publications in reputable journals, conferences and workshops in the areas of computer networks and performance evaluation. Dr. Hassanein has organized and served on the program committee of a number international conferences and workshops. He also serves on the editorial board of a number of International Journals. He is a senior member of the IEEE and serves as the Secretary of the IEEE Communication Society Technical Committee on Ad hoc and Sensor Networks (TC AHSN). Dr. Hassanein is the recipient of Communications and Information Technology Ontario (CITO) Champions of Innovation Research award in 2003. In March 2007, he received a best paper award at the IEEE Wireless Communications and Networks (a flagship IEEE communications society conference).

Selim G. Akl received his Ph.D. degree from McGill University in Montreal in 1978. He is currently a Professor of Computing at Queen’s University, Kingston, Ontario, Canada. His research interests are in parallel computation. He is author of Parallel Sorting Algorithms (Academic Press, 1985), The Design and Analysis of Parallel Algorithms (Prentice Hall, 1989), and Parallel Computation: Models and Methods (Prentice Hall, 1997), and a co-author of Parallel Computational Geometry (Prentice Hall, 1992). Dr. Akl is editor in chief of Parallel Processing Letters and presently serves on the editorial boards of Computational Geometry, the International Journal of Parallel, Emergent, and Distributed Systems, and the International Journal of High Performance Computing and Networking.

View full text