Elsevier

Computer Networks

Volume 51, Issue 12, 22 August 2007, Pages 3427-3447
Computer Networks

Adaptive energy-efficient registration and online scheduling for asymmetric wireless sensor networks

https://doi.org/10.1016/j.comnet.2007.02.007Get rights and content

Abstract

Increasing onboard processing capabilities of sensors enable self-organization in wireless sensor networks to dynamically adapt to ad hoc topologies and to react to task or network changes. Such self-organization, however. comes at a cost of additional energy consumption for the sensor nodes with already limited power resources. As energy limitations in unattended environments raise a major concern, such organizations need to trade-off between power consumption and topology maintenance. In this paper we present our adaptive energy-efficient registration and online scheduling (AEROS) protocol that exploits application based data flow characteristics to reduce power consumption during self-organization. Asymmetric data flow characteristics is used to govern route selection, and to formulate an organized transmission schedule with risk-free sleeping time. Our simulation results suggest that AEROS’s transmission schedule allows the minimum number of data message exchanges and guarantees a collision-free communication. AEROS provides significant energy savings in steady state using a low number of control messages.

Introduction

Recent advances in wireless technology and the development of small, low-cost, low-energy electronics has led to highly-distributed unattended sensor networks for various applications. These sensors result in highly dynamic topologies due to their ad hoc deployment, and also malfunctions, deactivations or displacements by natural incidents, etc. Self-organization, therefore, is a major requirement to organize the network after deployment and to maintain the functionality of the network. However, despite increased on-board processing capabilities, there are serious resource limitations that make self-organization a challenging task in sensor networks. At top of all, the lifetime of a sensor depends on a limited energy supply, and therefore, energy-efficiency is a major concern. Self-organization should, therefore, consume energy only parsimoniously.

In order to enable efficient self-organization in the network, it is important to observe the data flow characteristics in the network. For most sensor network applications, a single event, such as a change in the environment, triggers communications of a high number of sensors within the region. For instance, all neighbors that detect a local event attempt to communicate with the environment. In an uncoordinated environment, the signals of sensors that transmit simultaneously will collide, garbling the message, and requiring the sensors to retransmit. Such retransmissions result in additional energy consumption [4]. In addition, they necessitate additional mechanisms to enable reliable communications that is especially important at the time of peak activity.

Medium access control (MAC) aims at avoiding collisions. For instance, a sensor can sense the medium before transmitting using carrier sense medium access with collision avoidance (CSMA-CA) or apply contention-based approaches like IEEE 802.11. A major disadvantage with these types of MAC approaches is that all nodes need to constantly listen to the medium to make sure that they do not miss sent packets. Such idle listening can cause significant power consumption [27]. Various MAC protocols [11], [17], [31], [21] have been proposed to reduce the power consumption for sensor networks. These protocols, however, are designed for symmetric data flows and they do not exploit application-based flow patterns. In practice, the traffic flow in a sensor network is highly asymmetric, i.e., most of the data flow is from the sensor nodes towards the base station or the sink node (the node responsible for collecting the observations from the network). Symmetric allocations, therefore, result in unnecessary consumption of scarce network resources. Other solutions, e.g., [13], on the other hand, impose significant control message overhead to enable collision-free communication.

In this paper we propose a novel and adaptive sensor registration and scheduling protocol that exploits asymmetry of data flow characteristics to reduce power consumption for sensor networks. Our protocol accomplishes self-organization without any initial knowledge of the network topology. AEROS (adaptive energy-efficient registration and online scheduling) enables a publish/subscribe [8] based protocol that organizes the communication medium into independent control and data channels. Over the control channels, data channel reservations for non-conflicting schedules ensure a single transmission per observation during the operations of the network. In our study we exploit the fact that sensor network nodes, even though apply self-organization individually, have the opportunity to collaborate with each other to accommodate the overall data flow in the network. With its organized communication, AEROS allows collaboration between sensors such that observations can be verified or aggregated along the path to the sink node.

Our contributions can be summarized as follows:

  • 1.

    We consider the problem of MAC layer scheduling, routing and in-network data processing in direct relationship to each other. We enable adaptive routing paths towards the data collection points and we minimize the data transmissions such that each observation is transmitted only once per node using collision-free scheduling that is enabled even while the network is organizing.

  • 2.

    A major factor for the energy consumption is the amount of transmissions in the networks. Data aggregation is an approach that allows reductions in the amount of data that needs to be communicated. The chances of data aggregation during routing is maximized using application-aware publish/subscribe organization.

  • 3.

    We only depend on a reasonable resource usage during self-organization and application characteristics for asymmetric wireless sensor networks. Our experiment results show that AEROS significantly reduces the resources usage and the energy consumption.

In the rest of the paper Section 2 describes the AEROS protocol and its contributions. Section 3 presents a power modeling followed by simulation results from our performance analysis study in Section 4. Section 5 discusses previous works. We conclude with Section 6.

Section snippets

AEROS protocol

In this section we present the motivation behind our study and discuss the objectives of our protocol. In Section 2.2 we provide highlights from our protocol. We introduce the organization of the communication channel for control and data message exchanges in Section 2.3. We then summarize sensor registration for in-network communications and describe the scheduling of the data channel in Section 2.4.

Energy consumption modeling

In this section we provide a power model analysis for AEROS for a static network topology. In AEROS, each sensor will have a schedule as shown in Fig. 1. The allocations in the data channel describe when a sensor needs to be awake to receive or to transmit data. In all other times it can go to sleep to save power. Recall that sensors need keep polling the control channel to stay up-to-date with the changes in the network or application. After self-organization is completed, the power

Experiments

We study the performance AEROS in comparison to a baseline offline algorithm in addition to SMACS, TRAMA and B-MAC. We included the offline scheduling algorithm in our evaluation to examine the upper bounds of performance,. The offline algorithm is based on greedy centralized vertex coloring [19]. This algorithm assumes perfect knowledge of the complete topology and data flow pattern. Based on this assumed knowledge it organizes the communication channels for data transmissions. Therefore, it

Related work

Traditional objective of MAC protocols for wireless networks is to maximize bandwidth utilization, fair channel access for all nodes, and to reduce latency. In sensor networks, the major objective is to accommodate resource-constraints, i.e., to reduce power usage. In sensor network applications, the maximum bandwidth usage requirement is relaxed in comparison to traditional MAC protocols. Fairness is also not as important, since all nodes collaborate with each other for a certain task. In

Conclusions

Self-organization is a major requirement for ad hoc deployments of sensor networks. However, self-organization requires energy consumption due to control messages exchanged, and sensor networks have serious energy limitations. Therefore, self-organization should consume energy only parsimoniously. An opportunity that can be exploited for low-power optimizations arises from application characteristics of sensor networks. At top of all, data flow is highly asymmetric where the amount of

Acknowledgement

This work is supported in-part by a research grant from CENIC.

Saravanan Balasubramanian is a Software Engineer at Cisco Systems working on infrastructure and high availability. He has received his MS degree in Computer Science from the University of California, Davis. He has received his BS degree in Computer Science from California Polytechnic State University, San Luis Obispo.

References (34)

  • D. Aksoy, PLASMA: Planetary scale monitoring architecture, in: Proceedings of the ACM Multimedia, Singapore,...
  • D. Aksoy

    Information source selection for resource constrained environments

    ACM Sigmod Record

    (2005)
  • V. Annamalai, S.K.S. Gupta, L. Schwiebert, On tree-based convergecasting in wireless sensor networks, in: Proceedings...
  • S. Balasubramanian, D. Aksoy, Automatic registration for dynamic and reconfigurable sensor networks, in: Proceedings of...
  • S. Balasubramanian et al.

    Energy-efficient channel allocation in GeoSensor networks

  • S. Balasubramanian, D. Aksoy, Adaptive online scheduling for wireless sensor networks, in: Proceedings of the IEEE ISCN...
  • L. Bao, J. Garcia-Luna-Aceves, Hybrid channel access scheduling in ad hoc networks, in: Proceedings of the...
  • A. Carzaniga et al.

    Design and evaluation of a wide-area event notification service

    ACM Transactions on Computer Systems

    (2001)
  • A. Cerpa, D. Estrin, ASCENT: Adaptive self-configuring sensor networks topologies, in: Proceedings of the INFOCOM,...
  • U. Cetintemel, A. Flinders, Y. Sun, Power-aware data dissemination in wireless sensor networks, in: Proceedings of the...
  • T. van Dam, K. Langendeon, An adaptive energy-efficient MAC protocol for wireless sensor networks, in: Proceedings of...
  • B. Hohlt, L. Doherty, E. Brewer, Flexible power scheduling for sensor networks, in: Proceedings of the Information...
  • IEEE, Wirelesn LAN medium access control (MAC) and physical layer specifications, ANSI/IEEE Standard 802.11,...
  • L. Lui, E.L. Lloyd, A distributed protocol for adaptive link scheduling in ad hoc networks, in: Proceedings of the...
  • M. Miller, N. Vaidya. Minimizing energy consumption in sensor networks using a wakeup radio, in: Proceedings of the...
  • J. Mirkovic, G.P. Venkataramani, A self-organizing approach to data forwarding in large-scale sensor networks, in:...
  • J. Polastre, J. Hill, D. Culler. Versatile low power media access for wireless sensor networks, in: Proceedings of the...
  • Cited by (9)

    View all citing articles on Scopus

    Saravanan Balasubramanian is a Software Engineer at Cisco Systems working on infrastructure and high availability. He has received his MS degree in Computer Science from the University of California, Davis. He has received his BS degree in Computer Science from California Polytechnic State University, San Luis Obispo.

    Demet Aksoy has received her Ph.D. from the University of Maryland, College Park in 2000. She has been at the University of California, Davis as a tenure-track professor since then. She focuses on interdisciplinary problems of wireless networking and distributed information systems. She has authored more than 25 papers in networking and databases. She has been on the Technical Program Committees of a number of conferences, some of which are highly prestigious in databases.

    View full text