Copyright © 2008 Erchin Serpedin et al. This is an open access article distributed under the
Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Recent advances in micro-electromechanical systems (ME-MSs)
technology have enabled the design of low-power low-cost smart sensors equipped
with multiple onboard functions such as sensing, computing, and communications.
Such intelligent devices networked through wireless links have been referred to
as wireless sensor networks and recognized as one of the most important
technologies for the 21st century. Wireless sensor networks hold the promise to
revolutionize the sensing technology for a broad spectrum of applications,
including infrastructure monitoring and surveillance, disaster management,
monitoring the health status of humans, plants, animals, and industrial
machines, and so forth.
Wireless sensor networks can be viewed as a special case of wireless
ad hoc networks, and assume a multihop communication framework with no
centralized infrastructure and where the sensors cooperate spontaneously by
forwarding each other's packets for delivery from a source to a destination
node. The multihop nature of sensor networks is imposed by energy-consumption
reasons because of the superlinear power loss of wireless transmissions with
respect to the propagation distance.
In general, the design of wireless sensor networks is
subjected to a number of challenges: low energy consumption which is manifested
in minimal energy expenditure in each sensor node and efficient usage of
power-saving sleep/wake-up modes, scalability in the presence of a large number
of sensors, possibility of frequent node failures and network topology changes, collaborative
signal processing and data aggregation techniques to cope with the large number
of sensors which might congest the network with information, and efficient
communication protocols to deal with the
special broadcast communication paradigm and the increased possibility of
packet collisions and congestions for
nodes operating in closely spaced transmission ranges.
The goal of this special issue is to present the
state of the art and emerging distributed signal processing techniques that
deal with some of the above-mentioned design challenges. This special issue
consists of seven papers that treat important signal processing aspects such as
compression, quantization, estimation, detection, synchronization, and
localization in wireless sensor networks. A short description of the
contributions brought by these papers is next presented.
In the paper “Energy-constrained optimal
quantization for wireless sensor networks,” X. Luo and G. B. Giannakis deal
with the important problem of designing efficient quantizers that ensure optimal
reconstruction at the fusion center of the measurements yielded by a sensor as
well as the estimation of a deterministic parameter by exploiting the
measurements collected by a set of sensors. The design is carried out under
power constraints and information such as channel propagation effects,
modulation, and energy consumed by transceiver circuitry is considered into the
analysis. The effect of channel coding on the reconstruction performance is
also studied, and the optimum number of quantization bits and energy levels are
derived.
The problem of designing an optimal-level
distributed transform for wavelet-based spatiotemporal data compression in
wireless sensor networks is addressed by S. Zhou et al. in the paper “Ring-based optimal-level
distributed wavelet transform with arbitrary filter length for wireless sensor networks.”
This paper proposes a distributed optimal-level spatiotemporal
compression algorithm based on the ring model for general wavelets with
arbitrary supports. The proposed compression algorithm accommodates a broad
range of wavelet functions, effectively exploits the temporal and spatial
correlation of data measurements, and achieves significant reduction in energy
consumption and delay for data gathering in sensor clusters.
In “Distortion-rate bounds for distributed estimation using
wireless sensor networks,” D. Schizas et
al. address the problem of
centralized and distributed rate-constrained estimation of random signal
vectors by exploiting a network of wireless sensors (encoders) that communicate
with a fusion center (decoder). Within the proposed framework, the
authors determine lower and upper bounds on the corresponding distortion-rate
function using both centralized as well as distributed estimation techniques.
The paper “Distributed event region detection in wireless sensor networks,”
coauthored by J. Fang and H. Li, proposes a graph-based method for distributed
event-region detection in wireless sensor networks. The proposed detection
scheme exploits a graphical model to take into account the fact that events
occurring in geographically neighboring sensors present a statistical
dependency. This scheme also admits energy and bandwidth efficient distributed
implementations.
Q. Chaudhari and E. Serpedin, in the paper “Clock estimation
for long-term synchronization in wireless sensor networks with exponential delays,” deal with the maximum
likelihood estimation of the clock parameters (phase, skew, and drift) in
two-way timing exchange mechanisms and in networks with exponentially
distributed delays. The paper entitled “Extension of pairwise broadcast clock
synchronization for multicluster sensor networks,” coauthored by K. L. Noh et al., proposes a novel clock
synchronization protocol to minimize the overall energy consumption in wireless
sensor networks that assume general multicluster topologies. The proposed
synchronization approach relies on a receiver-only synchronization approach and
it can be viewed as a generalization of the pairwise broadcast synchronization (PBS)
protocol. Like PBS, the proposed synchronization approach exhibits the distinct
advantage that the number of sensor nodes can be synchronized by only over-hearing
time message exchanges between pairs of nodes, and therefore it reduces significantly
the overall network-wide energy consumption by decreasing the number of
required timing messages for synchronization.
Finally, in “Optimization of sensor locations and sensitivity
analysis for engine health monitoring using minimum interference algorithms,”
P. Cotae et al. address the problem of optimal placement of sensors in the presence of additive
white Gaussian noise (AWGN) by
considering the sensors as systems that present full communications
capabilities and by minimizing the RF-interference induced by the wireless
communication channels among the sensor nodes. Numerical simulations and a
sensitivity analysis study are presented to illustrate the robustness of the
proposed algorithm.
Acknowledgment
The editors of this special issue would like to express their heartfelt “Thank You!”
to all the people (editors, authors, and reviewers) who supported the
publication of this special issue.
Erchin Serpedin
Hongbin Li
Aleksandar Dogandžić
Huaiyu Dai
Paul Cotae