- Overview
- Research Agenda
- Participants
- Major Publications
- Funding Source
- Related Links
Overview
Recent technological advances have led to the emergence of small,
low-power devices that integrate sensors and actuators with limited
on-board processing and wireless communication capabilities. Pervasive
networks of such sensors and actuators open new vistas for
constructing complex control systems. Unlike traditional wired or
wireless networks, sensor networks possess certain characteristics
which warrant their treatment as a special class of ad hoc networks:
(1) Data-centric: Sensor networks are largely data-centric,
with the objective of delivering data collected, in a timely fashion,
to the required destination rather than performing intensive
computation on the data.
(2) Application-oriented: While traditional wired and
wireless networks are expected to cater to a variety of user
applications, a sensor network is usually deployed to perform specific
tasks. This property makes it possible to enable nodes to respond in
an application-aware fashion. Data can be collected, appropriately
aggregated with consideration of the requirement of the applications,
and then forwarded to a controller node (rather than simple end-to-end
data transfer).
(3) Collaborative: Because of the application-oriented
nature of sensor networks, how nodes collaborates with each other to
realize the global system objective outweigh the objective of
achieving
fairness of individual connections. This is in sharp contrast to
conventional wired and wireless networks in which provisioning of
fairness to users is an important design criterion.
(4) Energy-constrained: As most of the low-power devices in
sensor networks have limited battery life and replacing batteries on
tens of thousands of these devices is infeasible, any
protocol/algorithm that will be eventually deployed in sensor networks
has to be energy aware.
A major requirement of this type of sensor networks is for the
sensors to reliably disseminate information within a time frame that
allows the controller to take necessary action, even in the case of
poor spatial distribution of sensor devices, wireless/acoustic
interference, and malicious destruction. Out-of-date information is of
no use, as the object that was tracked may no longer be in the
vicinity when the information is received. This presents a key
technical challenge in cooperative engagement --- how to effectively
coordinate and control sensors over an unreliable wireless ad hoc
network. In particular, due to the unique characteristics of
data-centric sensor networks, many new design issues arise and
protocols originally designed for wireline and/or generic ad hoc
networks have to be adapted or entirely re-designed. For example, when
sensors are placed in open fields for environmental applications, they
may not be evenly distributed over a region. One may either use (and
have to coordinate the movement of) mobile ``router'' sensors to fill
the ``holes'' and maintain network connectivity, or exercise topology
and power control in a hierarchical, clustering fashion. The latter,
in turn, implies that conventional, flat ad hoc routing protocols may
not render the best performance.
Research Agenda
(click here for details)
To address the issues of effectively coordinating and controlling
sensor devices for information dissemination, we have carried out
tasks along the following research avenues:
Participants
- Wei-Peng Chen (Ph.D. 2004)
- Chunyu Hu (Ph.D. 2006)
- Rong Zheng (Postdoc 2004-2005;
Ph.D. 2004)
- Guanghui He (Postdoc 2004-2005)
- Hyuk Lim (Postdoc 2003-2006)
Major Publication
- Rong Zheng, Ye Ge, Jennifer C. Hou, Sandy R. Thuel, A case
for mobility support with temporary home agents, ACM Mobile
Computing and Communications Review, Vol. 6, No. 1, pp. 32--47,
January 2002.
- Wei-Peng Chen, Yung-Ching Hsiao, Jennifer C. Hou, Ye Ge, and
Michael Fitz, Syndrome: A light-weight approach to improve
the TCP performance in mobile wireless networks, Journal of
Wireless Communications and Mobile Computing, Special Issue on
Reliable Transport Protocols for Mobile Computing, Vol. 2, pp.
37--57, February 2002.
- Wei-Peng Chen and Jennifer C. Hou, Dynamic, ad-hoc source
routing with connection-aware link-state exchange and
differentiation, Proc. IEEE Globecom 2002, November 2002.
- Rong Zheng and Robin Kravets, On-demand power management for ad
hoc networks, Proc. IEEE INFOCOM 2003, April 2003
(acceptance ratio=20%).
- Chunyu Hu, Yefei Hong, and Jennifer C. Hou, On mitigating the
broadcast storm problem in MANET with directional antennas, Proc.
IEEE Int'l Conf. on Communications, May 2003.
- Rong Zheng, Jennifer C. Hou, and Lui Sha, Asynchronous wakeup
for ad hoc networks, Proc. of 4th ACM International Symposium on Mobile Ad
Hoc Networking and Computing (MobiHoc'03), June 2003 (acceptance
ratio = ~15%).
- Wei-Peng Chen, Jennifer C. Hou, and Lui Sha, Dynamic clustering
for acoustic target tracking in wireless sensor networks, Proc.
IEEE 11th Int'l Conference on Network Protocols, November 2003
(acceptance ratio = ~15%).
- Rong Zheng, Jennifer C. Hou, and Lui Sha, Performance analysis
of the IEEE 802.11 power saving mode, Proc. SCS Western
Multiconference on Computer Simulation -- communication networks
and distributed systems modeling and simulation conference,
January 2004.
- Chunyu Hu and Jennifer C. Hou, LISP: an link-indexed statistical
traffic prediction approach to improving IEEE 802.11 PSM, Proc. of
IEEE Int'l Conf. on Distributed Computing Systems (ICDCS'04),
March 2004 (acceptance ratio = %18).
- Wei-Peng Chen and Lui Sha, An energy-aware data-centric generic
utility based approach in wireless sensor networks Proc. of
3rd ACM/IEEE Symposium on Information Processing in Sensor
Networks, April 2004.
-
Ning Li and Jennifer C. Hou, ``BLMST: A Decentralized,
Power-Efficient Broadcast Algorithm for Wireless Sensor Networks,''
Proc. IEEE 1st Int'l Conf. on Quality of Service in Heterogeneous
Wired/Wireless Networks (QShine'04), October 2004.
-
Srikanth Kandula, Jennifer C. Hou, and Lui Sha, ``The case for
resource heterogeneity in large sensor networks,'' Proc. IEEE Milcom
2004, November 2004 (invited paper).
-
Rong Zheng, Jennifer C. Hou, and Lui Sha, "On time-out driven
power management policies in wireless networks," Proc. IEEE Globecom
2004, December 2004 (acceptance ratio = 30%).
- Paul Havinga, Jennifer C. Hou, Mani Srivastava, and Feng Zhao,
"Wireless Sensor Networks," editorial for the special issue of wireless
sensor networks, IEEE Wireless Communications Magazine,
Vol. 11, No. 6, pp. 4--5, December 2004.
-
Hyuk Lim and Jennifer C. Hou, ``Localization for anisotropic
sensor networks,'' Proc. IEEE INFOCOM, Miami, Florida, March 2005
(acceptance ratio = 17%).
- Chunyu Hu and Jennifer C. Hou, "A link-indexed statistical traffic
prediction approach to improving IEEE 802.11 PSM," Elsevier
Ad Hoc Networks Journal, vol. 3 No. 5, special Issue on
"Data communication and topology control in ad hoc networks,"
pp. 529--545, September 2005 (invited paper).
-
Wei-Peng Chen, Jennifer C. Hou, Lui Sha, and Marco Caccamo, ``A
distributed, energy-aware, utility-based approach for data transport
in wireless sensor networks,'' Proc. of IEEE Milcom, October 2005
(invited paper).
-
Guanghui He and Jennifer C. Hou, ``Tracking targets with quality
in wireless sensor networks,'' IEEE Int'l Conf. on Network Protocols
(ICNP'05), November 2005 (acceptance ratio = 15%).
- Wei-Peng Chen and Jennifer C. Hou, ``Data gathering and fusion in sensor
networks,'' Handbook of Sensor Networks: Algorithms and Architectures,
pp. 493--526, Ivan Stojmenovic (Ed), John Wiley and Sons, 2005.
- Rong Zheng, Jennifer C. Hou, and Lui Sha, Performance analysis
of power management policies in wireless networks,
IEEE Trans. on Wireless Communications, Vol. 5, No. 6,
pp. 1351--1362, June 2006.
- Ning Li and Jennifer C. Hou, "A scalable, power-efficient broadcast
algorithm for wireless networks," ACM Baltzer Wireless Networks (WINET),
Vol. 12, No. 4, pp. 495--509, August 2006.
- Rong Zheng, Jennifer C. Hou, and Lui Sha, "Optimal Block Design for
Asynchronous Wakeup and Its Applications in Multi-hop Wireless Networks,"
IEEE Trans. on Mobile Computing, Vol. 5, No. 9, pp. 1228--1241, September 2006.
- Rong Zheng and Jennifer C. Hou, ``Power management and control in wireless
networks,'' Ad Hoc and Sensor Networks, Yi Pan and Yang Xiao (Eds),
Nova Science Publishers, 2006.
-
Chunyu Hu, Rong Zheng, and Jennifer C. Hou, ``A microscopic study
of power management in IEEE 802.11 wireless networks,'' International
Journal of Wireless and Mobile Computing, special issue on Medium
Access Control for WLANs, WPANs, Ad Hoc Networks, and Sensor Networks,
to appear.
-
Fan Ye, Honghai Zhang, Songwu Lu, Lixia Zhang, and Jennifer C. Hou, "A randomized
energy-conservation protocol for resilient sensor networks."
ACM Wireless Network (WINET), Volume 12, Number 5, pp. 637-652, October 2006.
- Guanghui He and Jennifer C. Hou, ``Tracking targets with quality in wireless sensor
networks,'' ACM Trans. on Sensor Networks, to appear.
- Hyuk Lim and Jennifer C. Hou, ``Localization in wireelss sensor networks,''
Wireless Communications and Mobile Computing, special issue on "Distributed
systems of sensors and actuators," to appear (invited).
Funding Source
Related Links
Detailed Description of Research Agenda
A. A unified framework for
designing, and reasoning the effectiveness of, sensor networks
The issues of effectively coordinating and controlling sensor devices for
information dissemination have recently attracted much attention. However,
existing work usually limits its focus to one specific problem. Such a
specialization is advantageous if the functionalities required (or to be
realized) can be clearly divided among layers. Unfortunately this is not the
case in sensor networks, as there exist several issues, such as
application-aware data transport, combined cluster formation and cluster-based
routing, tradeoff between power management and routing efficiency, that cannot
be solved/handled in a single layer, but require cross-layer support. As a
matter of fact, for any scheme to be viable in sensor networks, it must consider
the close relationship between topology/power control, routing, medium access
control, and physical layer characteristics, and ensure compatibility among
component solutions across the layers. To this end, we have surveyed several
application scenarios of data-centric sensor networks, define an abstract
problem for information dissemination in sensor networks, and identify a set of
requirements and objectives. Candidates for key requirements include: efficiency
(defined as the non-redundant data bits that arrives timely/power used
(bits/watts)), robustness (defined as the number of node failures that the
network can sustain), and schedulability (defined as the link utilization below
which the deadline requirements of time-sensitive messages can be met). Then, we
configure the required functionalities into modules in/across layers, and figure
in their functional dependency. Under this unified framework (Figure
1), one can
design protocols specific to a layer without negligence of their interaction,
and compatibility, with protocols in other layers. Also, cross layer issues can
be identified and appropriately addressed.
B. Dynamic, tracking-based
cluster formation.
We have devised and evaluated a fully decentralized,
light-weight, dynamic clustering algorithm for target tracking. Instead of
assuming the same role for all the sensors, we envision a hierarchical sensor
network that is composed of (a) a static backbone of sparsely placed
high-capability sensors which will assume the role of a cluster head (CH) upon
triggered by certain signal events; and (b) moderately to densely populated
low-end sensors whose function is to provide sensor information to CHs upon
request. A cluster is formed and a CH becomes active, when the acoustic signal
strength detected by the CH exceeds a pre-determined threshold. The active CH
then broadcasts an information solicitation packet, asking sensors in its
vicinity to join the cluster and provide their sensing information (Figure
2).
Figure 2. An illustration of dynamic
clustering for tracking systems.
We address and devise solution approaches (with the use of Voronoi diagram) to
realize dynamic clustering:
(I1) how CHs cooperate with one another to ensure that only
one CH (preferably the CH that is closest to the target) is active with high
probability;
(I2) when the active CH solicits for sensor information,
instead of having all the sensors in its vicinity reply, only a sufficient
number of sensors respond with non-redundant, essential information to determine
the target location; and
(I3) both packets with which sensors respond to their CHs and
packets that CHs report to subscribers do not incur significant collision.
Through both probabilistic analysis and simulation, we show
with the use of Voronoi diagram, the CH that is usually closest to the target is
(implicitly) selected as the leader and and that the proposed dynamic clustering
algorithm effectively eliminates contention among sensors and renders more
accurate estimates of target locations as a result of better quality data
collected and less collision incurred.
C. Energy-aware,
utility-based data transport in wireless sensor networks.
We have formulated the problem of data transport in sensor
networks as an optimization problem whose objective function is to maximize the
amount of information (utility) collected at sinks (subscribers), subject to the
flow, energy and channel bandwidth constraints. In particular, we introduce
energy constraints and the notion of quality of data into the formulation. Also,
based on a Markov model extended from Bianchi's work, we derive the link delay
and the node capacity in both the single and multi-hop environments, and figure
them in the problem formulation. In contrast to most existing utility-based
approaches, our proposed approach not only considers energy constraints but also
differentiates treatment of packets with respect to their quality of
information. We show that the formulated optimization problem is general enough
to encompass a wide variety of applications in sensor networks, each with a
different objective function and subject to different constraints. We then
conduct three special case studies under the generic problem formulation. In
particular, we show in the first two cases that the issue of routing in an
environment monitoring system and the bandwidth allocation problem can be
treated as special cases under the generic problem formulation. In the third
case study, we derive an energy aware flow control solution, and investigate via
simulation its performance. The simulation results show that as compared with
the Ad hoc On Demand Distance Vector (AODV) routing and load balancing routing,
the solution derived under the proposed approach achieves higher utility and
incurs lower latency.
D. Power control and
management.
Statistical Traffic Prediction Approach to Improving IEEE
802.11 PSM
Power management is an important technique to prolong the
lifetime of battery-powered wireless ad hoc networks. The fact that the energy
consumed in the idle state dominates the total power consumed on wireless
network interfaces motivates use of protocols that put the radio into the sleep
mode in the absence of communications activities. IEEE 802.11 PSM is a
representative of such protocols. However, the performance of IEEE 802.11 (with
respect to the end-to-end delay) is significantly degraded under PSM because of
the wake-up latency thus introduced. In this part of the project, we propose a
novel and complementary mechanism, called Link-Indexed Statistical traffic
Predictor (LISP) to improve IEEE 802.11 PSM. Essentially LISP employs a simple,
light-weight traffic prediction method and enables each node to seek the
inherent correlation between ATIM_ACKs and incoming traffic. Once such a
correlation is identified, a node en route stays awake in the beacon interval
(BI) in which a packet is anticipated to arrive, thus building a ``freeway'' for
the packet to traverse the route. In this manner, a packet can travel from the
source to the destination within one BI ideally. Meanwhile, the number of duty
cycles is reduced and more energy conserved. LISP differs from previous work in
that it does not trade energy saving for better end-to-end performance at low to
moderate traffic loads. Instead more energy is saved because of the reduction in
duty cycles. We have conducted analytical and simulation studies, and
investigated the impact of traffic load, number of hops (of routes which
connections traverse), ATIM window size and packet sizes on the performance in
both tandem networks and networks of arbitrary topologies. LISP demonstrates a
performance improvement of 65-75% over PSM with respect to the end-to-end delay,
and 6% and 178% over IEEE 802.11 with and without PSM, respectively, with
respect to energy efficiency.
Asynchronous Wakeup Mechanism
Existing wakeup mechanisms falls into three categories:
on-demand wakeup, scheduled rendezvous, and asynchronous wakeup. In on-demand
wakeup mechanisms, out-band signaling is used to wake up sleeping nodes in an
on-demand manner. In scheduled rendezvous wakeup mechanisms, low-power sleeping
nodes wake up at the same time periodically to communicate with one another.
This is the mechanism used in IEEE 802.11 power saving mode (PSM). In
asynchronous wakeup mechanisms, each node follows its own wakeup schedule in
idle states, as long as the wakeup intervals among neighbors overlap. To meet
the requirement, nodes usually have to wake up more frequently compared to in
scheduled rendezvous mechanisms. On the flip side, asynchronous wakeup is easier
to implement and can ensure network connectivity even in highly dynamic
networks. In other words, asynchronous wakeup trades energy consumption for
robustness of network connectivity. A key challenge is to derive schedules that
have minimum idle state energy consumption with bounded neighbor discovery
latency. In this part of the project, we take a systematic approach to designing
asynchronous wakeup mechanisms in ad hoc networks. We intend to address the
following questions related to asynchronous wakeup mechanisms,
(1) Given a desirable delay for neighbor discovery, what is
the minimum percentage of time a node has to be awake?
(2) Does there exist an optimal schedule that can achieve
such minimum value?
(3) How to design a wakeup protocol using the optimal
schedule?
(4) How can power management be performed with asynchronous
wakeup?
To answer the former two questions, we first formulate the
problem of generating wakeup schedules as a block design problem in
combinatorics. We then derive the theoretical limit of the wakeup schedule and
give an optimal solution that can achieve minimum idle state energy consumption
with bounded neighbor discovery latency. To answer the latter two questions, we
study, after the theoretical base is laid, two protocol design issues: (i)
efficient implementation of the wakeup schedule and (ii) power management using
asynchronous wakeup. In the former issue, we design a neighbor discovery and
neighbor schedule bookkeeping protocol that can operate without requiring slot
boundaries be aligned. The protocol is also resilient to packet collision and
network dynamics. In the latter issue, we consider two design choices:
slot-based power management and on-demand power management, to determine how a
node transitions among different power management modes. In slot-based power
management power management states are managed slot by slot based the number of
buffered packets for a particular neighbor, while in on-demand power management
the transition between power management states are triggered by the
presence/lack of certain communication events. In both cases, a desirable
communication schedule can be overlaid over the wakeup schedule. To verify the
effectiveness of the design, we implement our proposed schemes using IEEE 802.11
MAC (without the power management component) in ns-2. Simulation studies
indicate that our wakeup schedule design guarantees that any two neighboring
hosts can detect each other in finite time without global clock synchronization.
In conjunction with the power management policies, the proposed wakeup protocol
can achieve communication efficiency comparable to the case without power
management while saving energy up to 70% (Figure 3)
|
