BORDER SECURITY USING

WIRELESS INTEGRATED NETWORK SENSORS (WINS)

By

K.Krishna chaitanya &

B.Vijayakumar
R:no- 08121A1528 R:no-
08121a1505
3rd year B.tech 3rd year
B.tech
Dept. of CSSE Dept. of CSSE

Sree Vidyanikethan
Engineering College
[Autonomous]

Mail id: [email protected]

Mailid:[email protected]
Abstract: environmental monitoring. On a national scale,

Wireless integrated Network Sensors transportation systems, and borders will be

(WINS) now provide a new monitoring and monitored for efficiency, safety, and security.

control capability for monitoring the borders of 1. Introduction:

the country. Using this concept we can easily Wireless Integrated Network Sensors
identify a stranger or some terrorists entering (WINS) combine sensing, signal processing,
the border. The border area is divided into decision capability, and wireless networking
number of nodes. Each node is in contact with capability in a compact, low power system.
each other and with the main node. The noise Compact geometry and low cost allows WINS
produced by the foot-steps of the stranger is to be embedded and distributed at a small
collected using the sensor. This sensed signal fraction of the cost of conventional wire line
is then converted into power spectral density sensor and actuator systems. On a local,
and the compared with reference value of our wide-area scale, battlefield situational
convenience. Accordingly the compared value awareness will provide personnel health
is processed using a microprocessor, which monitoring and enhance security and
sends appropriate signals to the main node. efficiency. Also, on a metropolitan scale, new
Thus the stranger is identified at the main traffic, security, emergency, and disaster
node. A series of interface, signal processing, recovery services will be enabled by WINS.
and communication systems have been On a local, enterprise scale, WINS will create
implemented in micro power CMOS circuits. A a manufacturing information service for cost
micro power spectrum analyzer has been and quality control. The opportunities for
developed to enable low power operation of WINS depend on the development of
the entire WINS system. scalable, low cost, sensor network
Thus WINS require a Microwatt of
power. But it is very cheaper when compared
to other security systems such as RADAR
under use. It is even used for short distance
communication less than 1 Km. It produces a
less amount of delay. Hence it is reasonably
faster. On a global scale, WINS will permit
monitoring of land, water, and air resources for
architecture. This requires that sensor distributions with an emphasis on recovering
information be conveyed to the user at low bit environment information. Multihop
rate with low power transceivers. Continuous communication yields large power and
sensor signal processing must be provided to scalability advantages for WINS networks.
enable constant monitoring of events in an Multihop communication, therefore, provides
environment. Distributed signal processing
and decision making enable events to be
identified at the remote sensor. Thus,
information in the form of decisions is
conveyed in short message packets. Future
applications of distributed embedded
processors and sensors will require massive
numbers of devices. In this paper we have
concentrated in the most important
application, Border Security.

an immediate advance in capability for the

2. WINS System Architecture:
WINS narrow Bandwidth devices. However,
Conventional wireless networks are
WINS Multihop Communication networks
supported by complex protocols that are
permit large power reduction and the
developed for voice and data transmission for
implementation of dense node distribution. The
handhelds and mobile terminals. These
multihop communication has been shown in
networks are also developed to support
the figure 2. The figure 1 represents the
communication over long range (up to 1km or
general structure of the wireless integrated
more) with link bit rate over 100kbps. In
network sensors (WINS) arrangement.
contrast to conventional wireless networks, the
WINS network must support large numbers of
3. WINS Node Architecture:
sensors in a local area with short range and
The WINS node architecture (Figure 1)
low average bit rate communication (less than
is developed to enable continuous sensing,
1kbps). The network design must consider the
event detection, and event identification at low
requirement to service dense sensor
power. Since the event detection process must
occur continuously, the sensor, data converter,
data buffer, and spectrum analyzer must all
operate at micro power levels. In the event that
an event is detected, the spectrum analyzer sensor nodes that include sensor signal
processing, control, and a wireless network
interface. Distributed network sensor devices
must continuously monitor multiple sensor
systems, process sensor signals, and adapt to
changing environments and user requirements,
while completing decisions on measured
signals. For the particular applications of
military security, the WINS sensor systems

Continuous operation low duty cycle must operate at low power, sampling at low

Figure .The wireless integrated network frequency and with environmental background

sensor (WINS) architecture limited sensitivity. The micro power interface

circuits must sample at dc or low frequency
where “1/f” noise in these CMOS interfaces is
output may trigger the microcontroller. The
large. The micro power signal processing
microcontroller may then issue commands for
system must be implemented at low power and
additional signal processing operations for
with limited word length. In particular, WINS
identification of the event signal. Protocols for
applications are generally tolerant to latency.
node operation then determine whether a
The WINS node event recognition may be
remote user or neighboring WINS node should
delayed by 10 – 100ms, or longer.
be alerted. The WINS node then supplies an
attribute of the identified event, for example,
the address of the event in an event look-up-
4. WINS micro sensors:
table stored in all network nodes. Total Source signals (seismic, infrared, acoustic and

average system supply currents must be less others) all decay in amplitude rapidly with
radial distance from the source. To maximize
than 30µA. Low power, reliable, and efficient
detection range, sensor sensitivity must be
network operation is obtained with intelligent
optimized. In addition, due to the fundamental

Figure 2. WINS nodes (shown as disks)

 limits of background noise, a Figure 3.

Thermal Infrared Detector

5. Routing between nodes:
maximum detection range exists for any
The sensed signals are then
sensor. Thus, it is critical to obtain the greatest
routed to the major node. This routing is done
sensitivity and to develop compact sensors
based on the shortest distance. That is the
that may be widely distributed. Clearly,
distance between the nodes is not considered,
microelectromechanical systems (MEMS)
but the traffic between the nodes is
technology provides an ideal path for
considered. This has been depicted in the
implementation of these highly distributed
figure 4. In the figure, the distance between the
systems. The sensor-substrate “Sensorstrate”
nodes and the traffic between the nodes has
is then a platform for support of interface,
been clearly shown. For example, if we want to
signal processing, and communication circuits.
route the signal from the node 2 to node 4, the
Examples of WINS Micro Seismometer and
shortest distance route will be from node 2 via
infrared detector devices are shown in Figure
node 3 to node 4. But the traffic through this
3. The detector shown is the thermal detector.
path is higher than the path node 2 to node 4.
It just captures the harmonic signals produced
Whereas this path is longer in distance.
by the foot-steps of the stranger entering the
border. These signals are then converted into
their PSD values and are then compared with
the reference values set by the user.
 6.Shortest distance algorithm:

Figure 4. Nodal distance and Traffic

Figure 5. Subnet with line capacities
In this process we find mean packet delay, if
the capacity and average flow are known.
From the mean delays on all the lines, we
calculate a flow-weighted average to get mean
packet delay for the whole subnet. The weights
on the arcs in the figure 5 give capacities in
each direction measured in kbps.
 The tabular column listed below gives
waiting factor for each path.

7. WINS digital signal processing:

If a stranger enters the border, his foot-
steps will generate harmonic signals. It can be
detected as a characteristic feature in a signal
power spectrum. Thus, a spectrum analyzer
must be implemented in the WINS digital
Figure 6.s Routing Matrix signal processing system. The spectrum
In fig 6 the routes and the number of analyzer resolves the WINS input data into a
packets/sec sent from source to destination low-resolution power spectrum. Power spectral
are shown. For example, the E-B traffic gives 2 density (PSD) in each frequency “bins” is
packets/sec to the EF line and also 2 computed with adjustable band location and
packets/sec to the FB line. The mean delay in width. Bandwidth and position for each power
each line is calculated using the formula spectrum bin is matched to the specific
Ti =1/(µc-λ) detection problem. The WINS spectrum

Ti = Time delay in sec analyzer must operate at µW power level. So

C = Capacity of the path in Bps the complete WINS system, containing
µ = Mean packet size in bits
λ = Mean flow in packets/sec. controller and wireless network interface
components, achieves low power operation by
The mean delay time for the entire maintaining only the micro power components
subnet is derived from weighted sum of all the in continuous operation. The WINS spectrum
lines. There are different flows to get new analyzer system, shown in Figure 7, contains a
average delay. But we find the path, which has set of parallel filters.
the smallest mean delay-using program. Then
we calculate the Waiting factor for each path.
The path, which has low waiting factor, is the
shortest path. The waiting factor is calculated
using
W = λi / λ
λi = Mean packet flow in path
λ = Mean packet flow in subnet
WINS micropower spectrum analyzer WINS systems present novel
architecture. requirements for low cost, low power, short
range, and low bit rate RF communication.
Simulation and experimental verification in the
8. PSD Comparision:
field indicate that the embedded radio network
Each filter is assigned a coefficient set must include spread spectrum signaling,
for PSD computation. Finally, PSD values are channel coding, and time division multiple
compared with background reference values In access (TDMA) network protocols. The
the event that the measured PSD spectrum operating bands for the embedded radio are
values exceed that of the background most conveniently the unlicensed bands at
reference values, the operation of a 902-928 MHz and near 2.4 GHz. These bands
microcontroller is triggered. Thus, only if an provide a compromise between the power cost
event appears, the micro controller operates. associated with high frequency operation and
Buffered data is stored during continuous the penalty in antenna gain reduction with
computation of the PSD spectrum. If an event decreasing frequency for compact antennas.
is detected, the input data time series, The prototype, operational, WINS networks are
including that acquired prior to the event, are implemented with a self-assembling, multihop
available to the micro controller. The micro TDMA network protocol.
controller sends a HIGH signal, if the The WINS embedded radio
difference is high. It sends a LOW signal, if the development is directed to CMOS circuit
difference is low. For a reference value of technology to permit low cost fabrication along
25db, the comparison of the DFT signals is with the additional WINS components. In
shown in the figure 8. addition, WINS embedded radio design must
address the peak current limitation of typical
battery sources, of 1mA. It is critical, therefore,
to develop the methods for design of
micropower CMOS active elements. For LC

oscillator phase noise power, Sφ, at frequency

Figure 8. Comparator plot
offset of δω away from the carrier at

9. WINS Micropower embedded frequency ω with an input noise power,

Snoise and LC tank quality factor, Q, phase
radio:
noise power is:
 entering the border. The border area is divided
into number of nodes. Each node is in contact
Now, phase noise power, Snoise, at the with each other and with the main node. The
transistor input, is dominated by “1/f” noise. noise produced by the foot-steps of the
Input referred thermal noise, in addition, stranger are collected using the sensor. This
increases with decreasing drain current and sensed signal is then converted into power
power dissipation due to the resulting decrease spectral density and the compared with
reference value of our convenience.
Accordingly the compared value is processed
using a microprocessor, which sends
appropriate signals to the main node. Thus the
stranger is identified at the main node. A series
of interface, signal processing, and
communication systems have been
implemented in micro power CMOS circuits. A
micro power spectrum analyzer has been
developed to enable low power operation of
the entire WINS system. Thus WINS require a
Microwatt of power. But it is very cheaper
when compared to other security systems such
as RADAR under use. Wireless Integrated
Network Sensors (WINS) combine sensing,
signal processing, decision capability, and
in transistor transconductance. wireless networking capability in a compact,
The tunability of micropower CMOS low power system. Compact geometry and low
systems has been tested by implementation of cost allows WINS to be embedded and
several VCO systems to be discussed below. distributed at a small fraction of the cost of
The embedded radio system requires narrow conventional wireline sensor and actuator
band operation and must exploit high Q value systems. On a local, wide-area scale,
components. Wireless Integrated Network battlefield situational awareness will provide
Sensors (WINS) now provide a new monitoring personnel health monitoring and enhance
and control capability for monitoring the security and efficiency. Also, on a metropolitan
borders of the country. Using this concept we scale, new traffic, security, emergency, and
can easily identify a stranger or some terrorists disaster recovery services will be enabled by
WINS. On a local, enterprise scale, WINS will the entire WINS system. Thus WINS require a
create a manufacturing information service for Microwatt of power. But it is very cheaper
cost and quality control. The opportunities for when compared to other security systems such
WINS depend on the development of scalable, as RADAR under use. It is even used for short
low cost, sensor network architecture. This distance communication less than 1 Km. It
requires that sensor information be conveyed produces a less amount of delay. Hence it is
to the user at low bit rate with low power reasonably faster. On a global scale, WINS will
transceivers. Continuous sensor signal permit monitoring of land, water, and air

processing must be provided to enable resources for environmental monitoring. On a

constant monitoring of events in an national scale, transportation systems, and
environment. Distributed signal processing and borders will be monitored for efficiency, safety,
decision making enable events to be identified and security.
at the remote sensor.
References:
10. Conclusion: 1. www.techonline.com
A series of interface, signal processing,
2. www.ieee.org
and communication systems have been
implemented in micropower CMOS circuits. A 3. en.wikipedia.org/wiki
micropower spectrum analyzer has been 4. portal.acm.org/citation.cfm
developed to enable low power operation of