Optimum Power Allocation in Sensor Networks for Active Radar Applications
Abstract:
We investigate the power allocation
problem in distributed sensor networks that are used for target object
classification. In the classification process, the absence, the presence, or
the type of a target object is observed by the sensor nodes independently. Since these local observations are
noisy and thus unreliable, they are fused together as a single reliable
observation at a fusion center. The fusion center uses the best linear
unbiased estimator in order to accurately estimate the reflection coefficient
of target objects. We utilize the average deviation
between the estimated and the actual reflection coefficient as a metric for
defining the objective function. First, we demonstrate that the
corresponding optimization of the power allocation leads to a signomial program
which is in general quite hard to solve. Nonetheless, by using the proposed
system model, fusion rule and objective function, we are able to optimize the
power allocation analytically and can hence present a closed-form solution.
Since the power consumption of the entire network may be limited in various
aspects, three different cases of power constraints are discussed and compared
with each other.
Architecture:
Existing System:
In general, the objective is to
maximize the overall classification probability, however, a direct solution to
the allocation problem does not exist, since no analytical expression for the
overall classification probability is available. we will see later that only a
single feasible stationary point exists.
Disadvantages:
However, they still suffer
inherent shortcomings such as limited radio coverage, poor system reliability,
and lack of security and privacy.
Proposed System:
First, we demonstrate that the corresponding optimization of the
power allocation leads to a signomial program which is in general quite hard to
solve. Nonetheless, by using the proposed system model, fusion rule and
objective function, we are able to optimize the power allocation analytically
and can hence present a closed-form solution. Since the power consumption of the entire network may be limited
in various aspects, three different cases of power constraints are discussed
and compared with each other. In addition, a sensitivity analysis of the optimal power
allocation with respect to perfect and imperfect parameter knowledge is worked
out. proposed system is the first region in which 2 g 2 g
holds, while a system operation in the third region, for which 2 g 2 g holds, should be avoided Advantages:
In addition, some
advanced attacks, such as traffic analysis and flow tracing, can also be launched
by a malicious adversary to compromise users’ privacy, including source
anonymity and traffic secrecy.
Modules:
1.
Analytical power allocation.
2. energy-efficient optimization.
3.
Threat models.
4.
distributed target
classification.
5.
network resource management.
6.
information fusion.
7.
information fusion
1.
Analytical power allocation:
we investigate the power allocation
problem in distributed sensor networks that are used for active radar
applications. the power allocation
problem for distributed wireless sensor networks, which perform object
detection and classification, is only treated for ultra-wide bandwidth (UWB)
technology. For active radar
systems an optimal solution to the power allocation problem is only known for
high signal-to-noise ratios (SNRs), see. This limits the
usability of this criterion for solving the power allocation problem. power allocation scheme
is still an open problem in order to improve the overall classification
probability. we present the
analytical solution of the power allocation problem for the case where the
average transmission power of each SN is limited by the output power-range
limitation Pmax 2 R+. Finally, we extend the power allocation problem to the
case where both constraints simultaneously hold and present the corresponding
optimal solution.
2.
energy-efficient optimization:
Note that the sum-power constraint Ptot is a reasonable approach to
compare energy-efficient radar systems. In order to save
energy, the value of Ptot should be as small as possible, which means that a
single SN or only a few SNs are active. Hence, if the number ~K
of active SNs is satisfactory in order to accurately estimate all important
parameters of the target, then the best energy-aware value of Ptot is equal to
~K Pmax.
3.
Threat models:
We consider the following two
attack models. Outside
Attacker:
An outside attacker can be considered as a global passive eavesdropper who has
the ability to observe all network links. An outside attacker can examine the
tags and message content, and thus link outgoing packets with incoming packets.
Further, even if end-to-end
encryption is applied to messages at a higher layer, it is still possible for a
global outside attacker to trace packets by analyzing and comparing the message
cipher text. Inside
attacker:
An inside attacker may compromise several intermediate nodes. Link-to -link
encryption is vulnerable to inside attackers since they may already have
obtained the decryption keys and thus the message plaintext can be easily
recovered. Both inside and outside attackers
may perform more advanced traffic analysis/flow tracing techniques, including
size correlation, time correlation, and message content correlation.
4.
Enhanced Privacy
against traffic analysis and flow tracing:
In the classification process, the absence, the presence, or the
type of a target object is observed by the sensor nodes independently. the power allocation problem
for distributed wireless sensor networks, which perform object detection and
classification, is only treated for ultra-wide bandwidth (UWB) technology. The main difficulty for
optimizing the power consumption is associated with finding a closed-form
equation for the overall classification probability. The sensing task and its
corresponding communication task for a single classification process are
performed in consecutive time slots. we are able to separate the power allocation problem from the
classification problem and optimize both independently.
5.network resource management:
Since the power
consumption of the entire network may be limited in various aspects, three
different cases of power constraints are discussed and compared with each other
Currently,
distributed detection
is rather discussed in the context of wireless sensor networks, where the
sensor units may also be radar nodes a network of amplify-and-forward SNs. Based on a simple
system model, we apply a linear fusion rule and utilize the average deviation
between the estimated and the actual reflection coefficient as a metric for
defining the objective function.
Algorithm:
Homomorphic encryption
algorithm.
We will take the the encryption
method when necessary. Note that the computational overhead is counted
independent of the underlying network coding framework. Actually, the scalar multiplicatively can be deduced from the
addictively, since 𝐸(𝑥 ⋅ 𝑡) = 𝐸(Σ𝑡𝑖=1 𝑥). Where the addition on plaintext can be achieved by performing
a multiplicative operation on the corresponding cipher text, i.e., 𝐸(𝑥1 + 𝑥2) = 𝐸(𝑥1) ⋅ 𝐸(𝑥2). Further, the
following two equations can be easily derived:
𝐸(𝑥 ⋅ 𝑡) = 𝐸𝑡(𝑥) 𝐸(Σ𝑖 𝑥𝑖 ⋅ 𝑡𝑖) =Π𝑖 𝐸𝑡𝑖 (𝑥𝑖)
SYSTEM SPECIFICATION
Hardware Requirements:
v
System : Pentium IV 2.4 GHz.
v
Hard Disk : 40 GB.
v
Floppy Drive : 1.44 Mb.
v
Monitor :
14’ Colour Monitor.
v
Mouse : Optical Mouse.
v
Ram :
512 Mb.
Software Requirements:
v Operating system :
Windows 7 Ultimate.
v
Coding Language :
ASP.Net with C#
v
Front-End : Visual Studio 2010 Professional.
v Data Base : SQL Server 2008.
