Toward a Statistical Framework forSource Anonymity in Sensor Networks
ABSTRACT:
In certain applications, the locations of events reported by a sensor network need to remain anonymous. That is,unauthorized observers must be unable to detect the origin of such events by analyzing the network traffic. Known as the sourceanonymity problem, this problem has emerged as an important topic in the security of wireless sensor networks, with variety oftechniques based on different adversarial assumptions being proposed. In this work, we present a new framework for modeling,analyzing, and evaluating anonymity in sensor networks. The novelty of the proposed framework is twofold: first, it introduces thenotion of “interval indistinguishability” and provides a quantitative measure to model anonymity in wireless sensor networks; second, itmaps source anonymity to the statistical problem of binary hypothesis testing with nuisance parameters. We then analyze existingsolutions for designing anonymous sensor networks using the proposed model. We show how mapping source anonymity to binaryhypothesis testing with nuisance parameters leads to converting the problem of exposing private source information into searching foran appropriate data transformation that removes or minimize the effect of the nuisance information. By doing so, we transform theproblem from analyzing real-valued sample points to binary codes, which opens the door for coding theory to be incorporated into thestudy of anonymous sensor networks. Finally, we discuss how existing solutions can be modified to improve their anonymity.
ARCHITECTURE:
EXISTING SYSTEM:
While transmitting the “description” of a sensed event ina private manner can be achieved via encryption primitives, hiding the timing and spatial information ofreported events cannot be achieved via cryptographicmeans.
Encrypting a message before transmission,for instance, can hide the context of the message fromunauthorized observers, but the mere existence of theciphertext is indicative of information transmission.
In the existing literature, the source anonymity problemhas been addressed under two different types of adversaries,namely, local and global adversaries. A localadversary is defined to be an adversary having limitedmobility and partial view of the network traffic. Routingbasedtechniques have been shown to be effective in hidingthe locations of reported events against local adversaries.
A global adversary is defined to bean adversary with ability to monitor the traffic of the entirenetwork (e.g., coordinating adversaries spatially distributedover the network). Against global adversaries, routingbasedtechniques are known to be ineffective in concealinglocation information in event-triggered transmission. This isdue to the fact that, since a global adversary has full spatialview of the network, it can immediately detect the originand time of the event-triggered transmission
DISADVANTAGES OF EXISTING SYSTEM:
The source anonymity problem in wireless sensor networksis the problem of studying techniques that providetime and location privacy for events reported by sensornodes. (Time and location privacy will be used interchangeablywith source anonymity throughout the paper.)
The source anonymity problem has been drawing increasingresearch attention recently.
PROPOSED SYSTEM:
In this paper, we investigate theproblem of statistical source anonymity in wireless sensornetworks. The main contributions of this paper can besummarized by the following points.
We introduce the notion of “interval in-distinguishability”and illustrate how the problem of statisticalsource anonymity can be mapped to the problem ofinterval indistinguishability.
We propose a quantitative measure to evaluatestatistical source anonymity in sensor networks.
We map the problem of breaching source anonymityto the statistical problem of binary hypothesis testingwith nuisance parameters.
We demonstrate the significance of mapping theproblem in hand to a well-studied problem inuncovering hidden vulnerabilities. In particular,realizing that the SSA problem can be mapped tothe hypothesis testing with nuisance parametersimplies that breaching source anonymity can beconverted to finding an appropriate data transformationthat removes the nuisance information.
We analyze existing solutions under the proposedmodel. By finding a transformation of observed data,we convert the problem from analyzing real-valuedsamples to binary codes and identify a possibleanonymity breach in the current solutions for theSSA problem.
We pose and answer the important research questionof why previous studies were unable to detect thepossible anonymity breach identified in this paper.
We discuss, by looking at the problem as a codingproblem, a new direction to enhance the anonymityof existing SSA solutions.
ADVANTAGES OF PROPOSED SYSTEM:
Removes or minimize the effect of the nuisance information
SYSTEM CONFIGURATION:-
HARDWARE REQUIREMENTS:-
Processor-Pentium –IV
Speed-1.1 Ghz
RAM-512 MB(min)
Hard Disk-40 GB
Key Board-Standard Windows Keyboard
Mouse-Two or Three Button Mouse
Monitor-LCD/LED
SOFTWARE REQUIREMENTS:-
Operating System: LINUX
Tool: Network Simulator-2
Front End: OTCL (Object Oriented Tool Command Language)
REFERENCE:
Basel Alomair, Member, IEEE, Andrew Clark, Student Member, IEEE,Jorge Cuellar, and Radha Poovendran, Senior Member, IEEE, “Toward a Statistical Framework forSource Anonymity in Sensor Networks”, IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 12, NO. 2, FEBRUARY 2013.