UPM Institutional Repository

Efficient and secured compression and steganography technique in wireless sensor network


Tuama, Ammar Yaseen (2016) Efficient and secured compression and steganography technique in wireless sensor network. Masters thesis, Universiti Putra Malaysia.


Wireless Sensor Networks (WSNs) have emerged as one of the most promising so- lutions for wireless communication. They can be used in a wide variety of appli- cations ranging from military tasking, healthcare servicing, disaster prediction and indoor positioning. However, the need to use less complex and low-cost sensor de- vice results in constraints in computational power, communication bandwidth, and operational energy. In fact, the growing demands for new and much complex WSN applications require optimising both efficiency and security of data communication archetype in order to counterbalance their intrinsic limitations. In this study, to ad- dress these issues, we propose two techniques, one for minimising the transmitted data size in order to improve the efficiency of the WSN and the other for securing the sensed data transmission. First, the new data compression algorithm is proposed for compressing sensed data before it gets transmitted to the sink. The proposed solution is designed to be less complicated, low energy consumption and resource efficient with the ability to provide a lossless compression for a variety of data size. We analyse the solution and compare with a range of well-known algorithms in terms of compression ratio, memory usage, the number of instructions, compression speed and energy consumption. Two datasets have been used in the experiment, generated data set and Harvard Sensor Lab data set, in order to validate the performance of the proposed solution. The result shows that the proposed solution can compress both small and large data efficiently with up to 60% compression rate, 10 times faster compression speed and 4 times lower energy consumption compared to existing al- gorithms. Second, an improved steganographic algorithm based on the infamous Least Significant Bit (LSB) is proposed for hiding the sensed data scheduled for transmission. The proposed solution comes with low complexity and is used to en- hance the security of the standard LSB algorithm by replacing an originally less secured sequential data hiding with a random pixel selection. This random pixel se- lection is achieved via the use of an Elliptic Curve equation. In terms of security, the proposed solution is studied against brute-force attacks and the analysis shows that the new algorithm can withstand this type of attack with an ample amount of hiding possibilities that make the process of retrieving the message extremely diffi- cult. Furthermore, some analyses on hiding quality show that our algorithm retains the cover image quality as high as that of standard LSB algorithm. Apart from being able to work with various limitations of the sensor node, both techniques can pre- serve the resource without sacrificing the performance of the nodes, security level of the data and lifetimes of the WSN, and therefore are good candidates for future implementation into the sensor node.

Download File

FSKTM 2016 24 IR.pdf

Download (2MB) | Preview

Additional Metadata

Item Type: Thesis (Masters)
Subject: Wireless sensor networks
Subject: Confidential communications - Data processing
Call Number: FSKTM 2016 24
Chairman Supervisor: Mohamad Afendee Bin Mohamed, PhD
Divisions: Faculty of Computer Science and Information Technology
Depositing User: Ms. Nur Faseha Mohd Kadim
Date Deposited: 10 Jul 2019 01:25
Last Modified: 10 Jul 2019 01:25
URI: http://psasir.upm.edu.my/id/eprint/69360
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item