ZEAL FOR ZENITH...........

My Software Design: At a Glance


Scientific Calculator Using Graphical User Interface


This is a basic Graphical User Interface (GUI) design that is designed by me. This is a scientific calculator and there are options for base functions and logarithmic functions. For scientific, base or logarithmic operations, one has to select the buttons. But for normal algebraic operations, ADD, SUB, MUL and DIV buttons are sufficient enough to execute the results. For that calculator I have used MatLab version R2009a.


Voice Verification Software for User Authentication


For large monitory transactions, added security scheme is incorporated in the proposed banking system. We propose a voice authentication scheme for client verification to ensure safe data transmission between the client and the mobile server. During the process of registration the client will be offered the voice authentication. If the client agrees to add this enhanced security, he will be receiving a call from the mobile server requesting to utter the PIN code in his own voice, which will be recorded as a template for voice authentication. The client’s voice will be saved into the mobile server voice database. During the time of large monitory transaction, mobile server will proceed only after successful verification of client’s voice. The verification task will be carried out at the mobile server. For a faster operation, we propose to use only linear prediction coding (LPC) parameters as a voice feature.


Server Design for GSM Banking

For banking transactions only via GSM technology, I have used only text messages or SMS. Here I have designed two servers: Mobile server and banking server. At the very commencing, client request will be forwarded to mobile server. Mobile server has particular segments. The options are: Text server, voice database, user caller option, online communication using http and watermarking section. Another server is Banking server which have the segments of Account Database, transaction confirmation and updating database, http and watermarking decoding end. I this process, client will send his/her SMS containing transaction info to the mobile server. At mobile server it will be watermarked for the security. Then this watermarked SMS will be forwarded to Banking server. After reaching this, it will be decoded by watermarking software. Then according to the transaction request, clients account will be updated and functioned. For the database and server design, I have used Java and graphical user interface.


Audio Watermarking Software Design


Digital watermarking, which is a well established technology for copy control and media identification, is an effective methodology for hiding one information (text, image, audio or video) into another. The common practice of digital watermarking suggests that the information of interest is preferably hidden into an audio or video. A small audio file can be a good carrier for hiding SMS information. Instead of sending the SMS directly to the banking server, we propose to use the desired SMS as a watermark within an audio file. In view of developing a more secured audio watermarking, instead of directly working in time domain, embedding operation is carried out in frequency domain. We propose to use direct sequencing in the 4th layer LSB, which offers a greater imperceptibility and robustness against attack. DCT domain watermarking offers better robustness against noise and attacks. Data integrity is the main concern in m-commerce. From this perspective, here we have analyzed the immunity of the proposed watermarking scheme against pseudorandom noise. Considering fixed sequencing, the effect of noise with different signal to noise ratio (SNRs) is observed at several LSB layer entries of information bit. It is obvious that bit error rate (BER) decreases significantly with the higher layer of LSB used to embed information bit.


Image Watermarking Software Design


We propose a technique using digital image watermarking to make the SMS information non transparent to an eavesdropping intruder while passing through the ISPs server broadband internet services. An image of JPEG format contains three matrices of RGB information which are chosen as the target for watermark insertion. The three matrices, namely red, green, and blue, basically contain integer numbers ranging from 0 to 255 representing different pixels. Choice of matrices and sequence of choosing depends on encryption algorithm. On the other hand the information signal to be watermarked is plain text. Each text characters are defined by an ASCII value which requires 7 bits to be symbolized in binary. There are 160 characters in a single SMS needing 1120 bits (160x7) for binary ASCII representation. These bits are the information signal to be inserted as watermark in the integer RGB values sieved from the chosen matrix of the image under consideration. In the proposed scheme, instead of working directly on spatial domain, in order to enhance the level of security, we have performed watermarking in the discrete cosine transform. In the proposed method, watermarking is
performed on the LSB (least significant bit) layer of the host signal, which provides very low error and distortion. After performing the watermarking, inverse DCT (IDCT) is operation is carried out to obtain the image with watermark bit embedded in it. On the receiving end, data bit is extracted from the LSB of the selected DCT coefficients of the image signal.


Text Watermarking Software Design


The SMS text sent by the client can be embedded into any audio, video or image file while forwarding from mobile server to banking server via internet. If any intruder or so called ‘hacker’ wants to snatch the banking transaction information, he would be baffled. Even if he would know there is hidden information, it would be almost impossible to extract them. We can use two techniques for data hiding.-Standard LSB method or 4th bit LSB method. In standard LSB method the information text is converted into a binary stream and this stream is inserted into the LSB of the sample values of carrier file (audio, video, image etc.). But it is unwise in a way that LSB method is frail to noise attack. Again if we use all sample values sequentially it will be easier for an intruder to decode the information text. So from the viewpoint of robustness to attack, we emphasize on 4th bit LSB algorithm. However, considering the obscurity to intruder we propose to employ a random sequencing in selecting sample values to be watermarked. Here’s a drawback of using 4th LSB as it may cause a change of sample values as large as 16 (2^4) which in turn causes defect in carrier file and raise suspicion to the intruder.


Basic Watermarking Software Design


This software is designed based on basic watermarking technology. It is the developed version of the previously watermarking software. In that software, image, audio or text any one can be used. It is the prime feature of this software. Design principle and algorithms are approximately same. Just like all other watermarking software, it also has two parts: encoding and decoding parts. For that software, I have used MatLab.

Topic: Software Devalopment


Nice work

Bravo...Keep it up

Search site

© Syed Bahauddin Alam, 2011