International Journal of Engineering Trends and Technology (IJETT) - Volume4 Issue6- June 2013

Design For The Automation Of Inter-Networked Banking Using Universal Sim

T.Suganya Thevi, Assistant Professor, Department of ECE, PSNACET,Dindigul. suganyaathevi@gmail.com Contact no : 97510 56856 C. Chitra, Professor, Department of ECE, PSNACET,Dindigul. chitrasrivi@gmail.com Contact no:97504 40704

ABSTRACT Automated teller machines (ATMs) are well known devices typically used by individuals to carry out a variety of personal and business financial transactions and/or banking functions. ATMs have become very popular with the general public for their availability and general user friendliness. But the drawback in the existing ATM system is that user should carry the ATM card without fail. And also ATM card can be misused by the unauthenticated person. To overcome this problem, creation of new generation ATM machine can be operated without ATM card. In this proposed system the image recognition technique can provide the important function of security maintenance such as identification of users and surveillance of the ATM environment. By using this system need of ATM card is completely eliminated. We can operate the ATM machine by using our SIM itself. 1. INTRODUCTION

When we insert our SIM in the reader unit of the ATM machine it transfers the mobile to the server. In the server, related information of the mobile number, (i.e) users account details, their photo etc is collected. The camera near the ATM machine will capture the users image and compare it with the user image in the server using MATLAB. Only when the image matches it asks the pin number and further processing starts. Otherwise the process is terminated. So by using this system need of ATM card is completely eliminated. Malfunctions can also be avoided. Our transaction will be much secured. Mobile scanning device scans SIM number through GSM Modem. Collected data is given to the Teller machine for further processing. At the same time, web camera captures the images and compares using digital signal processing. If images and PIN number are same then further processing is continued. Otherwise it gives alarm through alert module to the users mobile. 2. IMPLEMENTATION

Existing ATMs are convenient and easy to use for most consumers. Existing ATMs typically provide instructions on an ATM display screen that are read by a user to provide for

interactive operation of the ATM. Having read the display screen instructions, a user is able to use and operate the ATM via data and information entered on a keypad. However the drawback in the existing system is that the user should carry their ATM card without fail. But in many cases we forget it. This system helps us to use the ATM machine without the ATM card. By using this system, ATM machine can be operated by our SIM in the mobile phone.

To implement this automation of internetworked banking, microcontroller is programmed. Now it is wondering that the nonmentioning of memory space meant for the program storage is the most important part of any embedded controller. Originally this 8031 architecture was introduced with on chip, one time programmable version of Program Memory of size 4K X 8. Intel delivered all these microcontrollers (8051) with users program fused inside the device. The memory portion was mapped at the lower end of the Program Memory area. But, after getting devices, customers couldnt change anything in their program code, which was already made available inside during device fabrication. So, very soon Intel introduced the 8031 devices (8751) with re-programmable

ISSN: 2231-5381

http://www.ijettjournal.org

Page 2417

International Journal of Engineering Trends and Technology (IJETT) - Volume4 Issue6- June 2013

type of Program Memory using built-in EPROM of size 4K X 8. Like a regular EPROM, this memory can be re-programmed many times. Later on Intel started manufacturing these 8031 devices without any on chip Program Memory. The AT89S51 is a low-power, high-performance CMOS 8-bit microcontroller with 4K bytes of In-System Programmable Flash memory. The device is manufactured using Atmels highdensity nonvolatile memory technology and is compatible with the Indus-try-standard 80C51 instruction set and pin out. The on-chip Flash allows the program memory to be reprogrammed in-system or by a conventional nonvolatile memory pro-grammar. By combining a versatile 8-bit CPU with In-System Programmable Flash on a monolithic chip, the Atmel AT89S51 is a powerful microcontroller which provides a highly-flexible and cost-effective solution to many embedded control applications. The AT89S51 provides the following standard features: 4K bytes of Flash, 128 bytes of RAM, 32 I/O lines, Watchdog timer, two data pointers, two 16-bit timer/counters, a five-vector two-level interrupt architecture, a full duplex serial port, on-chip oscillator, and clock circuitry. Since universal SIM is used for the communication of information between the ATM and the server, a wireless modem is needed. Here GSM modem is used. A GSM modem is a wireless modem that works with a GSM wireless network. A wireless modem behaves like a dial-up modem. It can be an external device or a PC Card / PCMCIA Card. Typically, an external GSM modem is connected to a computer through a serial cable or a USB cable. A GSM modem in the form of a PC Card / PCMCIA Card is designed for use with a laptop computer. It should be inserted into one of the PC Card / PCMCIA Card slots of a laptop computer. Like a GSM mobile phone, a GSM modem requires a SIM card from a wireless carrier in order to operate. As mentioned in earlier sections of this SMS tutorial, computers use AT commands to control modems. Both GSM modems and dial-up modems support a common set of standard AT commands. You can use a GSM modem just like a dial-up modem. In addition to the standard AT commands, GSM modems support an extended set of AT commands. These extended AT commands are defined in the GSM standards. With the extended AT commands, you can do things like Reading, writing and deleting SMS messages.

Sending SMS messages. Monitoring the signal strength. Monitoring the charging status and charge level of the battery. Reading, writing and searching phone book entries. GSM provides recommendations, not requirements. The GSM specifications define the functions and interface requirements in detail but do not address the hardware. The reason for this is to limit the designers as little as possible but still to make it possible for the operators to buy equipment from different suppliers. The GSM network is divided into three major systems: the switching system (SS), the base station system (BSS), and the operation and support system (OSS). The basic GSM network elements are shown in below figure 1.

Fig.1 GSM Network Elements The types of software tools used here are, MPLAB IDE MATLAB PCA Algorithm ORCAD - Capture 2.1 PCA ALGORITHM RECOGNITION SYSTEM FACE

Automatic face recognition systems try to find the identity of a given face image according to their memory. The memory of a face recognizer is generally simulated by a training set. In this project, our training set consists of the features extracted from known face images of different persons. Thus, the task of the face recognizer is to find the most similar feature vector among the training set to the feature vector of a given test image. Here, the identity of a person where an image of that

ISSN: 2231-5381

http://www.ijettjournal.org

Page 2418

International Journal of Engineering Trends and Technology (IJETT) - Volume4 Issue6- June 2013

person (test image) is given to the system is recognized. PCA is itself a feature extraction algorithm. In the training phase, the feature vectors for each image are extracted in the training set. Let A be a training image of person A which has a pixel resolution of M N (M rows, N columns). In order to extract PCA features of A, the image is converted into a pixel vector A by concatenating each of the M rows into a single vector. The length (or, dimensionality) of the vector A will be M N. Use the PCA algorithms a dimensionality reduction technique which transforms the vector A to a vector A which has a dimensionality d where d M N. For each training image i, you should calculate and store these feature vectors i . In the recognition phase (or, testing phase), you will be given a test image j of a known person. Let j be the identity (name) of this person. As in the training phase, you should compute the feature vector of this person using PCA and obtain j. In order to identify j, you should compute the similarities between j and all of the feature vectors is in the training set. The similarity between feature vectors can be computed using Euclidean distance. The identity of the most similar i will be the output of our face recognizer. If i = j, it means that we have correctly identified the person j, otherwise if i j, it means that we have misclassified the person j. Schematic diagram of the face recognition system that will be implemented is shown in Figure 2.

3.1.2 USE OF PCA The use of PCA as a feature extractor is explained. Assume that we have p training images: i, i = 1,2,, p. For each training image, you should form pixel vectors i where i k<k;(k = M N). Our aim is to compute feature vectors i where i < d, (d k). In order to apply PCA to the training set, you should first form a training data matrix A which contains p rows: at each row is are stored. Thus the dimensionality of A is p k. First, you should compute the covariance matrix of A: CA. Then the eigen values and their corresponding eigenvectors of CA should be computed. There will be k eigen value and eigenvector pairs where each eigenvector is of dimensionality k. Sort the eigen values in decreasing order, and select the biggest d eigen value and eigenvector pairs. Form the transformation matrix by simply putting the selected eigenvectors as columns of . You will use to compute is from is. The computation of i is simply by i= TiT (1) where Tand iT are the transposes of and i, respectively. Note that each column of corresponds to an eigenvector which is of length k. This is equal to M N which is the dimensionality (resolution) of input images. Thus, you can convert each eigenvector to an image by reversing the concatenation operation. These converted eigenvector images are called eigen faces since they are similar to human faces. Figure 2 shows 20 eigen faces that correspond to the largest 20 eigen values of the ORL face database. Once you obtain is using the largest d eigen vectors, you can reconstruct the image of person i. If you use all k eigen vectors instead of d when forming , the reconstructed image will the same as image i.However, since our aim is dimensionality reduction and d k, reconstructed image i will be an approximation of the actual i. You can reconstruct i by converting the pixel vector: i = (i) T to an image of resolution MN. Figure 2 shows the reconstructed images of two persons using different number of eigenvectors. Notice that if you use more eigen vectors, then the reconstructed image is more similar to the original face image.

Fig.2. Schematic diagram of a face

ISSN: 2231-5381

http://www.ijettjournal.org

Page 2419

International Journal of Engineering Trends and Technology (IJETT) - Volume4 Issue6- June 2013

The implementation details of PCA are given below. You will use the ORL face database which contains 10 different images of each of the 40 subjects. For each subject five images (instances) will be put into training set and the rest of the images will be put into the test set. Training and test set images will be under train images and test images directories, respectively. A sample MATLAB code will be provided to you which automatically reads the images under these directories. The pseudo-code of MATLAB code is as follows: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. the sample

16. 17.

Set image resolution parameter (im_res) Set PCA dimensionality parameter (PCA_DIM) Read training images Form training data matrix (M_train_data) Form training class labels matrix (M_train_labels). Calculate PCA transformation matrix (tmatrix). Calculate feature vectors of all training images using tmatrix. Store training feature vectors in a matrix. Read test faces. For each test face do step 4 to 8. Compute the distances between test feature vector and all training vectors. Store the distances together with the training class labels Initialize error count to zero. For each test face do step 1 to 8. Using the distance data, determine the person ID of the most similar training vector. If the found ID is not equal to the ID of the test image increment error count. Output the correct recognition accuracy : (1 - (error count/ total test image count))*100 recognizer.

Web cam window will open automatically. Press enter button to take input image. Add database image in the same manner. A dialog box will be open. It shows the message database successfully added to database. PCA algorithm compares the input and database image. If the two images are same then the Message Authorized person will be shown. Then it will ask bank name. Select the bank and click sign in. Give pin no and click sign in. In the next step, if we go with withdraw option, it will ask to enter the amount to withdraw. Enter the amount and click process. Then the transaction will be completed. If the two images do not match it will show the message Not authorized and ask the port id and gsm id number. It is shown in figure 3.

Fig. 3: Unauthorized person In case of emergency, we may send our relations or friends to take money. If it is not authorized, a random pin no will be automatically sent to our mobile. We can give the pin number to them for authentication. Thus the transaction will be completed successfully. The completed transaction is shown in figure 4.

4. RESULTS & DISCUSSIONS The steps to be done to perform this design are discussed below. Open the MATLAB file and run the program. A window will be open. Click the Input button to give the input image.

ISSN: 2231-5381

http://www.ijettjournal.org

Page 2420

International Journal of Engineering Trends and Technology (IJETT) - Volume4 Issue6- June 2013

SMC , Vol.SMC-6, No.8, Oct. 1977, pp. 562-570. 2. M. Ejiri, H. Yoda, H.Sakou and Y. Sakamoto, Knowledge-Directed Inspection for Complex Multilayered Patterns, Machine Vision and Applications, vol.2, no.3, 1989, pp.155166H. Jain, R. Bolle, and S. Pankati, Biometrics: personal identification in networked society , Kluwer Academic Publishers, 1998. H. Sako, M. Seki, N. Furukawa, H. Ikeda and A. Imaizumi, Form Reading based on Form-type Identification and Form-data Recognition, Proc. of International Conference on Document Analysis and Recognition (ICDAR 2003), 2003, pp. 926-930. S. Hoque, M.C. Fairhurst, F. Deravi, W.G.J. Howells, On the Feasibility of Generating Biometric Encryption Keys IEE Electronics Letters, 41(6), 309-311, 2005. Cellan-Jones, R., London starts digital cash trial in Technology correspondent, BBC News Online, http://news.bbc.co.uk/1/hi/tec hnology/7117213.stm. 2007: UK. H. Fujisawa, M. Koga, H. Ogata, T. Kagehiro and H. Sako, Recognition Strategies for Japanese Mailpiece Recipient Addresses, Proc. of 2nd International Conference on Multimodal Interface (ICMI '99), 1999, pp. III6974.

3.

Fig. 4: Completed transaction CONCLUSION Today, single factor authentication, e.g. passwords, is no longer considered secure in the internet and banking world. Easy-to-guess passwords, such as names and age, are easily discovered by automated password-collecting programs. Two factor authentication using mobile phones has recently been introduced to meet the demand of organizations for providing stronger authentication options to its users. The proposed system has two option of running, either using a free and fast connection-less method or a slightly more expensive SMS based method. Both methods have been successfully implemented and tested, and shown to be robust and secure. The system has several factors that make it difficult to hack. REFERENCES 1. S. Kashioka, M. Ejiri and Y. Sakamoto,A Transistor Wire-Bonding System Utilizing Multiple Local Pattern Matching Techniques, IEEE Trans. On

4.

5.

6.

7.

ISSN: 2231-5381

http://www.ijettjournal.org

Page 2421