17-09-2016, 12:23 PM
1455107747-appwirlessmbcom11.doc (Size: 202 KB / Downloads: 5)
ABSTRACT
Cellular communication industry has grown leaps and bounds to become the essence of day to day science. Cell phone is the heart of cellular industry. The number of cell phone users is growing exponentially. But the privacy of user and security of cell phone remains a challenging question. Stealing cell phone has become the eye catching business for culprits. There should be a reliable method of protecting the instrument from burglars. Our paper proposes a new innovative, reliable, fool proof security measure for cell phones using VOICE RECOGNITION (a Biometric Approach).
Voice recognition is the identification of an individual identity using speech as the identifying characteristic. The voice of the user is processed (Speech Processing) using a digital signal processor which is the prime part of a cell phone. We are programming this DSP to implement this protection technique.
Firstly an original voice database of the user is created, and stored in the Flash ROM which is available inside the cell phone. Then whenever the user speaks through the cell phone, part of the speech sample is taken and stored. This is compared with the original database to check the identity of the user. If the user is authorized, he is allowed to continue his talk. If not the transmission is cut abruptly by making the DSP in idle state. Thus the cell phone is being protected from any unauthorized user. Even if the cell phone was stolen or missed it won’t be useful for any other person.
IMPLEMENTATION:
The analog input is got from the user and is converted to digitized form using A to D Convertor. In our implementation this digitized output is fed as input to a C program. This is compared with the already stored templates held in databases. If the comparison is positive, the user is proved to be authenticative. Else the connection cuts down abruptly.
INTRODUCTION:
The word Cell Phone has become the greatest buzz word in the Cellular Communication Industry. Cell phones are some of the most intricate devices people play with on a daily basis. Modern digital cell phones can process millions of calculations per second in order to compress and decompress the voice stream. The protection systems in the Cell phones are not good enough to give true security. Our paper tells about a new system of Cell phone protection using Voice Recognition.
NEED FOR CELL PHONE PROTECTION:
Cell phones have become an essential part of human day to day life. The number of cell phone users is increasing day by day. The threat for cell phones is also increasing. Stealing a cell phone and selling them in black market has grown into a profiting business. Hence there arises a need to adopt a fool proof security system to prevent these unlawful activities, unauthorized usage and safeguard the user.
EXISTING SECURITY MEASURES:
Cell phones of today have numerous security systems like pass word protection and personal identification number. But these measures are needed to be remembered and are not fool proof in the ever-growing technology. . In order that a password cannot be guessed, it should be as long as possible, not appear in a dictionary, and include symbols such as +, -, %, or #. Moreover, for security purposes, a password should never be written down, never be given to another person, and should be changed at least every three months. Besides these, anyone a seditive technician can break these passwords and PIN numbers. There should be a reliable method of protecting the instrument from burglars.
In our paper we have proposed a new innovative, reliable, fool proof security measure using VOICE RECOGNITION.
BIOMETRIC AUTHENTICATION:
Biometrics is the science of authentication and identification of a person on the basis of his/her physical characteristics. In dividing the authentication mechanism into “what you are?”, “what you have?”, “what you know?”, Biometrics fall under the category “what you are?”. It measures the physical characteristics and matches against a stored profile. The physical characteristics are closely bounded to a person and thought to be more reliable, not easily forgotten, lost, stolen, falsified or guessed.
Biometrics can be used in two main ways – 1.Authentication.
2. Identification.
The former falls under “one to one” configuration in which the characteristics of that particular person whose identity is to be checked for authentication is matched with his corresponding stored profile. If matches it allows the person to proceed his work. Else the service will be cut abruptly.
The later comes under “one to many” configuration in which that particular person’s characteristics is matched with all the stored biometric templates and identity is found.
.By determining an individual's physical features in an authentication inquiry and comparing this data with stored biometric reference data, identification for a specific user can be determined and authentication for access can be granted. It has many features as ideal for the definite identification of even an identical twin. Therefore, the time has come to replace the password with a more user friendly solution -- biometric authentication.
Although all biometrics are effective and virtually fraud-proof, an individual's speech pattern has more unique qualities and is often recognized as the most natural. It is based on the frequency analysis spectrum of someone’s voice identification nearly as accurate as finger prints.
VOICE RECOGNITION:
Voice recognition is the identification or verification of an individual identity using speech as the identifying characteristic. There is no attempt in voice recognition to necessarily identify the content, or speech, of a sound-stream, merely to identify its auditory and vocal characteristics. An individual speech spectrum is of the form as shown here.
In this case the sound signal is digitized and then the digitized signal is compared to previously recorded samples held in a database. The process of recording the samples to be held in a database is called ENROLLMENT and the recorded samples are generally called Biometric templates. The process of comparison is based on the algorithms and analysis softwares. The result is a simple yes/no decision as to whether the speaker has been identified.
CELL PHONE ARCHITECTURE:
The basic internal architecture cell phone mainly consists of the following parts.
Kernel which is nothing but a microprocessor.
Vocoder is a DSP (digital signal processor).
Memory (SRAM, Flash RAM,ROM,&PROM)
RF interface for transmitting and receiving of data.
Base band interface.
System clock (an oscillator producing frequency)
SIM card interface
HEART OF CELL PHONE:
Vocoder is the heart of cell phone. The vocoder is nothing but a Digital signal Processor. The main purpose of this vocoder is for voice encoding and decoding functions. In our paper we recognize the particular person speech by means of the acoustic resonances of the vocal tract modulate the spectra of the sources. Different speech sound corresponds uniquely to different spectral shapes. Vocoders depend on a parametric description of the vocal tract transfer functions. There are various forms of vocoder
Channel vocoder
Linear Predictive Coefficients vocoder(LPC)
Formant vocoder
Autocorrelation vocoder
FEATURES OF VOCODER:
For digital transmission, the number of bits per correlation sample used by a time domain vocoder should be about twice as high for spectral samples in frequency domain vocoders. Therefore time domain vocoders are not desirable. Hence we are going for frequency vocoder. The features of this vocoder are
1. Complete GSM voice encoding and decoding functions
2. Fully asynchronous coding and decoding
3. A-law PCM conversion
4. Programmable microphone amplifier with two differential input ports
5. Two programmable audio output amplifiers
6. A master clock
7. DTX function.