Mobiilne biomeetriline tuvastusliides

Abstract

Autentimine arvutivõrkudes on protsess, mis üritab kontrollida digitaalset identiteeti, kasutades paroole. Parooli olemasolu aitab tagada kasutaja autentsuse. Kasutajaid saab biomeetiriliselt kindlaks teha ka nende sõrmejälje, hääle või näo järgi kasutades kindlat riistvara ja parooli toimingu kinnitamiseks. Samas, juhul kui paljud kasutajad on seotud sedatüüpi autentimisega, võib selline lahendus muutuda liiga aeglaseks ja suhteliselt kulukaks. Seetõttu vajab edukas biomeetriline autentimise süsteem põhjalikku läbimõeldust paljude tegurite osas (PBworks, 2007).

Minu magistritöö eesmärk on uurida erinevaid biomeetrilise autentimise meetodeid, luues kirjeldused mobiilseks kasutamiseks ning töötada välja ja realiseerida vastav prototüüp. Kui biomeetrilise testi andmed on kogutud, näiteks video või hale kaudu, siis saadud andmed läbivad töötluse, peale mida kasutatakse ainult unikaalset osa andmetest. See osa andmetest salvestatakse andmebaasi. Kuid, enne andmete kasutamist, tuleb telefonis rakendada mitmeid meetmeid.

Biomeetria puhul kasutatakse kahte peamist stsenaariumit: kontrollimine ja isikuandmete tuvastamine. Kontrollimise puhul võrreldakse ainult ühte andmebaasi, mis on kirjeldatav kui üks-ühele võrdulus. Isikuandmete tuvastamisel kasutatakse üks-mitmele andmebaaside võrdlust. Mõlemal juhul peavad salvestatud andembaasid toetama vastavaid stsenaariume. Magistritöös valmis vastav kasutajaliides nõutud omadustega.

Mobile Biometric Challenge Recording Interface

In computer network, authentication is the process of attempting to verify the digital identity using login passwords. Knowledge of the password is presumed to guarantee that the user is authentic. Likewise, Users may biometrically authenticate via their fingerprint, voice, face, or using provided hardware and password to confirm the operation. However, when many of users are involved this type of authentication is not suitable because this solution may become too slow and relatively expensive. In brief, successful distribution of a biometric authentication system needs careful deliberation of many factors (PBworks, 2007). My thesis core will be to study different biometric authentication techniques, derive requirements from these for a mobile interface, and implement and refine an interface prototype. When a biometric sample or trait is acquired, such as a video or a voice sample, the captured data goes through some amount of processing to prepare it for the extraction of a relatively small set of numbers, which represent the unique aspects of the data. This extracted set of data in record called a recording template. However, to prepare this recording optimally for biometric algorithms several measures already have to be taken on the phone. There are two basic usage scenarios for biometrics, known as verification and identification. Verification is the scenario where a newly created template is compared with only one other template in a database, which is described as an one-to-one comparison. The second biometric usage scenario is called identification, which is used when it is necessary to compare a newly created template too many enrolled templates, which is described as an one-to-many comparison.