Multicolor Physical Unclonable Functions


Eşidir A., Kiremitler N. B., Kalay M., Önses M. S.

16th NANOSCIENCE & NANOTECHNOLOGY CONFERENCE, Ankara, Türkiye, 5 - 08 Eylül 2022, ss.375

  • Yayın Türü: Bildiri / Özet Bildiri
  • Basıldığı Şehir: Ankara
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.375
  • Kayseri Üniversitesi Adresli: Evet

Özet

In today's world, counterfeiting is a major problem. Increasing counterfiting cause economic loss, health risks, and social issues. For this reason, academic and industrial circles have focused on the development of new generation of security labels/barcodes to provide effective anticounterfeiting measures. Physical encoding systems called “physically unclonable functions” (PUFs) have attracted great attention in last decade, due to the potential to provide unique encoding for each object/application that even the manufacturers cannot repeat thanks to randomness of process. Therefore, in order to use this advantage of PUFs effectively, it is essential to develop multiplexed PUFs fabricated by versatile and cost effective fabrication techniques. In this study, an unique approach that reinforces the advantages of PUFs with a versatile fabrication method is presented. Ourapproach is based on the practical fabrication of patterns consisting of polymeric features  deposited at spatially random positions using the electrospraying. Here, electrohydrodynamic instabilities during electrospraying provide the formation of completely random and complex polymeric structures. The additive nature of the process enables sequential deposition of multiple materials on the same substrate. Furthermore, the solution processing-based fabrication allows incorporation of functional materials into polymeric features and enables the formation of additional security layers. By taking advantage of the advantages mentioned earlier and enriching the technique, multicolor PUFs with multi-layered security measures have been fabricated. Specifically, PUFs containing layered security measures from many different polymer systems such as polystyrene, polymethyl methacrylate, polyethylene oxide, and polyvinylpyrrolidone have been produced and their practical applications have been demonstrated. Furthermore, systematic examination of the process parameters, the condition-structure relationship was revealed where electrosprayed PUF structures display a wide variety of distinct morphologies as well as extend in range from micro to nanoscale. It is demonstrated that security keys extracted from the response of the resulting features exhibit close to ideal values when tested for quantitative metrics (randomness, uniqueness, uniformity, reliability) that are widely used for assessing PUF performance. Electrosprayed PUFs, which stand out with their performance and characteristics, can be taken further and cooperated with fluorescent molecules directly during the process. The resulting multicolor PUFs can be authenticated at different levels, which can be performed independently. Here, the random distribution of features forms the first level of security, which can be verified by conventional optical microscopy imaging, and unique morphology of features forms the second security layer, which can be authenticated by advanced techniques (SEM, AFM, etc.), and lastly, the multiple responses from unique fluorescence from randomly located photoluminescent molecules form a third layer of security that can be authenticated with fluorescence imaging. Finallly, the high coding capacity was determined and the impenetrable layers of security defined procedurally, and the multicolor PUFs are applied directly as security label on the goods. The images taken from the application/good to be authenticated are verified by comparing the keys obtained with the binaryization and subsequent 256-bit code extraction procedures with those in the database. In addition to this common authentication method, much more practical and faster authentication mechanisms have been developed within the scope of our study, where many limitations are eliminated thanks to the direct implantation of image feature matching algorithms into PUF applications.