Millimeter Wave Chipless RFID Authentication Based on Spatial Diversity and 2-D Classification Approach

Abstract

In this paper, a mm-Wave chipless RFID tag is developed to operate in the V-band for authentication applications. A novel approach based on tag backscattered E-field measurements at different orientation angles for unitary classification is proposed. The concept is based on the hardness to identically reproduce materials due to the inherent randomness in the fabrication process. These uncertainties are transcribed in very small variations that can be observed in the tag electromagnetic response. A set of 16 identical tags were fabricated, each tag shares same fabrication mask and manufacture process method. Spatial diversity using the tag backscattering pattern (at two different angles) adds independent characteristics for estimating authenticity of each tag. To better exploit the large amount of data collect with this approach, a Machine Learning (ML) sighting classification is used, which enhance the system performance. The probability of error (PE) achieved with the method is around 1%. This PE is four times lower than the one obtained with a similar approach implemented in the X-band.

Chipless RFID Tag Implementation and Machine Learning Workflow for Robust Identification

“…Each pair of adjacent rings introduces a stopband resonance in the radar cross section (RCS) spectrum, whose frequency varies with the dimensions and spacing between the rings. According to [15], the RCS of chipless tags must be greater than -40dB and the resonance peaks (notches) used to code the ID of the tags should have a depth greater than 10dB from the maximum level of the RCS [22], [23] in order to be correctly identified with methods not involving ML. An enhancement of the RCS can be obtained by arranging the CRR cells periodically in the substrate, creating a CRR array.…”

A Review of Chipless RFID Measurement Methods, Response Detection Approaches, and Decoding Techniques

“…In [311], for example, PCA was used to detect and characterize cracks in metal samples. Machine learning based techniques have also been considered, including in [338] and [339] where quantile regression is used, in [325] where linear discriminant analysis is used, [43] that sends tag response data to a cloud database where various pattern recognition algorithms are applied for supervised learning purposes, and in [223] the Support Vector Machine approach is found to have the best performance among four machine learning approaches tested, to give just a few examples [207,340,341].…”

“…Just as there are application-specific measurement approaches, there are also application-specific post-processing methods. For example, post processing methods have been developed for specific sensor architectures [157,350,351], for authentication purposes [189,207,325,345,352], and for decoding responses measured from moving tags [337,346,[353][354][355]. Another application that has seen many specifically developed post-processing methods is collision avoidance/tag localization in multi-tag environments [247,248,[356][357][358].…”

Advances in the Design Techniques and Applications of Chipless RFIDs

“…The spatial technique is significantly different from a time domain or frequency domain, a Chipless RFID tag, illuminating the entire tag. An example [140] is comprehended in Fig. 13, which describes the difference between a spatial Chipless RFID tag reading technique and a frequency-based Chipless RFID tag reading technique.…”