Analysis and Improvements of the Pseudorandom Number Generation in Passive UHF-RFID Tags

Several works have been carried out in recent years aiming at improving the performance of RFID systems. Some proposals focus their attention on the communication protocol between the reader and passive tags, trying to avoid collisions or the capture effect during the identification process. Most of these works have a common factor: the required use of a pseudorandom number generator that has a great impact in the performance of the identification process. This paper analyzes the quality of the pseudorandom number generator in commercial passive UHF-RFID tags, to assess whether its outputs can be distinguished from truly random data. It then goes on to present a new mechanism that simplifies the generation of these numbers. A set of tests has been used to assess the randomness of the measured pseudorandom number sequences from commercial tags as well as the pseudorandom numbers obtained with the proposed method. This new method has then been implemented in a physical platform that emulates a real tag and its adequacy checked with commercial readers. The results thus obtained indicate that the proposed method generates valid pseudorandom number sequences while reducing the complexity and power consumption required for the tag.