Markov Chain Signal Generation Based on Single Magnetic Tunnel Junction
Xihui Yuan, Jiajia Jian, Zheng Chai, Suihuan An, Yawei Gao, Xue Zhou, J. F. Zhang, Weidong Zhang, Tai Min
Abstract
Markov chain (MC) is a stochastic model that describes a sequence of events where the probability of each event depends only on the previous state. Such memoryless property makes MC widely used in machine learning and encryption, but the hardware implementation of MC generation remains challenging. This paper presents a hardware solution for generating MC signals using only one industry-ready magnetic tunnel junction (MTJ). High quality standard MC signal has been generated with low error and good randomness. The proposed solution also demonstrates the potential in increasing the generation speed. The presented solution offers a hardware-friendly implementation of MC signal in semiconductor chips.