ABSTRACT
Artificial Intelligence (AI) inspired by the human brain is believed to dominate our future world. The basic topologies of AI stand on artificial neural networks (ANNs), the simpler versions of biological neural networks (BNNs). In order to match the functioning of these ANNs to BNNs, the key features of BNNs need to be retained. Out of these, non-volatility and connections with memory are the key features. To obtain these features, researchers started to learn about changing the resistance of a material by changing its physical state (amorphous ↔ crystalline, oxide ↔ reduced oxide etc.). This phenomenon leads to phase-change devices. In this chapter, the construction and operation of the phase-change devices such as phase-change memory (PCM), memristor, and resistive random access memory (RRAM) is presented. The parameters of importance for these phase-change devices are also discussed in detail. This is followed by the introduction of implementation techniques of these phase-change devices. Besides, a brief overview of the challenges faced for the implementation of these devices is provided. In addition, the applications of these phase-change devices in logic and memory design are discussed.
