Parallel Density‐Based Spatial Clustering with Dual‐Functional Memristive Crossbar Array

• Published in: Advanced functional materials Vol. 34; no. 8
• Main Authors: Cheong, Sunwoo, Shin, Dong Hoon, Lee, Soo Hyung, Jang, Yoon Ho, Park, Taegyun, Han, Janguk, Shim, Sung Keun, Kim, Yeong Rok, Han, Joon‐Kyu, Ghenzi, Néstor, Hwang, Cheol Seong
• Format: Journal Article
• Language: English
• Published: Hoboken Wiley Subscription Services, Inc 01.02.2024
• Subjects: Algorithms; analog Euclidean distance calculation; Arrays; Clustering; Data points; Datasets; DBSCAN; Density; dual‐mode memristors; Euclidean geometry; Gene sequencing; memristive crossbar arrays; Memristors; processing‐in‐memory; propagation clustering; Synthetic data
• ISSN: 1616-301X; 1616-3028
• DOI: 10.1002/adfm.202309108

Analog and digital switching performances in a Ta/HfO2/RuO2 (THR) memristor are studied to implement a density‐based spatial clustering of applications with noise (DBSCAN) algorithm in a low‐power, parallel‐computing memristor crossbar structure. In the analog mode THR memristor, more than 256 states can be stored through a fine‐tuning process with a denoising scheme. The analog mode crossbar array facilitates Euclidean distance calculation between any points in the given graphic dataset. In the digital mode, the on/off ratio of more than three orders of magnitude between the binary states is achieved, providing functionality to cluster the data points with a reduced number of operations. The parallel computing capacity of the adopted crossbar decreases the time complexity of the original DBSCAN from O(n2) to O(n). Through array‐level simulations, the effectiveness of hardware functionality is validated using representative synthetic datasets and single‐cell RNA sequences datasets. Analog and digital switching performances in Ta/HfO2/RuO2 memristor are studied to implement a density‐based spatial clustering of applications with noise algorithm in a low‐power, parallel‐computing memristor crossbar structure. The parallel computing capacity of the adopted crossbar decreases the time complexity of the original DBSCAN from O(n2) to O(n).