Storing analog data with an analog memory system
Abstract/Contents
- Abstract
- Data stored in the cloud or on mobile devices reside in physical memory systems with finite sizes. Today, huge amounts of analog data, e.g. images and videos, are first digitized and then algorithmically compressed (e.g., using the JPEG standard) to minimize the amount of physical storage required. However, the conventional approach to analog data storage is inefficient in accommodating the continuing exponential increase of data (most of the data are in analog form) in the era of artificial intelligence. Emerging non-volatile-memory technologies (e.g., phase change memory (PCM), resistive random access memory (RRAM)) provide the possibility to store information in an analog fashion to enhance memory use efficiency. This dissertation introduces a new concept of directly storing analog data in a compressed format into analog-valued memory. Taking a multi-disciplinary approach combining joint source channel coding (JSCC) from information theory, neural network, along with the use of emerging non-volatile memories that can directly store analog values as a set of analog electrical properties (e.g., resistances), we develop and demonstrate with experiments a novel image compression and storage scheme using analog-valued non-volatile memories (NVM), with PCM and RRAM arrays as examples. This scheme simultaneously delivers competitive performance for storing analog images compared with conventional image compression and storage with digital memory, resilience to the PCM and RRAM device technology non-idealities (e.g., defective cells, device variability, resistance drift, and relaxation), and adaptivity to different NVM characteristics. This work opens up new opportunities for addressing pressing demands for the storage of analog data.
Description
Type of resource | text |
---|---|
Form | electronic resource; remote; computer; online resource |
Extent | 1 online resource. |
Place | California |
Place | [Stanford, California] |
Publisher | [Stanford University] |
Copyright date | 2022; ©2022 |
Publication date | 2022; 2022 |
Issuance | monographic |
Language | English |
Creators/Contributors
Author | Zheng, Xin |
---|---|
Degree supervisor | Wong, Hon-Sum Philip, 1959- |
Thesis advisor | Wong, Hon-Sum Philip, 1959- |
Thesis advisor | Pop, Eric |
Thesis advisor | Wong, S. Simon |
Degree committee member | Pop, Eric |
Degree committee member | Wong, S. Simon |
Associated with | Stanford University, Department of Electrical Engineering |
Subjects
Genre | Theses |
---|---|
Genre | Text |
Bibliographic information
Statement of responsibility | Xin Zheng. |
---|---|
Note | Submitted to the Department of Electrical Engineering. |
Thesis | Thesis Ph.D. Stanford University 2022. |
Location | https://purl.stanford.edu/bb403bh6352 |
Access conditions
- Copyright
- © 2022 by Xin Zheng
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...