The objective of this research is to develop signal processing system design solutions to enable high-performance and low-cost hybrid solid-state/magnetic hard disk drives with beyond 1 Tbit per square inch areal density. The proposed research aims to cohesively develop powerful signal processing algorithms and low-cost VLSI architecture solutions that can simultaneously address the low signal-to-media-noise ratio and high system fault tolerance implementation cost faced by the emerging hybrid shingled written recording systems.
The intellectual merit of this proposal lies in the theme of developing unified signal processing system design solutions from both theory and practical implementation perspectives. New non-controllable trellis-based finite-state detectors and their efficient VLSI architectures will be developed to address the loss of signal-to-media-noise ratio caused by shingled writing. A unified read channel design approach and a data-management-aware signal processing strategy will be further developed to reduce the overall system signal processing implementation cost.
The broader impact of the proposed research is that it will enable the development of low-cost ultra-high capacity mass data storage drives to accommodate the exploding global data volume. It also contributes greatly to sustaining the historical magnetic recording areal density growth rate, providing a direct benefit to this strategically important high technology sector critical to the economic health of the nation. Extra effort will be made to recruit women and other under-represented groups as research assistants in this project, involve representatives of the Pacific Islander minority as undergraduate researchers, and reach out to high school students as part of outreach efforts.
