We have developed a holographic data storage system that can demonstrate real-time playback of beyond high definition video signals. In the proposed system, to increase the data-transfer rate of the reproduced data, we focused on improving the SNR of the reproduced data and on improving the signal processing speed, which the SNR of the reproduced data has a significant effect on. One of the factors that deteriorate the SNR is shrinkage in the medium. This shrinkage distorts recorded holograms and degrades the quality of the reproduced data. We investigated wavefront compensation as a means to improve the SNR of reproduced data degraded by hologram distortion and found that controlling the defocus component of the reference beam is effective. We have also been developing parallel signal processing to increase the data-transfer rate. We placed three GPUs in the signal processing unit: one for the reproduced data detection from the reconstructed image and two for the LDPC decoding for error correction of the reproduced data. The LDPC decoding required a lot more time than the data detection, so we designed a signal processing in which detected data in the GPU for the data detection were sent to the two GPUs for the LDPC decoding alternatively. We implemented wavefront compensation for the defocus component and developed parallel signal processing with three GPUs for our holographic data storage system. Using this system, we demonstrated real-time playback of beyond high definition video signals with 50 Mbps.

Date Published: 5 September 2014
PDF: 7 pages
Proc. SPIE 9201, Optical Data Storage 2014, 92010G (5 September 2014); doi: 10.1117/12.2063423

Published in SPIE Proceedings Vol. 9201:
Optical Data Storage 2014
Ryuichi Katayama; Thomas D. Milster, Editor(s)