Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Optical Computing

Sparse-Wavelength Angularly Multiplexed Volume Holographic Memory

Scott P. Campbell, Xianmin Yi, Pochi Yeh

Published Jan 1, 1995• 0 citations
Publisher

Abstract

Photorefractive-based optical memory systems have received considerable attention in recent years [1-6] In these systems, numerous volume holograms are stored within a single photorefractive crystal, with the writing and recall selectivity generally being performed via angle [1] wavelength [2,3], or phase [4] multiplexing. As well, non-photorefractive-based optical memory systems [7,8] have been proposed. In this paper, we present a hybrid system [6] that utilizes a combination of wavelength and angle multiplexing in a photorefractive medium. Such a hybridization is driven by photonic source generation and manipulation device limitations, as well as the desire to expand information throughput rates in optical memory systems via spectral parallelism [9]. Our design is essentially the summation over wavelength of many angularly multiplexed volume memory systems. We therefore compare our system’s parameters to those of purely wavelength or angle multiplexed ones.

DOI
10.1364/optcomp.1995.owa4

Related papers

High-Speed Storage of Wavelength-Multiplexed Volume Spectral Holograms

Xiao A. Shen, Yu Bai, Ravinder Kachru

One of the most attractive features of an optical memory is its ability to write and read data in a bit-parallel format, giving rise to theoretically very high (in excess of 1 Gbps) data transfer rates. However, such a potential has not been demonstrated experimentally because of various inherent technical difficulties associated with existing optical storage techniques. In a photorefractive or a persistent spectral hole-burning (PSHB) memory, for example, the time required to record one page varies from a fraction of a second to several seconds.1-3 For a page containing 1000 × 1000 bits of data, it translates to a bandwidth of approximately 1 Mbps, substantially slower than any existing semiconductor memories.

Optical Computing•Jan 1, 1995•0 citations

High-Speed Holographic Recording of 500 Images in Hole Burning Rare-Earth Doped Crystals

Ravinder Kachru, Xiao A. Shen

One of the unique and perhaps the most attractive features of an optical memory is its ability to store and retrieve information in a bit-parallel format.1-3 Recently, we have proposed a new scheme for parallel data storage in TDOM.4,5 This scheme was proven to be practical and can be implemented with existing technologies. In a proof-of-principle experiment, four wavelength-multiplexed single-page volume spectral holograms, generated with black-and-white transparencies, were successfully stored in a single spatial location at a speed of approximately 23 μs/frame. The experimental results project system's input/output bandwidth to exceed 40 Gbps, two orders of magnitude faster than a semiconductor cache memory.

Topical Meeting on Holography•Jan 1, 1996•0 citations

Dynamic photorefractive optical memory

Hajime SASAKI, Yeshaiahu Fainman, Joseph E. Ford +2 more

We investigate an architectural approach to dynamic three-dimensional volume storage that circulates holograms between two photorefractive crystals. Introduction of an optical amplifier into the system increases the effective write-erase asymmetry of the crystals and permits the amplification of the recalled images. This memory architecture is experimentally shown to provide write-erase storage with robustness to multiple optical readouts.

Optics Letters•Dec 1, 1991•45 citations

Encrypted optical storage with wavelength-key and random phase codes

Osamu Matoba, Bahram Javidi

An encrypted optical memory system that uses a wavelength code as well as input and Fourier-plane random phase codes is proposed. Original data are illuminated by a coherent light source with a specified wavelength and are then encrypted with two random phase codes before being stored holographically in a photorefractive material. Successful decryption requires the use of a readout beam with the same wavelength as that used in the recording, in addition to the correct phase key in the Fourier plane. The wavelength selectivity of the proposed system is evaluated numerically. We show that the number of available wavelength keys depends on the correlation length of the phase key in the Fourier plane. Preliminary experiments of encryption and decryption of optical memory in a LiNbO(3):Fe photorefractive crystal are demonstrated.

Applied Optics•Nov 10, 1999•91 citations

Optical Data Storage Using Persistent Spectral Hole Burning

Eric S. Maniloff, Stefan B. Altner, Stefan Bernet +2 more

The use of holography is attractive for optical data storage due to its poten­tial for implementing page oriented data storage systems. However, multiplexed recording using Bragg selective materials has not provided a large enough stor­age capability to justify the expense of these systems. Persistent spectral hole burning holography allows multiplexed recording using much smaller wavelength separations between the holograms, and hence dramatically increases the storage capacity possible in holographic memory systems [1]. A comparison between the wavelength selectivity obtained for transmission gratings recorded using hole burning in a dye doped polymer film versus Bragg selectivity in a 1 cm thick iron-doped lithium niobate crystal is shown in Fig. (1).

Joint International Symposium on Optical Memory and Optical Data Storage•Jan 1, 1993•0 citations

Status of 3D memory storage materials

Gajendra D. Savant, Tomasz P. Jannson, Winston Ho

The speed of modern supercomputers is primarily limited by the relatively slow speed of memory. Research over the past years has focused on developing optical memory systems that, by nature of optics, promise to achieve high speed and high memory capacity. Three- dimensional optical memory systems are well known for their storage capacity and achievable short access time. But the early attempts to exploit 3-dimensional media were limited to photorefractive crystals such as lithium niobate, due to the unavailability of any other suitable medium (alternative media are precluded by their diffraction limited storage densities and non- erasable characteristics). Although volume phase holograms can be recorded in photorefractive crystals, these materials have many obvious flaws such as strong scattering, phase distortion, self diffraction, fast decay, and low refractive index modulation. There have also been continuous efforts to develop new 3-D media (such as photon echo, photorefractive polymers, bacteriorhodusphin, memory access cache, etc., but none can be applied in real world 3-D device because these materials are still in the early stage of development.

Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE•Apr 27, 1993•0 citations
Photorefractive-based optical memory systems have received considerable attention in recent years [1-6] In these systems, numerous volume holograms are stored within a single photorefractive crystal, with the writing and recall selectivity generally being performed via angle [1] wavelength [2,3], or phase [4] multiplexing. As well, non-photorefractive-based optical memory systems [7,8] have been proposed. In this paper, we present a hybrid system [6] that utilizes a combination of wavelength and angle multiplexing in a photorefractive medium. Such a hybridization is driven by photonic source generation and manipulation device limitations, as well as the desire to expand information throughput rates in optical memory systems via spectral parallelism [9]. Our design is essentially the summation over wavelength of many angularly multiplexed volume memory systems. We therefore compare our system’s parameters to those of purely wavelength or angle multiplexed ones.
One of the most attractive features of an optical memory is its ability to write and read data in a bit-parallel format, giving rise to theoretically very high (in excess of 1 Gbps) data transfer rates. However, such a potential has not been demonstrated experimentally because of various inherent technical difficulties associated with existing optical storage techniques. In a photorefractive or a persistent spectral hole-burning (PSHB) memory, for example, the time required to record one page varies from a fraction of a second to several seconds.1-3 For a page containing 1000 × 1000 bits of data, it translates to a bandwidth of approximately 1 Mbps, substantially slower than any existing semiconductor memories.
One of the unique and perhaps the most attractive features of an optical memory is its ability to store and retrieve information in a bit-parallel format.1-3 Recently, we have proposed a new scheme for parallel data storage in TDOM.4,5 This scheme was proven to be practical and can be implemented with existing technologies. In a proof-of-principle experiment, four wavelength-multiplexed single-page volume spectral holograms, generated with black-and-white transparencies, were successfully stored in a single spatial location at a speed of approximately 23 μs/frame. The experimental results project system's input/output bandwidth to exceed 40 Gbps, two orders of magnitude faster than a semiconductor cache memory.
An encrypted optical memory system that uses a wavelength code as well as input and Fourier-plane random phase codes is proposed. Original data are illuminated by a coherent light source with a specified wavelength and are then encrypted with two random phase codes before being stored holographically in a photorefractive material. Successful decryption requires the use of a readout beam with the same wavelength as that used in the recording, in addition to the correct phase key in the Fourier plane. The wavelength selectivity of the proposed system is evaluated numerically. We show that the number of available wavelength keys depends on the correlation length of the phase key in the Fourier plane. Preliminary experiments of encryption and decryption of optical memory in a LiNbO(3):Fe photorefractive crystal are demonstrated.
The use of holography is attractive for optical data storage due to its poten­tial for implementing page oriented data storage systems. However, multiplexed recording using Bragg selective materials has not provided a large enough stor­age capability to justify the expense of these systems. Persistent spectral hole burning holography allows multiplexed recording using much smaller wavelength separations between the holograms, and hence dramatically increases the storage capacity possible in holographic memory systems [1]. A comparison between the wavelength selectivity obtained for transmission gratings recorded using hole burning in a dye doped polymer film versus Bragg selectivity in a 1 cm thick iron-doped lithium niobate crystal is shown in Fig. (1).
The speed of modern supercomputers is primarily limited by the relatively slow speed of memory. Research over the past years has focused on developing optical memory systems that, by nature of optics, promise to achieve high speed and high memory capacity. Three- dimensional optical memory systems are well known for their storage capacity and achievable short access time. But the early attempts to exploit 3-dimensional media were limited to photorefractive crystals such as lithium niobate, due to the unavailability of any other suitable medium (alternative media are precluded by their diffraction limited storage densities and non- erasable characteristics). Although volume phase holograms can be recorded in photorefractive crystals, these materials have many obvious flaws such as strong scattering, phase distortion, self diffraction, fast decay, and low refractive index modulation. There have also been continuous efforts to develop new 3-D media (such as photon echo, photorefractive polymers, bacteriorhodusphin, memory access cache, etc., but none can be applied in real world 3-D device because these materials are still in the early stage of development.