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

Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE

Impulse-equivalent time-domain optical memory

Xiao A. Shen, Ravinder Kachru

Published Sep 20, 1996• 0 citations
Publisher

Abstract

We propose a novel approach to time-domain storage of serial optical data. In this approach, two identical impulse- equivalent sequences are used as write and read pulses to store and later retrieve optical information. These pulse sequences are both amplitude- and phase-modulated in such a way that their autocorrelations resemble, as closely as theoretically possible, the autocorrelation of a single large brief pulse to ensure faithful data retrieval. Such impulse-equivalent sequences are known to exist for all lengths, permitting the storage of large bandwidth data. In a proof-of-concept experiment, we have successfully stored a 42-microsecond long stream containing 420 bits of data in a 40-MHz spectral channel in a Eu3+:Y2SiO5 crystal. The stored data was repeatedly recalled for up to eight times with high signal fidelity, showing no measurable sidelobe-induced noise.

DOI
10.1117/12.251971

Related papers

Impulse-equivalent time-domain optical memory

Xiao A. Shen, R. Hartman, Ravinder Kachru

We propose a novel approach to time-domain storage of serial optical data. In this approach, two identical impulse-equivalent sequences are used as write and read pulses to store and later retrieve optical information. These pulse sequences are both amplitude and phase modulated in such a way that their autocorrelations resemble, as closely as theoretically possible, the autocorrelation of a single large brief pulse to ensure faithful data retrieval. Such impulse-equivalent sequences are known to exist for all lengths, permitting the storage of large-bandwidth data. The advantages of this approach include modest laser power requirement, high signal fidelity, and encrypted recording.

Optics Letters•Jun 1, 1996•7 citations

Experimental demonstration of impulse-equivalent time-domain optical memory

Xiao A. Shen, Ravinder Kachru

We demonstrate the impulse-equivalent approach [Opt. Lett. 21, 833 (1996)] to time-domain storage of optical data. Using the 128-bit real Huffman-Ackroyd code as the write-read sequences, we have successfully stored a 42-micros-long stream containing 420 bits of data in a 40-MHz spectral channel in a Eu(3+):Y(2)SiO(5) crystal. The stored data were repeatedly recalled as many as eight times with high signal f idelity, showing no measurable sidelobe-induced noise. Data retrieval with the use of mismatched write-read pulses was also examined, and implications of the results for data encryption are discussed.

Optics Letters•Dec 15, 1996•4 citations

Use of biphase-coded pulses for wideband data storage in time-domain optical memories

Xiao A. Shen, Ravinder Kachru

We demonstrate that temporally long laser pulses with appropriate phase modulation can replace either temporally brief or frequency-chirped pulses in a time-domain optical memory to store and retrieve information. A 1.65-µs-long write pulse was biphase modulated according to the 13-bit Barker code for storing multiple bits of optical data into a Pr(3+):YAlO(3) crystal, and the stored information was later recalled faithfully by using a read pulse that was identical to the write pulse. Our results further show that the stored data cannot be retrieved faithfully if mismatched write and read pulses are used. This finding opens up the possibility of designing encrypted optical memories for secure data storage.

Applied Optics•Jun 10, 1993•3 citations

Time- and frequency-domain hybrid optical memory: 16-kbit data storage in Eu^3+:Y_2SiO_5

Masaharu Mitsunaga, Ryuzi Yano, Naoshi Uesugi

We propose and demonstrate a novel type of frequency-selective optical memory that writes and reads the data in both the time domain and the frequency domain. Temporal 16-bit data were stored by accumulated-photon-echo bit-by-bit storage at 103 frequency addresses within the inhomogeneous line of the (7)F(0)-(5)D(0) transition of Eu(3+):Y(2)SiO(5), which yields a total memory capacity of 1.6 kbits in a single spot of 240-microm diameter. The keys to the success of this experiment are this material's long dephasing time and lack of spectral diffusion.

Optics Letters•Dec 1, 1991•121 citations

Time-domain optical data storage using Eu3+ ions in crystals

Masaharu Mitsunaga

The photon-echo memory[1], or the time-domain optical data storage, resembles the hole-burning memory [2] in that the inhomogeneous absorption line of a material is employed as a storage medium and the information is stored in it as a frequency-dependent absorption change with the resolution determined by the homogeneous width. Having the common memory capacity N given by the inhomogeneous-to-homogeneous width ratio, both types of memories share common problems like low-temperature operation, spectral diffusion and so on. Although a storage medium is common, storage techniques are definitely different for the two memories. The photon-echo memory utilizes the temporal interference effect between a reference pulse and a data pulse and their interference pattern is stored as a population grating in the frequency domain. In this sense this is a time-domain analog of holography where the spatial interference pattern between a reference beam and a data beam is stored in a medium as a population grating in the real space. Like the spatially diffracted beam from the read beam gives the information stored in the hologram, the “temporally diffracted pulse", or the photon echo, after the read pulse can give the information stored in the inhomogeneous distribution.

Spectral Hole-Burning and Related Spectroscopies: Science and Applications•Jan 1, 1991•1 citation

Coherent saturation removal in time-domain optical memory by storage of frequency-chirped data pulses

Xiao A. Shen, Ravinder Kachru

We report a novel technique for storing digital optical data in coherent time-domain optical memory to circumvent the problem of coherent saturation. In this technique, the carrier frequency of a data stream to be stored is slowly chirped so its power is spread over the data bandwidth for maximum data storage. In a proof-of-concept experiment, a total of 500 bits of data were stored at a rate of 20 Mbits/s in a 45-MHz-wide channel within the inhomogeneous line of the (7)F(0)-(5)D(l) transition in Eu(3+):Y(2)SiO(5). The result suggests a minimum storage density of ~1.5 Gbits/cm(3) for this transition. Issues related to the storage of frequency-chirped data, such as the optimum chirp width and the saturation owing to peak-sidelobe interference, are also discussed.

Optics Letters•Nov 15, 1993•11 citations
We propose a novel approach to time-domain storage of serial optical data. In this approach, two identical impulse- equivalent sequences are used as write and read pulses to store and later retrieve optical information. These pulse sequences are both amplitude- and phase-modulated in such a way that their autocorrelations resemble, as closely as theoretically possible, the autocorrelation of a single large brief pulse to ensure faithful data retrieval. Such impulse-equivalent sequences are known to exist for all lengths, permitting the storage of large bandwidth data. In a proof-of-concept experiment, we have successfully stored a 42-microsecond long stream containing 420 bits of data in a 40-MHz spectral channel in a Eu3+:Y2SiO5 crystal. The stored data was repeatedly recalled for up to eight times with high signal fidelity, showing no measurable sidelobe-induced noise.
We propose a novel approach to time-domain storage of serial optical data. In this approach, two identical impulse-equivalent sequences are used as write and read pulses to store and later retrieve optical information. These pulse sequences are both amplitude and phase modulated in such a way that their autocorrelations resemble, as closely as theoretically possible, the autocorrelation of a single large brief pulse to ensure faithful data retrieval. Such impulse-equivalent sequences are known to exist for all lengths, permitting the storage of large-bandwidth data. The advantages of this approach include modest laser power requirement, high signal fidelity, and encrypted recording.
We demonstrate the impulse-equivalent approach [Opt. Lett. 21, 833 (1996)] to time-domain storage of optical data. Using the 128-bit real Huffman-Ackroyd code as the write-read sequences, we have successfully stored a 42-micros-long stream containing 420 bits of data in a 40-MHz spectral channel in a Eu(3+):Y(2)SiO(5) crystal. The stored data were repeatedly recalled as many as eight times with high signal f idelity, showing no measurable sidelobe-induced noise. Data retrieval with the use of mismatched write-read pulses was also examined, and implications of the results for data encryption are discussed.
We demonstrate that temporally long laser pulses with appropriate phase modulation can replace either temporally brief or frequency-chirped pulses in a time-domain optical memory to store and retrieve information. A 1.65-µs-long write pulse was biphase modulated according to the 13-bit Barker code for storing multiple bits of optical data into a Pr(3+):YAlO(3) crystal, and the stored information was later recalled faithfully by using a read pulse that was identical to the write pulse. Our results further show that the stored data cannot be retrieved faithfully if mismatched write and read pulses are used. This finding opens up the possibility of designing encrypted optical memories for secure data storage.
We propose and demonstrate a novel type of frequency-selective optical memory that writes and reads the data in both the time domain and the frequency domain. Temporal 16-bit data were stored by accumulated-photon-echo bit-by-bit storage at 103 frequency addresses within the inhomogeneous line of the (7)F(0)-(5)D(0) transition of Eu(3+):Y(2)SiO(5), which yields a total memory capacity of 1.6 kbits in a single spot of 240-microm diameter. The keys to the success of this experiment are this material's long dephasing time and lack of spectral diffusion.
The photon-echo memory[1], or the time-domain optical data storage, resembles the hole-burning memory [2] in that the inhomogeneous absorption line of a material is employed as a storage medium and the information is stored in it as a frequency-dependent absorption change with the resolution determined by the homogeneous width. Having the common memory capacity N given by the inhomogeneous-to-homogeneous width ratio, both types of memories share common problems like low-temperature operation, spectral diffusion and so on. Although a storage medium is common, storage techniques are definitely different for the two memories. The photon-echo memory utilizes the temporal interference effect between a reference pulse and a data pulse and their interference pattern is stored as a population grating in the frequency domain. In this sense this is a time-domain analog of holography where the spatial interference pattern between a reference beam and a data beam is stored in a medium as a population grating in the real space. Like the spatially diffracted beam from the read beam gives the information stored in the hologram, the “temporally diffracted pulse", or the photon echo, after the read pulse can give the information stored in the inhomogeneous distribution.
We report a novel technique for storing digital optical data in coherent time-domain optical memory to circumvent the problem of coherent saturation. In this technique, the carrier frequency of a data stream to be stored is slowly chirped so its power is spread over the data bandwidth for maximum data storage. In a proof-of-concept experiment, a total of 500 bits of data were stored at a rate of 20 Mbits/s in a 45-MHz-wide channel within the inhomogeneous line of the (7)F(0)-(5)D(l) transition in Eu(3+):Y(2)SiO(5). The result suggests a minimum storage density of ~1.5 Gbits/cm(3) for this transition. Issues related to the storage of frequency-chirped data, such as the optimum chirp width and the saturation owing to peak-sidelobe interference, are also discussed.