The present invention relates to high-speed high-density optical disk data storage, and in particular to a system and method for storing multiple-bits of data at a single diffraction limited pit.
High-speed high-density data storage technologies are required in next generation information superhighway and military warfare because of increased demand in information access. The data storage technology is critical for fast computing and processing for targeting, time critical communication and control, and real-time sensor-to-shooter operations. The data storage is also critical for fast access to large intelligent database for other command and control operations. Furthermore, high-density data storage is highly demanded in computer and network applications. All these applications place high requirement on data storage systems namely high storage capacity, fast data access, and rugged system packaging.
There are significant advancements in disk data storage technology in recent years. The traditional magnetic disk storage has now reached its performance limitations. Higher storage density becomes harder and harder to achieve on these magnetic disks and close head-to-media spacing makes high density drives no longer removable. To enhance the stored data integrity and achieve high storage density with large head-to-media spacing, optical disk storage is preferred.
Optical data storage is currently a hot research topic. There are constant developments to improve data storage density such as holographic data storage, optical disk data storage (evolution of DVD), and nano-structure technique such as near field micro-hole diffraction. Holographic data storage can demonstrate extremely high storage density using LiNbO3, BaTiO3, photopolymers, liquid crystals, Ge-doped silica glass, protein recording media, etc. Despite the high storage density achieved, the holographic storage requires critical vibration-free recording and readout setups. Many of them suffer from destructive readout problems. Volume holographic storage is presently far from commercial uses.
Optical disk data storage would be considered as preferred commercial technique since the disk architecture is well accepted, as long as the storage density bottleneck problem is solved. The data storage media under research and commercial applications include magneto-optic media, dye-polymer, phase change media, etc. The disk storage density bottleneck problem is mainly the storage coding technique. Of course, it is also related to available recording material. The conventional coding is binary coding achieved by recognizing different reflected light levels (high or low in binary format) from the storage disk (magneto-optic disk, dye-polymer disk, or phase change disk). The recording density is thus limited by the focused optical spot size used for recording and readout. The minimum focused spot size is the diffraction limited spot size from the focusing lens unit at the designated optical wavelength.
In addition, some novel technologies are being pursued in parallel towards accomplishing higher capacities per disk and higher data transfer rates. Several unconventional long-term optical data storage techniques promise data densities greater than 100 Gb/in2 and perhaps even exceeding Tb/in2. These include near-field optics, sub-Rayleigh criterion optics and probe storage techniques that hold promise for new optical data storage technologies. All these techniques can be considered as near field techniques since the storage pit size is further reduced that requires near field recording and readout. Employing parallel readout to an array of such optics is way to increase read/write access time and data throughput. Thus, it can be readily adapted into this technology. Although there have demonstrated some proof-of-principle results, a number of issues must be addressed before these technologies can be considered for commercial applications, such as overall system reliability, bit stability, tip/medium wear, limits of data rate, signal to noise ratio, and cost. These near field techniques even if successful would require high-resolution servo scanning that is very costly. Furthermore, the high precision scan readout is slow limited by the trade-off between scanning speed and scanning accuracy. Such technique cannot well use high-speed electronics to increase data access rate since the limitation comes from the readout scanning. Thus, the data access rate would be a major concern.
Reducing storage pit size is currently a general technique for improving data storage density (except the in-mature holographic storage technique). However, the data coding is still binary code. Such trend of reducing pit size increases recording and readout difficulty that makes newer technologies more difficult to be commercially implemented.
A high-density disk storage concept using micro holographic multiplexing method has been reported [e.g., H.J. Eichler, P. Kuemmel, S. Orlic, A. Wappelt, “High-density disk storage by multiplexed microholograms,” IEEE J. of Selected Topics in Quant. Electr. 4, 840-849 (1998)]. That method combines the bit-oriented storage of conventional optical disks and volume storage from the holographic approach, and thus it benefits from both technologies. This concept has been proved experimentally through generation of micro holograms with about 2 micron diameter, wavelength multiplexing by three-color hologram recording and picoseconds recording in a commercial photopolymer.
Although with successful demonstration of the micro holographic multiplexing method for high density data storage, there are some limitations that may obstruct this concept from practical commercial systems. First, the hologram recording needs highly coherent light sources (laser lines). For multiple bits storage on a single spot, the laser line number must be the same as the bit number. It is difficult to find multiple laser lines with nearly equal wavelength intervals in the sensitivity range of a recording material. The cost of using these lasers is high. Another issue is that the recording spot size cannot remain the same diffraction limited size along the propagation path of the beam in the recording medium. It requires a certain thickness of a medium for multiplexing the multiple holograms. This thickness will be much larger than the Rayleigh length of a laser waist. The divergent beam at the recording volume will cause cross talks among adjacent pits. This effect limits the achievable recording density.
It is therefore an object of the present invention to provide an optical disk data storage system that has high storage density and fast access rate by coding multiple bits on a single pit that may be a diffraction limited pit.
It is another object of the present invention to provide a novel optical disk data storage system that is compatible with existing optical disk drivers such as CD-ROM and DVD for data recording and readout.
It is another object of the present invention to provide an optical disk data storage system that can use various light sources including both coherent and non-coherent sources for recording and readout so as to reduce the system cost.
It is yet another object of the present invention to apply a specially designed objective lens system with property of extended depth of focus to create diffraction limited spot size with long focal depth of the recording beam for minimizing crosstalk between stored pits in a thick storage film.
According to the present invention, there is described a novel coding and implementation techniques for high-speed high-density optical disk data storage. Multiple narrowband spectral beams, either coherent or non-coherent, are combined together by optical fiber couplers or lenses assembly and are then focused into a photosensitive film with diffraction limited spot size through a hybrid diffractive/refractive lens. The focal depth of the focused beam is extended due to the use of the specially designed hybrid diffractive/refractive lens, so that the beam size remains diffraction limited size in the whole depth of the recording medium volume. Multiple gratings which are respectively corresponding to these used spectral bands are recorded in the medium since a reflection mirror which is attached at the back surface of the recording film reflects back the incident beam so that the incident and the reflected beams interfere each other to generate the gratings. The reflected beam from the reflection mirror can also be replaced by a second focused beam (without using reflection mirror) with the recording beam splitted using an optical fiber splitter from the incident recording beam and using an identical lens with the same extended depth of focus property. By using white light to readout these gratings and using a spectrometer or multi-wavelength reader to analyze the reflected light, some peaks corresponding to these recording wavebands can be found. By proper spectral curve operations, the recording information is recovered from this reflected spectrum. This present invention can write and read multiple bits information on a single storage pit, which is usually only one bit based on conventional storage technologies.
a shows the focal depth property of the hybrid diffractive/refractive lens for extending the depth of focus.
b shows the focal depth property of a conventional objective lens.
a and 2b show a comparison of focal depth property between an extended focal depth lens 16 and a conventional lens 23. To achieve long propagation depth of at least two times the film thickness (forward and return) of the diffraction limited recording beam spot, the present invention uses pseudo non-diffracting beam 21 instead of conventional Gaussian beam 22. Conventional Gaussian beam 22 has very limited propagation depth 24 at diffraction limited spot size. Pseudo non-diffracting beam 21 on the other hand can maintain its small central spot for a substantially long propagation distance 25 with minor beam spreading in transversal direction. The pseudo non-diffracting beam can be achieved by using a diffractive/refractive hybrid lens 16 which is fabricated by using a specially designed diffractive optical element 16a attached on a conventional objective lens 16b as shown in
Various existing optimization algorithms for diffractive optical element design can be adopted to design the pseudo non-diffracting beam shaper 16a.
Attaching the designed non-diffracting beam shaper 16a to a refractive lens 16b will generate the appropriate power hybrid refractive-diffractive lens 16 with extended focal depth. Due to opposite chromatic dispersion between diffractive element and refractive element, it is possible to obtain achromatic lens by combining the two types elements. When material specifications and geometric parameters of the adopted refractive lens are properly designed, perfect achromatic property can be achieved so that the hybrid lens is suitable to work in a wide waveband.
One embodiment schematic of multi-wavelength recording system is shown in
There is still another method shown in
Upon exposing to the multiple wavelengths light, gratings are generated in the photosensitive film 11. When illuminated with a white light source (54 in
To retrieve the recorded data information from the reflection spectral curve, data processing on the spectral curve can be carried out. If the spectrum is recorded by binary signal modulation, simply thresholding the spectral data at corresponding wavelength locations can recover the recorded information represented by a serial of binary bits.
If the recording signal is modulated in gray levels, the recorded information can be recovered by comparing detected spectral information, comparing spectrum curve curvatures, local spectrum curve slopes, and relative intensities in each wavelength region, correct spectral-coded information can be recognized independent of some white-light intensity fluctuation and slight readout spot misalignment. Here the readout is not based on determining the absolute transmittance at each wavelength since that is sensitive to readout alignment error and light source intensity fluctuation. Comparing spectrum curve curvature, slope, and relative intensity is a novel technique that offers reliable readout. The readout color (or spectral) is unique if we use conventional chromaticity evaluation technique. Although there are a lot of spectral details, the readout needs to employ grouping technique to identify unique features in each wavelength and use table lookup to minimize the recognition processing.
The spectral-coding concept uses different detailed spectroscopic contents on a recording media to represent different data. Different detected intensity patterns will be properly compared by an electronic processing circuit and to establish its corresponding binary value. Using photodetector array is similar to using many readout heads in parallel. This approach on spectral identification can be quite accurate by comparing many spectral data. The speed of spectral identification can be fast using advanced high-speed electronics including multi-chip modules and using table look-up technique. The proposed spectral coded storage allows further improvement of data access rate through electronic circuit improvement. Since each recording pit stores multiple bits of information, we can expect minor difference on spectral curve to represent different data. Spectral curve fitting and colorimetry evaluation of local curve curvature, slope, and comparison of relative intensities can quantify the spectral details with excellent accuracy.
The present invention of spectral coded data read/write system as shown in
White light to fiber coupling is another technical challenging issue. We use white light for readout because we can extract more spectral details at wavelengths besides the recording laser wavelengths. This can produce a continuous spectral curve to represent the multiple bits binary data. Because the relatively high lamp power, even with the poorer fiber coupling efficiency, we can still deliver enough white light to the read/write head for spectral readout.
With a single-mode fiber as input and output to the read/write head, we can modify the existing CD-ROM or DVD head for the spectral coded data readout.
We use the optical fiber to replace the original red laser diode as shown in
If a multi-wavelength combiner shown in