Investigation of a cerium- iron- doped lithium niobate photorefractive crystal hologram and its applications to optical signal processing

In this thesis, a novel Ce: Fe: doped LiNbO{dollar}sb3{dollar} crystal (both bulk and fiber form) is investigated. The applications of this specially doped crystal to large capacity holographic storage, fiber specklegram sensor, multi-target optical pattern recognition, and content-addressable polychromatic neural networks are also included.; At the beginning of this thesis, the growth procedure and the material properties of this Ce: Fe: doped LiNbO{dollar}sb3{dollar} crystal is introduced. The key points of growing this crystal such as well-mixing, proper furnace temperature distribution and the annealing process are stressed. After that, to estimate the real storage capacity of a photorefractive volume hologram, an analysis of angular and wavelength selectivity, cross-talk noise and dynamic range requirement for angular-multiplexed and wavelength-multiplexed holograms is also given. It is found that the reflection-type wavelength-multiplexed hologram offers a higher selectivity, lower cross-talk noise, and possibly less dynamic range requirement. Another important part of this thesis is using photorefractive fiber as fiber specklegram recording medium, which provides a more compact, robustness, and less expensive fiber specklegram sensor system. Applications of this crystal hologram to optical signal processing have also been investigated. First, since the reflection-type correlator has a larger shift invariant tolerance, we choose this type of architecture to make a multiplexed correlator. One of the major differences between the system reported in this thesis and the previous works is that it uses a tunable visible-light diode laser as the correlator light source. Second, in the neural network application aspect, by taking the advantage of the wide spectrum response of this crystal, a content-addressable polychromatic neural network is constructed. The major advantages of this system are the large storage capacity, fewer interconnection links, and content-addressable capability.; Finally some concluding remarks and brief discussions about future study are also given at the end of the thesis.