| Popis: |
NOTE: Text or symbols not renderable in plain ASCII are indicated by [...]. Abstract is included in .pdf document. Volume holographic data storage involves the superposition and independent recall of multiple pages of data within the same volume of a storage medium. These pages, stored as separate holograms, can be accessed by changing the angle of the reference laser beam used to store and retrieve them. Because all the data in a stored page is read out in parallel, the output data rate can be very large. At the same time, large storage capacity is available through the superposition of many data pages. The topic of this thesis is volume holographic memories using the 90° geometry. This configuration, where signal and reference beams enter orthogonal crystal faces, is attractive for angle multiplexing because of its high angular selectivity. We choose angle multiplexing because it gives us many options for rapid steering of the reference beams. Our goal is to develop read—write holographic memories which achieve high capacity and high output data rate. Our approach, in terms of recording material, is to work with what we have. In our case, the only photorefractive widely available in thicknesses greater than a centimeter is Fe-doped [...]. This material is relatively easy to make with high optical quality, and its performance shows no degradation after repeated record/thermal erase cycles. The disadvantages of [...]Fe include volatility of storage, which we will treat briefly, and poor dynamic range, which we will discuss extensively in the first part of the thesis. We start in Chapter 2 with a study of dynamic range in holographic storage, in order to determine what is required of a photorefractive crystal. One of the outcomes of this study is a concise metric—which we call the M/#—for measuring the dynamic range performance of a holographic storage system. Chapter 3 discusses the experimental measurement of this M/# as a function of the oxidation state of [...]. We find that there exists an optimal oxidation state (for maximum dynamic range performance), and in Chapter 4 we develop a theoretical model which predicts this optimum. In the remainder of Chapter 4, we extend this model to other parameters such as crystal size, doping, and modulation depth. Having squeezed as much performance as possible from our storage material, we turn to the design of a large–scale holographic memory. Our goal is to use angle, fractal, and spatial multiplexing to achieve large capacity—without sacrificing fast access to the stored holograms. In Chapter 5, we discuss our segmented mirror array, and how it makes such a design possible. Then in Chapter 6, we experimentally demonstrate the various features of this memory design. These demonstrations include storage using the mirror array, storage of 1000 holograms using an acousto–optic deflector, storage of 10,000 holograms in the same [...] volume of [...], and the demonstration of the 160,000 hologram system with the mirror array and mechanical scanners. In this last part of the thesis, we consider additional aspects of holographic storage, in preparation for proposing a bigger and better system. In Chapter 7, we discuss systems issues affecting holographic memory design. In this vein, we survey the methods of performing angle–multiplexing, and introduce and demonstrate a new device for angle steering: a silicon bulk–micromachined, magnetically–actuated micromirror. We also discuss time response and noise and error performance of holographic memories. Finally, in Chapter 8 we propose and discuss several next–generation designs for large–scale high–speed holographic memories. This includes a method for nonvolatile readout that combines several previously proposed methods. |
