This dissertation explores the effects of two-photon absorption and color center induced absorption on all-optical switching devices. The amount of allowable two-photon absorption was quantified by the parameter T = 2βλ/n₂, where λ is the operating wavelength, β is the two-photon absorption coefficient, and n₂ is the nonlinear refractive index coefficient, the latter two being measured at λ. If the value of T exceeds unity, the operation of all-optical switching devices is in general degraded beyond usable regimes. This result was demonstrated by numerical experiments on systems of equations modelling a nonlinear directional coupler, a prototypical all-optical switching device. The value of T was measured in two fibers, one made of lead silicate glass, and one made of TiO₂-doped silica. We find the value of T to be greater than unity at a wavelength of 1.06 μm in both fibers. Significant color center formation was seen in the lead glass fiber. These color centers were created through two-photon absorption and destroyed through one-photon absorption. Color center induced absorption was seen to mimic two-photon absorption in certain regimes. The nonlinear optical response of semiconductor-doped glasses, an example of a one-photon resonant nonlinearity, was studied. A relaxation time which is dependent on the carrier density was found to be important when modelling the response of these glasses.

No description available.

Firstly, for a optical switch that suffers from the intrinsic crosstalk problem when two optical signals cross at the same switch element, active connections must be node-disjoint in the switching network to avoid crosstalk. Applying a sequence of decomposition and merge operations, called crosstalk-free conjugate transformation, on each switch element to tackle this problem, the network resulting from this transformation is called crosstalk-free conjugate network. By using the numbering schemes of networks, we prove that if the route assignments in the original network are link-disjoint, their corresponding ones in the conjugate network would be node-disjoint. Thus, traditional nonblocking switching networks can be transformed into crosstalk-free optical switches in a routine manner. / Furthermore, we show that crosstalk-free or wavelength interleaving multicast switches can also be obtained from existing nonblocking multicast switches via the same crosstalk-free or wavelength-interleaving conjugate transformation in a straightforward manner, respectively. / On the other hand, high speed and easy integrated arrayed waveguide gratings (AWGs) and tunable wavelength converters (TWCs) are promising optical devices to construct wavelength division multiplexing (WDM) optical switches. In order to practically realize a WDM switch, the core research issue is how to provide wavelength-interleaving connections throughout the entire switch network of these optical devices. In this way, the topological transformation consisting of decomposition and merge operations can convert nonblocking space-division networks into wavelength-interleaving conjugate networks. The principle of this transformation relies on the parallel characters of internal contentions of space switches and wavelength interferences of WDM switches. We show that any nonblocking route assignments in the original space-division network can be mapped into the conjugate network and become wavelength interleaving under this topological transformation. / The emerging all optical network (AON) calls for the realization of high speed (user data rates from 10 Mbps to 10 Gbps), high capacity (∼NTbps) transparent optical network. One of the major obstacles of AON is the deployment of high-speed optical switching systems that can switch optical signals at the rate of several terabits per second. In all optical switches constructed by different optical devices, the crosstalk or wavelength interference may happen when any two optical signals cross at a single optical device or when two signals carried by the same wavelength in a single link, respectively. In the past serval decades, the nonblocking electronic switching networks have been widely studied and a mature theory has been erected. In this thesis, we propose an easy-to-implement conjugate transformation applied to the traditional nonblocking switching networks in two opposite manners to tackle these two problems, respectively. / Deng Yun. / "January 2007." / Adviser: Tony Tong Lee. / Source: Dissertation Abstracts International, Volume: 68-09, Section: B, page: 6176. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (p. 101-108). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract in English and Chinese. / School code: 1307.

So Kin Tai. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2000. / Includes bibliographical references (leaves 65-66). / Abstracts in English and Chinese. / Chapter Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- The 0-1 sorter and concentrator --- p.1 / Chapter 1.2 --- Review of literature on constructions --- p.3 / Chapter 1.2.1 --- Odd-even merging network --- p.4 / Chapter 1.2.2 --- The Fast Knockout algorithm --- p.5 / Chapter 1.2.3 --- Reverse Banyan network prepended by running sum adder --- p.10 / Chapter 1.2.4 --- Recursive construction using iterative cells --- p.14 / Chapter 1.2.5 --- Comparison of construction algorithms --- p.17 / Chapter Chapter 2 --- Compressor based on baseline-swap network --- p.22 / Chapter 2.1 --- Bit permutation induced exchange --- p.22 / Chapter 2.2 --- Compressor --- p.26 / Chapter 2.3 --- The baseline-swap network --- p.28 / Chapter 2.4 --- New algorithm for running parity initialization --- p.31 / Chapter 2.5 --- Input fairness --- p.42 / Chapter Chapter 3 --- The general architecture of 0-1 sorter --- p.46 / Chapter 3.1 --- Recursive 2X-construction --- p.46 / Chapter 3.2 --- Control a 2X-interconnection network as a cyclic 0-1 sorter --- p.50 / Chapter 3.3 --- Recursive construction of larger 0-1 sorter --- p.56 / Chapter Chapter 4 --- Epilogue --- p.59 / Chapter 4.1 --- Directions of further studies --- p.59 / Chapter 4.1.1 --- Synchronization within the same stage --- p.59 / Chapter 4.1.2 --- Layout complexity --- p.60 / Chapter 4.1.3 --- Statistical initialization of running parity --- p.62 / Chapter 4.2 --- Conclusion --- p.63 / REFERENCES --- p.65

by Lam Wan. / Thesis submitted in: November 1997. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1998. / Includes bibliographical references (leaves 96-102). / Abstract also in Chinese. / Part I / Chapter 1 --- Introduction to Integrated Intelligent Personal Communication System --- p.1 / Chapter 2 --- The Switching Architecture --- p.5 / Chapter 2.1 --- The Overall Switching Architecture --- p.6 / Chapter 2.2 --- Switching Module --- p.10 / Chapter 2.2.1 --- Traffic Routing in Switching Module --- p.11 / Chapter 2.2.2 --- Structure of Switching Module --- p.15 / Chapter 2.2.3 --- Wireless Base Interface --- p.16 / Chapter 2.2.4 --- Trunk Interface --- p.18 / Chapter 2.2.5 --- Analog Interfaces --- p.18 / Chapter 2.3 --- Network Intelligence --- p.19 / Chapter 2.4 --- Wireless Part --- p.21 / Chapter 2.4.1 --- Call-Setup in IIPCS --- p.24 / Chapter 2.4.2 --- Handoff --- p.25 / Chapter 2.4.3 --- Wireless Base --- p.27 / Chapter 2.5 --- Downstream Wired Extensions --- p.28 / Chapter 2.6 --- Upstream Wired Part --- p.28 / Chapter 2.7 --- Voice System --- p.28 / Chapter 2.8 --- Features of the IIPCS --- p.29 / Chapter 3 --- Concluding Remarks --- p.33 / Chapter 3.1 --- Summary --- p.35 / Chapter 3.2 --- Directions for Further Research --- p.36 / Part II / Chapter 4 --- Introduction to Next-Generation Switch --- p.37 / Chapter 5 --- Architecture of Next-Generation Switch --- p.41 / Chapter 5.1 --- Overall Architecture of Next-Generation Switch --- p.42 / Chapter 5.1.1 --- Interface module --- p.44 / Chapter 5.1.2 --- Packetizer --- p.46 / Chapter 5.2 --- Concentration Fabric --- p.50 / Chapter 5.3 --- Shared-Buffer Memory Switch --- p.53 / Chapter 6 --- Concentration Networks --- p.56 / Chapter 6.1 --- Background of Concentration Networks --- p.56 / Chapter 6.2 --- k-Sorting --- p.63 / Chapter 6.3 --- Concentrator --- p.72 / Chapter 6.3.1 --- Nk-to-k Concentrator --- p.73 / Chapter 6.3.2 --- Match between Circles with Cost Reduction --- p.75 / Chapter 6.4 --- The Structure of a Molecule --- p.78 / Chapter 6.5 --- Summary --- p.81 / Chapter 7 --- Lock-Latch Algorithm --- p.82 / Chapter 8 --- Performance Evaluation --- p.88 / Chapter 9 --- Concluding Remarks --- p.93 / Chapter 9.1 --- LSI Implementation --- p.94 / Chapter 9.2 --- Summary --- p.95 / Bibliography

Cheng, Lap Kei. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references. / Abstracts in English and Chinese. / ABSTRACT --- p.I / 摘要 --- p.II / ACKNOWLEDGEMENTS --- p.V / TABLE OF CONTENTS --- p.IV / INTRODUCTION --- p.0 / Chapter 1.1 --- Different ways to achieve all-optical tunable delay --- p.2 / Chapter 1.1.1 --- Optical buffer realized with optical switching --- p.2 / Chapter 1.1.2 --- Slow light technique --- p.3 / Chapter (i) --- Basics of slow light --- p.4 / Chapter (ii) --- Slow light via electromagnetically induced transparency (EIT) --- p.6 / Chapter (iii) --- Slow light via coherent population oscillation (CPO) --- p.7 / Chapter (iv) --- Slow light via optical parametric amplification (OPA) --- p.8 / Chapter (v) --- Slow light via stimulated Raman and Brillouin scattering --- p.8 / Chapter 1.1.3 --- Tunable delay using wavelength conversion together with chromatic dispersion --- p.10 / Chapter 1.1.4 --- Comparison of different schemes for constructing all-optical delay line --- p.11 / Chapter 1.2 --- Overview of the thesis --- p.12 / References --- p.14 / ALL-OPTICAL SWITCHING OF DPSK SIGNAL IN AN SOA USING NONLINEAR POLARIZATION ROTATION --- p.18 / Chapter 2.1 --- Introduction --- p.19 / Chapter 2.2 --- Birefringence and nonlinear polarization rotation --- p.20 / Chapter 2.3 --- Differential-phase-shift keying (DPSK) modulation format --- p.22 / Chapter 2.4 --- Experimental setup --- p.23 / Chapter 2.5 --- Experimental results --- p.25 / Chapter 2.6 --- Conclusion --- p.29 / References --- p.30 / WIDEBAND SLOW LIGHT VIA STIMULATED BRILLOUIN SCATTERING IN AN OPTICAL FIBER USING A PHASE-MODULATED PUMP --- p.32 / Chapter 3.1 --- Introduction --- p.33 / Chapter 3.2 --- Stimulated Brillouin scattering (SBS) --- p.34 / Chapter 3.3 --- Slow light via SBS --- p.35 / Chapter 3.4 --- Experimental setup --- p.37 / Chapter 3.5 --- Experimental result --- p.39 / Conclusion --- p.42 / References --- p.43 / SIGNAL WAVELENGTH TRANSPARENT SBS SLOW LIGHT USING XGM BASED WAVELENGTH CONVERTER AND BRILLOUIN FIBER LASER --- p.45 / Chapter 4.1 --- Introduction --- p.46 / Chapter 4.2 --- Brillouin fiber laser and XGM wavelength converter --- p.47 / Chapter 4.3 --- Operating principle --- p.50 / Chapter 4.4 --- Experimental setup and results --- p.51 / Conclusion --- p.56 / References --- p.57 / ALL-OPTICAL TUNABLE DELAY LINE FOR CHANNEL SELECTION IN A 40-GB/S OPTICAL TIME DIVISION MULTIPLEXING SYSTEM --- p.59 / Chapter 5.1 --- Introduction --- p.60 / Chapter 5.2 --- Principle of four-wave mixing --- p.61 / Chapter 5.3 --- Channel selection in an OTDM system --- p.63 / Chapter 5.4 --- Experimental setup --- p.64 / Chapter 5.5 --- Experimental results --- p.67 / Conclusion --- p.70 / References --- p.71 / TUNABLE OPTICAL DELAY WITH CSRZ-OOK TO RZ-OOK OPTICAL DATA FORMAT CONVERSION USING FOUR-WAVE MIXING WAVELENGTH CONVERSION AND GROUP VELOCITY DISPERSION --- p.73 / Chapter 6.1 --- Introduction --- p.74 / Chapter 6.2 --- Carrier-Suppressed Return-to-Zero --- p.76 / Chapter 6.3 --- Operating Principle --- p.77 / Chapter 6.4 --- Experimental setup --- p.79 / Chapter 6.5 --- Experimental result --- p.81 / Conclusion --- p.86 / References --- p.87 / CONCLUSION --- p.90 / Chapter 7.1 --- Summary of work --- p.90 / Chapter 7.2 --- Prospects of future work --- p.92 / APPENDIX: LIST OF PUBLICATIONS A

Not available / text

published_or_final_version / abstract / Electrical and Electronic Engineering / Doctoral / Doctor of Philosophy

The focus of this dissertation is the development of resource management schemes for integrated networks, with the major contributions being: (i) the development of an optimal adaptive buffer management scheme for the packet-switched subsystem, (ii) the integration of a moveable-boundary hybrid switching scheme with the time assigned speech interpolation technique for implementing a congestion control mechanism for the packet-switched subsystem, and (iii) the development of an adaptive hierarchical scheme for implementing the access control and routing functions within the circuit-switched subsystem. The problem of buffer management at an integrated network node is formulated as a nonlinear programming problem with a convex objective function and an interative solution technique with fast convergence is proposed for a real-time implementation of the buffer management scheme in practical environments. In order to exercise an additional degree of control over the packet-blocking probability at each hybrid-switched link within the network, a new multiplexing scheme based on the integration of the moveable-boundary hybrid switching scheme and the time assigned speech interpolation technique is presented in this dissertation. The tradeoff between the corresponding decrease in the packet blocking probability and the increase in the circuit freezeout fraction is demonstrated by a detailed queueing analysis of the multiplexer. Specific algorithms are also presented in this dissertation for the solution of the access control and routing problems within the circuit-switched subsystem. In particular, an access control scheme is developed by solving an integer programming problem formulated using the policy of complete partitioning of the available bandwidth among the competing user classes. As an alternative to the completely partitioned approach, the problem of traffic routing is considered in a network that supports homogeneous traffic classes based on the policy of complete sharing. Finally, for the general case of networks with heterogeneous traffic classes, a hierarchical scheme is developed for the implementation of the access control and the routing functions at two functional levels, where the access control is implemented by the network supervisor who solves an appropriate linear integer programming problem periodically, and the routing function is handled by the individual nodes of the network on a distributed basis. (Abstract shortened with permission of author.)

Lam Wan. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2001. / Includes bibliographical references (p. 139-145). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.