Amphibious landings and small boat operations are normally conducted only in benign wave conditions. An unexpected encounter with an isolated freak wave may damage equipment and prevent mission accomplishment. This study examines the occurrence of unusually large waves using data sets obtained with bottom mounted pressure sensors and wave buoys in the DUCK 94, SHOWEX, and SAX 04 experiments. All of the experiments include wave records from high energy events. After correcting the raw pressure data for hydrodynamic attenuation over the water column, the statistics of wave heights were evaluated and compared with the theoretical Rayleigh distribution of a narrow-band linear wave field. Observations from deep water sites follow the Rayleigh distribution well, even in extreme sea states, indicating that strong nonlinearity does not have a major effect on wave height statistics. However, during high energy events at shallow water sites, there are significantly less measured wave heights in the right-hand tail of the distribution of wave heights than the theoretical Rayleigh distribution would predict. These results show that waves become more homogeneous in height as they propagate into shallower water, possibly owing to breaking and nonlinear effects. While the observed wave statistics do not suggest a frequent occurrence of freak waves, isolated large waves were indeed observed, even in benign conditions. Further studies are needed to assess their risk to Navy operations.

Centralized light source (CLS) at the central office is an attractive solution for low-cost implementation of WDM-PON, as it eliminates the need of wavelength-specific transmitters and wavelength management at the optical network units (ONU). CLS can be realized by either a carrier-distributed scheme or a remodulation scheme. In both schemes, however, the upstream signal is susceptible to the interferometric crosstalk induced by the beating between the upstream signal and the back-reflected light due to the intrinsic Rayleigh Backscattering (RB), both of which are of the same wavelength. We propose and demonstrate a simple and novel scheme to suppress the RB noise in the carrier-distributed WDM-PON. Differential phase-shift keying (DPSK) is used as the upstream modulation format and the destructive port of the delay-interferometer (DI) is employed to demodulate the upstream DPSK signal. As the spectrum of RB towards the OLT is narrow due to the very narrow spectrum of the distributed carrier, the RB noise can be considerably rejected by the notch filter-like destructive port of the DI at the OLT, which is used simultaneously to demodulate the upstream DPSK signal. The scheme can also be extended to the application in the remodulation-based WDM-PON, as long as the downstream signal has a narrow spectrum (i.e. via reducing downstream modulation depth). A unique feature of the DPSK signal with reduced modulation depth (RMD-DPSK) is that it can be demodulated by DI's destructive port without extinction ratio (ER) degradation, whereas the demodulated signal from DI's destructive port has a very low ER and can be used as the source for upstream remodulation. We also proposed a novel offset-Manchester coding to suppress Rayleigh noise in electrical domain via a postdetection high-pass filter. / Foreseeing the rapidly growing demand for multimedia services and the trend of service convergence, the penetration of optical fiber in access network is an ultimate solution to break the last-mile bottleneck imposed by the 100-year-old copper network. One of the most promising solutions to realize optical access is the passive optical network (PON), in which the network infrastructure is shared by many subscribers and has no active elements between the central office and the customer. Thus a PON requires neither electrical power nor active management, leading to effective reduction in operational expenses. Time-division-multiplexed passive optical networks (TDM-PONs) such as Ethernet PON (EPON) and Gigabit PON (GPON) are being widely deployed in current fiber-based access networks for providing broadband access, offering triple-play services including video, data and voice. In the near future, wavelength-division-multiplexed passive optical network (WDM-PON) can be the enabler of the next-generation optical broadband access that requires large dedicated and symmetric bandwidth, data privacy, and upgrade flexibility. TDM-PONs also can benefit from WDM technologies for capacity upgrade. / With more diverse multimedia and data services available for broadband access, the access network has to be flexible enough to cope with various data or video delivery such as broadcast/multicast services, in addition to the point-to-point traffic. Multicast is more attractive, compared to broadcast, as it allows selective control of the connection for each subscriber individually. Multicast can be easily realized in TDM-PONs as it employs power-splitting at the remote node (RN). However, it is more challenging in WDM-PONs, due to the dedicated connection between the optical line terminal (OLT) and each ONU. Many studies have been carried out to solve this problem. The prior schemes either need relatively complicated multicast control and/or cannot support future proof 10-Gb/s symmetric point-to-point (PtP) transmission. We proposed a novel multicast control scheme for a WDM-PON with 10-Gb/s symmetric bit rate. The multicast data encoded in DPSK format is superimposed onto all PtP channels modulated in inverse return-to-zero format. With an athermal Dr being used at the ONU to demodulate the DPSK signal, the multicast data can be effectively disabled by slight detuning the laser wavelength at OLT, which has negligible effect on the PtP data. The proposed scheme differs from all the previous schemes in that, the multicast control is realized via the inherent wavelength management of WDM systems, rather than via any other additional adjustment such as ER, synchronization, and polarization. v / Xu, Jing. / Adviser: Lian-Kuan Chen. / Source: Dissertation Abstracts International, Volume: 73-06, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2011. / Includes bibliographical references. / 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, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.

D. Tech. Mathematical Technology. / Investigates hyperbolic and pseudohyperbolic equations and the results are applied to higher-order rod approximations for the propagation of the longitudinal stress waves in elastic rods. The main objectives of this thesis are: 1. Provide a unified approach to the derivation of the families of one-dimensional hyperbolic differential equations simultaneously with the associated natural and essential boundary conditions describing longitudinal vibration of finite length rods. 2. Establish a new theoremto shorten the derivation of equations of motion and the corresponding boundary conditions, modelling longitudinal wave propagation in the rod. 3. Prove that, when deriving the higher-order rod equations, the lower-order are still included, thus increasing the number of deformations in the rod or the accuracy of the model. 4. Provide mathematical tools for the classification of the obtained equations. 5. Compare the accuracy of the above-mentioned vibration theories in elastic rods based on the investigation of their frequency spectrums which are not available in the literature. 6. Show how two of the above vibration theories, the Rayleigh-Bishop and Mindlin-Herrmann theories, can be applied to predict wave propagation in a piezoelectric circular cylinder and isotropic conical rod. In both cases a numerical example is given as a simulation of the solution.7. Find general methods for solving problems of longitudinal vibration of finite length rods for all of the above-mentioned theories.