No description available.

Brian Harold May's thesis (doctoral)--Imperial College of Science, Technology and Medicine, London, 2007. / Includes bibliographical references (p. [205]-215).

Thesis--Columbia University. / Photocopy (positive) of typescript, made by University Microfilms. Bibliography: leaves 244-258.

No description available.

Thesis (Ph. D.)--University of Texas at Austin, 2000. / Vita. Includes bibliographical references. Available also from UMI/Dissertation Abstracts International.

Thesis (Ph. D.)--University of Texas at Austin, 2000. / Vita. Includes bibliographical references (leaves 279-281). Available also in a digital version from Dissertation Abstracts.

Debris discs are often modelled assuming compact dust grains, but more and more evidence for the presence of porous grains is found. We aim at quantifying the systematic errors introduced when modelling debris discs composed of porous dust with a disc model assuming spherical, compact grains. We calculate the optical dust properties derived via the fast, but simple effective medium theory. The theoretical lower boundary of the size distribution - the so-called 'blowout size' - is compared in the cases of compact and porous grains. Finally, we simulate observations of hypothetical debris discs with different porosities and feed them into a fitting procedure using only compact grains. The deviations of the results for compact grains from the original model based on porous grains are analysed. We find that the blowout size increases with increasing grain porosity up to a factor of 2. An analytical approximation function for the blowout size as a function of porosity and stellar luminosity is derived. The analysis of the geometrical disc set-up, when constrained by radial profiles, is barely affected by the porosity. However, the determined minimum grain size and the slope of the grain size distribution derived using compact grains are significantly overestimated. Thus, the unexpectedly high ratio of minimum grain size to blowout size found by previous studies using compact grains can be partially described by dust grain porosity, although the effect is not strong enough to completely explain the trend.

Not available / text

Thesis Imperial College of Science, Technology and Medicine, London, 2007. / Includes bibliographical references.

Satellite IP networks are characterized by high bit error rates, long propagation delays, low bandwidth feedback links, and persistent fades resulting from varying weather patterns. A new unicast transport protocol is designed to address all the above challenges. Two new algorithms, Jump Start and Quick Recovery, are presented to replace the traditional Slow Start algorithm and to recover rapidly from multiple segment losses within one window of data. The characteristics of satellite IP networks also distinguish satellite multicasting from multicasting in terrestrial wirelined networks. A reliable data multicast transport protocol, TCP-Peachtree, is proposed to solve the acknowledgment implosion and scalability problems in satellite IP networks. Developments in space technology are enabling the realization of deep space missions. The scientific data from these missions need to be delivered to the Earth successfully. To achieve this goal, the InterPlaNetary Internet is proposed as the Internet of the deep space planetary networks, which is characterized by extremely high propagation delays, high link errors, asymmetrical bandwidth, and blackouts. A reliable transport protocol, TP-Planet, is proposed for data traffic in the InterPlaNetary Internet. TP-Planet deploys rate-based additive-increase multiplicative-decrease (AIMD) congestion control and replaces the inefficient slow start algorithm with a novel Initial State algorithm that allows the capture of link resources in a very fast and controlled manner. A new congestion detection and control mechanism is developed and a Blackout State is incorporated into the protocol operation. Multimedia traffic is also one part of the aggregate traffic over InterPlaNetary Internet backbone links and it has additional requirements such as minimum bandwidth, smooth traffic, and error control. To address all the above challenges, RCP-Planet is proposed. RCP-Planet consists of two novel algorithms, i.e., Begin State and Operational State. The protocol is based on a novel rate probing mechanism and a new rate control scheme to update the media rate smoothly based on the observed rate for the probing sequence.