Selective epitaxial growth in silicon on insulator: Planarity and mass flow

Seeded zone-melt recrystallization using a dual electron beam system has been performed on silicon-on-insulator material, which was prepared with single-crystal silicon filling of the seed windows by selective epitaxial growth. The crystal quality has been assessed by a variety of microscopic techniques, and it is shown that single-crystal films 0.5-1.0 μm thick over 1.0 μm of isolating oxide may be prepared by this method. These films have considerably less lateral variation in thickness than standard material, in which the windows are not so filled. The filling method is suitable for both single- and multiple-layer silicon-on-insulator, and gives the advantages of excellent layer uniformity after recrystallization and improved planarity of the whole chip structure. Experiments using various amounts of seed window filling have shown that the lateral variations of silicon film thickness seen in unplanarized material are due to stress relief in the cap oxide when the silicon film is molten, rather than the effect previously postulated in which they were assumed to be due to the contraction of silicon on melting.

Dual-service wireless LAN antenna-remoting system using uncooled directly modulated D'FB laser diodes

We report a fibre-optic wireless distribution system, which allows antenna-remoting of a dual-service IEEE 802.11b/g WLAN operating at 2.4GHz up to 700m over low-bandwidth 62.5/125μm MMF using highly linear uncooled directly modulated laser diodes. © 2004 Optical Society of America.

Electronic load interface for improving PV dual-converter system operational margin

An electronic load interface (ELI) for improving the operational margin of a photovoltaic (PV) dual-converter system under dynamic conditions is presented. The ELI - based on a modified buck-boost converter - interfaces the output of the converters and the load system. It improves the operational margin of the PV dual-converter system by extending the conditions under which the dual-converter system operates at the maximum power point. The ELI is activated as and when needed, so as minimise system losses. By employing the ELI, utilisation and efficiency of a PV dual-converter system increases. In general, the concept of the ELI can be applied to multi-converter PV systems - such as multi-converter inverters, and multi-converter DC-DC converter systems - for performance and efficiency improvement. © 2013 The Institution of Engineering and Technology.

Development of the 2003 CU-HTK conversational telephone speech transcription system

This paper describes the development of the 2003 CU-HTK large vocabulary speech recognition system for Conversational Telephone Speech (CTS). The system was designed based on a multi-pass, multi-branch structure where the output of all branches is combined using system combination. A number of advanced modelling techniques such as Speaker Adaptive Training, Heteroscedastic Linear Discriminant Analysis, Minimum Phone Error estimation and specially constructed Single Pronunciation dictionaries were employed. The effectiveness of each of these techniques and their potential contribution to the result of system combination was evaluated in the framework of a state-of-the-art LVCSR system with sophisticated adaptation. The final 2003 CU-HTK CTS system constructed from some of these models is described and its performance on the DARPA/NIST 2003 Rich Transcription (RT-03) evaluation test set is discussed.

Unsupervised training for mandarin broadcast news and conversation transcription

A significant cost in obtaining acoustic training data is the generation of accurate transcriptions. For some sources close-caption data is available. This allows the use of lightly-supervised training techniques. However, for some sources and languages close-caption is not available. In these cases unsupervised training techniques must be used. This paper examines the use of unsupervised techniques for discriminative training. In unsupervised training automatic transcriptions from a recognition system are used for training. As these transcriptions may be errorful data selection may be useful. Two forms of selection are described, one to remove non-target language shows, the other to remove segments with low confidence. Experiments were carried out on a Mandarin transcriptions task. Two types of test data were considered, Broadcast News (BN) and Broadcast Conversations (BC). Results show that the gains from unsupervised discriminative training are highly dependent on the accuracy of the automatic transcriptions. © 2007 IEEE.