7 research outputs found
RF circuit and antenna optimization using space mapping technique
No full textThe use of EM simulation in circuit optimization in RF domain is very common. There are some problems with this kind of optimization. Firstly, in many cases EM simulations are expensive process. Another problem is the relationship between circuit response and design variable is not an easy function in many cases.
In order to overcome these problems a technique named space mapping (SM) was suggested. This technique has been experimentally proved as effective and efficient technique to do optimization including RF circuit optimization.
From large number of SM algorithm variants there are only a few that are designed specially for constrained optimization. In this report some constrained SM algorithms using projection method are investigated. These algorithms are variants of Aggressive Space Mapping (ASM) and Aggressive Output Space Mapping (AOSM) and designed to handle convex constraints. The methods are used to optimize multilayer LTCC bandpass filter. The numerical results shows good convergence rate.
All of the constrained SM algorithms designed in this report use equivalent circuit and embedded knowledge in coarse model. Parameter extraction is done using circuit tuning based on physical augmentation. In order to test the efficiency of the tuning method, this tuning method is used to tune multilayer LTCC bandpass filter equivalent circuit. The result shows this tuning algorithm is effective and efficient.
This circuit tuning algorithm is a derivative of a modeling algorithm based on physical augmentation. The use of this modeling technique in producing equivalent circuit for antenna is investigated. It is shown experimentally that this modeling algorithm is effective and efficient.
Although some of proposed constrained SM algorithms are designed to provide global convergence, the author can not prove the global convergence mathematically. The author can only provide the proof for local convergence of some constrained SM algorithms. Further developments may be done in the direction of finding the mathematical analysis of these SM algorithms, refining the methods especially the AOSM based methods and combining the ASM and AOSM.Bachelor of Engineerin
MULTIMEDIA-BASED INTERACTIVE LEARNING MEDIA FOR DEAF AND MENTALLY DISABLED ELEMENTARY STUDENTS ON SLBN 1 LENGAYANG
No full textThe ability to remember and understand is quite difficult in the learning activities of mentally retarded and deaf students at SLBN 1 Lengayang. This study aims to examine the use of interactive media to assist the learning activities of deaf and mentally retarded students in improving understanding and memory. The data collection method used is the interview method and literature study. While the method used for the design and manufacture of the author's application uses UML and Storyboard which is implemented with Adobe Flash CS6. The result of the design is the Learning Application to Know Our Body Parts and a companion book to Know the Parts of the Body. Data analysis shows that using interactive learning media for deaf and mentally retarded children can improve memory and understanding with a more interesting and interactive learning process. The use of interactive multimedia in the form of applications and companion books is a new learning medium in the learning process that provides a different learning experience for deaf and mentally retarded students
Automatic Detection of Oral Squamous Cell Carcinoma from Histopathological Images of Oral Mucosa Using Deep Convolutional Neural Network
No full textWorldwide, oral cancer is the sixth most common type of cancer. India is in 2nd position, with the highest number of oral cancer patients. To the population of oral cancer patients, India contributes to almost one-third of the total count. Among several types of oral cancer, the most common and dominant one is oral squamous cell carcinoma (OSCC). The major reason for oral cancer is tobacco consumption, excessive alcohol consumption, unhygienic mouth condition, betel quid eating, viral infection (namely human papillomavirus), etc. The early detection of oral cancer type OSCC, in its preliminary stage, gives more chances for better treatment and proper therapy. In this paper, author proposes a convolutional neural network model, for the automatic and early detection of OSCC, and for experimental purposes, histopathological oral cancer images are considered. The proposed model is compared and analyzed with state-of-the-art deep learning models like VGG16, VGG19, Alexnet, ResNet50, ResNet101, Mobile Net and Inception Net. The proposed model achieved a cross-validation accuracy of 97.82%, which indicates the suitability of the proposed approach for the automatic classification of oral cancer data
Coherent control of a strongly driven silicon vacancy optical transition in diamond
No full text© 2017 The Author(s). The ability to prepare, optically read out and coherently control single quantum states is a key requirement for quantum information processing. Optically active solid-state emitters have emerged as promising candidates with their prospects for on-chip integration as quantum nodes and sources of coherent photons connecting these nodes. Under a strongly driving resonant laser field, such quantum emitters can exhibit quantum behaviour such as Autler-Townes splitting and the Mollow triplet spectrum. Here we demonstrate coherent control of a strongly driven optical transition in silicon vacancy centre in diamond. Rapid optical detection of photons enabled the observation of time-resolved coherent Rabi oscillations and the Mollow triplet spectrum. Detection with a probing transition further confirmed Autler-Townes splitting generated by a strong laser field. The coherence time of the emitted photons is comparable to its lifetime and robust under a very strong driving field, which is promising for the generation of indistinguishable photons
