QD photodectors, among many field of applications, will probably be most interesting if we can use it in Night Vision Camera, i.e. thermal imaging.
Electron in QD is like “particle in a 3D box”, its energy bands are discrete, spike-type Density of State. Thats why Quantum Dot is called “Artificial Atom”.
Due to this atom-like behavior , QD detector is fundamentally resistant to thermal noise. Unfortunately, we still have to operate these QD detectors at very low temperature, say, at 80 degree kelvin, and the cooling system increases both cost and size of detectors.
However, if researchers can find a way to remove some problems in QD fabrication technique (i.e. uniform growth of QDs), it will definitely outdo all present IR detectors both in terms of performance and costing. Thats the key for us, the engineers !
Dr. Harun-Ur-Rashid is working on wireless interconnects to replace global interconnects on chip for quite some time. This is an excellent approach as the delay caused by global interconnects increases with device scaling e.g. 45nm technology will have more delay caused by global interconnects than 90nm tech. In this wireless interconnect topology, my part was to design an delay element for receiver. This delay element is needed to recover data stream from BPSK modulated wave. Design is optimized for 200ps delay (two clock cycles for 10GHz data pulse). Three 2nd order all-pass filters are used in this design.
Gesture Recognition provides an efficient human-computer interaction for interactive and intelligent computing. It is becoming increasingly popular for applications in ubiquitous computing environment. Accelerometer based methods have proven themselves to be competitive in terms of both portability and recognition accuracy. But there are only a few algorithms which have achieved moderate accuracies in user independent gesture recognition. In this work, we address the problem of user independent gesture recognition by using Dynamic Time Warping algorithm. A novel accelerometer based user independent hand gesture recognition algorithm is proposed. For validation of accuracy the test was run over 3200 samples collected from 5 users over 5 days. Simulation results reveal a superior performance, in terms of accuracy. Our algorithm also achieves a very high accuracy in user dependent gesture recognition and can be simultaneously applied in user dependent case. To tackle the computational complexity and timing constraints, the acceleration of DTW algorithm is proposed and implemented. The improvement of speed obtained is sufficient to develop a continuous recognition system in future. Furthermore, a functional prototype is implemented to demonstrate the algorithm in a discrete device. A high recognition accuracy and possibility of simpler implementation gives this algorithm an upper hand among competitive methods published in recent literature.