Electrical and Electronics

The system consists of a standard microprocessor and a hardware accelerator for Gaussian mixture model (GMM) emission probability calculation implemented on a field-programmable gate array. The GMM accelerator is optimized for timing performance by exploiting data parallelism. In order to avoid large memory requirement, the accelerator adopts a double buffering scheme for accessing the acoustic parameters with no assumption made on the access pattern of these parameters. Experiments on widely used benchmark data show that the real-time factor of the proposed system is 0.62, which is about three times faster than the pure software-based baseline system, while the word accuracy rate is preserved at 93.33%. As a part of the recognizer, a new adaptive beam-pruning algorithm is also proposed and implemented, which further reduces the average real-time factor to 0.54 with the word accuracy rate of 93.16%. The proposed speech recognizer is suitable for integration in various types of voice (speech)-controlled applications.

Images containing faces are essential to intelligent vision-based human computer interaction, and research efforts in face processing include face recognition, face tracking, pose estimation, and expression recognition. The rapidly expanding research in face processing is based on the premise that information about a user’s identity, state, and intent can be extracted from images and that computers can then react accordingly, e.g., by knowing person’s identity, person may be authenticated to utilize a particular service or not. A first step of any face processing system is registering the locations in images where faces are present. However, face registration for whole database is a challenging task because of variability in scale, location, orientation (up-right, rotated), and pose (frontal, profile). Facial expression, occlusion, and lighting conditions also change the overall appearance of face. The Image registration algorithm will register all these images present in the database. The face recognition algorithm which is insensitive to large variation in lighting direction and facial expression is to be implemented. Taking a pattern classification approach, each pixel in an image can be considered as a coordinate in a high-dimensional space. The advantage of this is that the images of a particular face, under varying illumination but fixed pose, lie in a 3D linear subspace of the high dimensional image space—if the face is a Lambertian surface. However, since faces are not truly Lambertian surfaces and do indeed produce self-shadowing images will deviate from this linear subspace. Rather than explicitly modeling this deviation, project the image into a subspace in such a manner which discounts those regions of the face with large deviation. This is achieved by using dimension reduction techniques like Principal component analysis (PCA), Linear Discriminant analysis (LDA), Laplacian faces, and other modified approaches like A Priori Laplacian and PCDA.

Number of plant species, in present time, is on the verge of extinction. Thus owing to this threat to plant species their identification has become very important. This paper presents a technique that has been implemented for the identification of plant species based on leaf images. The plant species identification involves three steps such as pre-processing, feature extraction and classification.

Space-based solar power (SBSP) is the concept of collecting solar power in space for use on Earth. It has been in research since the early 1970s.This paper projects how SBSP would differ from current solar collection methods in that the means used to collect energy would reside on an orbiting satellite instead of on Earth's surface. Besides the cost of implementing such a system, SBSP also introduces several new hurdles, primarily the problem of transmitting energy from orbit to Earth's surface for use. Since wires extending from Earth's surface to an orbiting satellite are neither practical nor feasible with current technology, SBSP designs generally include the use of some manner of wireless power transmission. The collecting satellite would convert solar energy into electrical energy on-board, powering a microwave transmitter or laser emitter, and focus its beam toward a collector (rectenna) on the Earth's surface. Radiation and micrometeoroid damage could also become concerns for SBSP

An adequate blood supply is required for the normal functioning of all organs in the body. However this flow can be impeded due to several reasons, thrombus or clot being a major one. Detection of such clots is done with the help of an imaging technique called “DOPPLER ULTRASONOGRAPHY” by transmitting an ultrasound pulse and calculating the frequency of the received signal. If there is no clot the reception is always a homogenous signal, whereas if a certain part being sonographed has a clot the received signal shows a variance in frequency which depends upon the distance from the receiver. Hence blood clot detection in ultrasonography is based on obtaining the frequency variance and applying adaptive thresholding. In this work we propose a ZOOM FFT based technique followed by automated adaptive thresholding to detect the clot. Data files are synthesized by simulation in MATLAB. Also the entire system is developed in a Matlab environment for the system to be both simple and cost-effective. Experimental results show that the accuracy of the system is very high even under noisy conditions.

Distance learning over the past several years has gained in popularity not only as a way to offer instruction in locations without local expertise, but also as a cost effective method where limited enrollment at one location would not normally warrant offering the course. In this paper, we use the embedded board with interface module to implement a remote electronic measurement system which includes the power supply, signal generator and oscilloscope. This design uses an embedded board to replace a computer, since the embedded board has the advantage of being easily carried, a real-time operation, a low cost, and programmable. In addition users can operate this measurement system with the help of the operating system and the TCP/IP modules to connect to the Internet. By using the on board operating system, our design provides the step by step function to help user operate, such as keying in the waveform parameters with the embedded board keyboard, providing the waveforms and then connecting the circuit to the embedded electronic measurement system. This design can also show the waveform measured by the embedded measurement system in the embedded board LCM (Liquid Crystal Monitor). Because the file size of measured waveforms is very small around 1 KB, the prototype system show that the main delay time to receive the measurement waveform at the server site from the client site is within 2.5 seconds In the server, the observers can observe many waveforms from the different clients. If the client’s waveform is error, the observer can also send the required waveforms to the client’s embedded measurement system and input these waveforms to the testing circuits and re-observe the circuit measurement and help the user to debug.

Power has been an important issue for the present day microelectronic circuits of Soc designs. In the entire phase of design controlling power and dealing with power dissipation is very important. There are six leakage components in a MOS transistor. About 50 % of the total power consumption is through leakage components alone. Out of this 40 % power dissipation is through transistors. Out of them subthreshold leakage and gate leakage are of important concern for sub 100 nm devices. In this work we presented the main leakages in a 6T- SRAM cell. Later we presented an existing variable body biasing method and its performance. We proposed two novel methodologies HSVR, GSVE to reduce the subthreshold and gate leakage components. These methods are toplogy based leakage reduction methods applied for leakage reduction in the circuit level. We observed considerable reduction of both the components through the proposed methods.

The buildings situated on hill slopes in earthquake prone areas are generally irregular, torsionally coupled & hence, susceptible to serve damage when affected by earthquake ground motion. Such buildings have mass & stiffness varying along the vertical & horizontal planes, resulting the center of mass & center of rigidity do not coincide on various floors, hence they demand torsional analysis, in addition to lateral forces under the action of earthquakes. These unsymmetrical buildings require great attention in the analysis & design. Analysis of hill buildings is somewhat different than the buildings on leveled ground, since the column of hill building rests at different levels on the slope. The shorter column attracts more forces & undergoes damage, when subjected to earthquakes.

This paper proposes a new Tabu Search Enhanced Artificial Bee Colony (TSEABC) algorithm that integrates the concepts of Tabu Search (TS) and Artificial Bee Colony (ABC) techniques. The proposed algorithm is applied to solve Economic Load Dispatch(ELD) problems considering of prohibited operating zones and ramp rate limits of thermal units, capacity limits and power balance constraints. In the proposed TSEABC algorithm, the best attributes of both TS and ABC are utilized, and it is capable of finding the better optimal solution for the combinatorial optimization problems. For validating the proposed algorithm, it has been tested on the standard three, six and fifteen unit test systems. The comparison of results show that the proposed TSEABC algorithm is well suitable for solving non-linear economic dispatch problems, and it is doing better than the EP, PSO and other modern metaheuristic optimization methods reported in the recent literatures.

This paper presents the development and implementation of a new and practical fuzzy switching position controller for process industry control. Compatible with the structure of a bang–bang controller, a two-input fuzzy switching position controller with five rules is proposed. The idea is based on constructing a fuzzy switching control law which is functionally analogous to a traditional bang–bang controller. Fuzzy logic rules are used to enable an improved version of the conventional bang–bang control. The objective is to replace the conventional bang–bang controller with the proposed fuzzy switching position controller. A modified three-stage bang–bang controller is also designed and implemented. This controller is tested under the same conditions as that of the proposed fuzzy switching position controller, and its performance is used as a basis of comparison by which both controllers are measured. Both controllers are developed using MATLAB using the position control of a brushless dc motor drive found in the majority of these new robotic devices. Simulation Results shows that the fuzzy switching position controller produces adequate control performance, particularly in handling nonlinearities and external disturbances. The efficacy of the fuzzy switching position controller is demonstrated by its positive results, practicality, and feasibility in the process industry sector, when compared with a modified traditional bang–bang controller.