Piezomachined ultrasonic transducer (PMUT) arrays are commonly found in applications in the field of ultrasonography and gesture recognition systems. Their application for bio and chemical sample preparation is another possibility, based on their beam steering and acoustic field manipulation capabilities. Post-fabrication non-destructive measurement of key device temporal and spatial parameters is required in order to adjust either simulation models or tune fabrication steps. In this work we report an optical testing setup for measuring the acoustic spectrum of PMUT devices and arrays, characterize maximum deflection of PMUTs and piezopumps and investigate the load effect of electrical contacts on the spatial and temporal oscillation behavior of these piezoelectric structures. Spatial parameters are evaluated with digital holography. We employ this testing setup to measure our own designed PMUT structures which were fabricated at IME-Singapore, evaluating the relative merits of the PMUT design parameters.
WSN devices forensics as a field within forensic science is at an early stage when compared to traditional computer forensics. The increase and proliferation of WSN has noted the urgent need for the creation of new analysis tools and techniques. This experimental research implements an approach to capture Flash storage sensor node. The storage data log is extracted from multiple sensor nodes to identify any change on the log storage. Our major contribution is a mechanism for the extraction, analysis of forensic data for IRIS WSN deployments by a tool which can capture storage flash dumps from devices running TinyOS. Two parts of software were used; one installed on the mote itself to extract the storage data then send it to the PC serial port, the second part is resides on the PC that captures data received from serial port.
It is common to analyze and evaluate people's strategies in very isolated situations in which the strategy space is not rich and consequently - not realistic. In this paper we present Junior High Game - a game that has the following attributes which, when combined, make the game unique (and we argue more realistic) than other games analyzed in the literature: * A large number of people and machine agents interact with each other repeatedly. * Interactions are personal: individual-to-individual. * Interactions include opportunities to cooperate, defect, and punish. * Each person and machine has limited resources and unequal capabilities. We conducted a series of user studies from which we observe and evaluate the strategies used by people in this game and we present results that give notion of the possible successful and unsuccessful strategies in Junior High Game.
The essence of object classification lies in the attributes of an object. State-of-the-art object detectors typically use shape information as a low level feature representation to capture the local structure of an object. This paper presents the concept of merging the edge and color information, which could lead to significant improvement in performance for object detection. In real life scenarios, often, we need to recognize objects that are at captured at different angles. In this paper, we have identified the color along the edge as the importance features for object recognition which we have used with the provided video sequence of an objects. Our approach of extracting these features along the edge of an object with the help of edge detection algorithm can be seen as a simple but important features, as we are able to prove with good results using Random Forest ensemble learning approach.
Probabilistic forecasts account for the uncertainty in the prediction helping the decision makers take optimal decisions. With the emergence of renewable technologies and the uncertainties involved with the power generated through them, probabilistic forecasts can come to the rescue. Solar power is an emerging technology and as the technology matures there will be a need for forecasting the power generated days ahead. In this study, an ensemble approach for probabilistic forecasting is used with different machine learning algorithms and different initial settings assuming normal distribution for the forecasts. It is observed that having multiple models with different initial settings gives exceedingly better results when compared to individual models.
It is expected that financial infrastructure of smart grids would involve smart meter integrated token based payment system. Therefore it is important to identify security requirements of such systems. In this paper, we identified security requirements for token based payment system, the Bitcoin. The requirements are elicited using SQUARE method and evaluated. The security analysis has been conducted.
File carving is a type of digital forensics recovery technique which focuses on recovering files from digital media without using file system metadata. This technique can be used in several situations such as recovering deleted files or recovering files from storage media with corrupted or unknown file systems. The aim of this research is to develop a file carving technique that is capable of recovering fragmented video files without using file system metadata.
Supervisory Control and Data Acquisition (SCADA) systems are process control systems which monitor and control the physical processes in industrial facilities such as factories, power generation and distribution systems, oil and gas facilities and nuclear power plants. SCADA collects data about the physical status and sends the commands to control the physical processes in a feedback control network. In this paper we present SCADA log analysis approach by which we detect any malicious behavior if exist in SCADA network. The collection of the log files from a real SCADA system is currently impossible due to the sensitivity nature of such systems, an alternative good sources were Honeynet log from an open source and logs from a small PLC based SCADA system which was built in another project. We performe in-depth analysis process using Splunk which provides very good capabilities for log and unstructured data analysis.
Supervisory Control and Data Acquisition (SCADA) is a set of systems which are used to monitor and control remote equipment that can be found in general power plants, nuclear power plants, and any other critical infrastructure entities. As the technology changes the needs came to connect the SCADA into the internet; which means increasing the chances of exposing the critical infrastructure into cyberattacks. In this paper, we present a new approach to monitor Programmable Logical Controllers (PLC) that is focused on analyzing packets from PLC and open sources to understand the normal behavior of the equipment and reflect it into analyst monitoring dashboard. As a proof of concept, we will build small PLC temperature sensor to proof the benefits of our approach in protecting PLC from cyberattacks.
There has been a growing interest in wind resources in the Gulf region, not only for evaluating wind energy potential, but also for understanding and forecasting changes in wind. In this study, the seasonal autoregressive integrated moving average (SARIMA) models with different combinations of parameters are applied to the observed monthly average of wind speed time series in the United Arab Emirates (UAE), measured from the international airport of Abu Dhabi. This model deals with apparent seasonality in the data to help understand the characteristics of original and modified wind speed time series, and forecast the future wind speed. The best SARIMA model is selected based on the root mean square(RMSE) and relative RMSE of each model. Results indicate that the SARIMA models provide a good fit to wind data in the UAE.