Search

This paper presents the deep learning model by Convolution Neural Network (CNN) in training the crop disease classifier via image classification. A camera will be equipped and applied in artificial intelligent drone to operate as a crop monitoring system used for agriculture. Agriculture productivity is a key component of country economy. Crop diseases can lead to a drop in the quality and quantity of agricultural products. Famers are facing problems to detect the crop diseases accurately in huge region of crops. Therefore, CNN based method for crop disease detection is proposed. Dataset contains of 16,257 color images which has a total of categories have been fed into the model, out of which 10 categories are of diseased crop leaves. The CNN model contains 7 convolution layers with the number of filters 32, 64, two layers with 128 filters, three layers with 256 filters and filter size $3\times 3$ is the proposed approach to perform crop disease classification, with the best testing accuracy of 99.02%. The crops are classified correctly using the suggested CNN design. The suggested CNN design is validated and evaluated which achieves accuracy of 99.02%, 0.98% error, 99% recall, 99% precision and 0.99 score of F-measure. In this paper, achievement of the proposed CNN model is reaching a promising result and simulated successfully in classifying the crop disease.

Throughout the years, authentication processes of individuals’ identities have become essential parts of our modern daily life. These authentication processes also introduced the heavy use of Personally Identifiable Information (PII) in various applications. On the other hand, the continuous increase of using identifiable information –the unauthorized use of such PII–has created rich business opportunities for identity protection service providers. These services usually consist of a monitoring system that continuously searches through the Internet for incidents that supposedly indicates identity theft activities. However, these solutions are largely based on case studies and a quantified method is missing among different identity protection services. This research offers a tool that provides quantitative analysis among different identity protection services. By bringing together previous work in the field, namely the UT Center for Identity (CID) Identity Ecosystem (a Bayesian network mathematical representation of a person’s identity), real world identity theft data, stochastic game theory, and Markov decision processes, we generate and evaluate the best strategy for defending against the theft of personal identity information. One of the research problems that this paper addresses is the computation complexity of quantitatively evaluating identity protection strategies with real world data. In a real world database like Identity Threat Assessment and Prediction (ITAP) project which the UT CID Identity Ecosystem is built on, the number of PII attributes in use are normally in the order of 103. We propose a reinforcement learning algorithm for solving the optimal strategy to protect the user’s identity against a malicious and efficient attacker. We aim to understand how initial individual PII exposure evolves into crucial PII breaches over time in terms of the dynamic integrity of the Identity Ecosystem. Real world identity protection strategies are then translated into the system and we evaluate/simulate their effectiveness against a malicious attacker for quantitative comparison in our experiment. We present the survival analysis to these strategies and calculate the survival gap between these strategies against our active protection strategy as our experiment result. This study is aimed to understand the evolutionary process of identity under attack which may inspire a new direction for future identity protection strategies.

Following another vehicle too close is known as tailgating. Tailgater may not have enough space and time to avoid a collision. Thus there is a need for a fundamental understanding on factors affecting safe distance between vehicles. This paper aims to study such factors and to adopt certain rules and principles into the proposed traffic cellular automata model. Rule of seconds is applied and some deductions are made to attain a realistic traffic model. The simulation also entail human reaction time in perceiving safe distance with the leading vehicle and the simulation results are as expected. However there is a trade off between reaction time and an optimum safe following distance. Hence ideal reaction time of 0.75s is selected to emulate real life traffic flow without deteriorating traffic stability.

In this paper, we leverage previous work in the Identity Ecosystem, a Bayesian network mathematical representation of a person's identity, to create a framework to evaluate identity protection systems. Information dynamic is considered and a protection game is formed given that the owner and the attacker both gain some level of control over the status of other PII within the dynamic Identity Ecosystem. We present a policy iteration algorithm to solve the optimal policy for the game and discuss its convergence. Finally, an evaluation and comparison of identity protection strategies is provided given that an optimal policy is used against different protection policies. This study is aimed to understand the evolutionary process of identity theft and provide a framework for evaluating different identity protection strategies and future privacy protection system.

This paper is based on the research topic of the intelligence Internet of Things(IoT) telemedicine health care for the elderly living alone system, which can be applied to the living system of the elderly living alone. The telemedicine system is mainly based on heart rate monitoring, myoelectric signal acquisition, and blood flow monitoring system. The physiological signals are taken from the patient's side and presented on a portable tablet or mobile phone. Generally, the normal physiological signals are firstly used by elderly people living alone. The body is taken down (no abnormal signal has occurred) and is defined as a normal baseline physiological signal (Base Line Data) for the patient.

Unintended Salicidation is a yield killing defect and found in 0.18um technology node which uses Cobalt salicidation. Unintended salicidation happens at salicide block area causing electrical failure, which can cause many electrical failures such as data retention issue of EEPROM structure and resistance change in poly resistor etc. It occurs during the process of wet-etching of the salicide block oxide. The photoresist is undercut and the salicide block oxide area underneath the photo resist will then be attacked by HF chemical. The mechanism of defect formation and various improvement methods will be discussed in this paper. The evaluation includes photolithography spin rinsing efficiency, post bake temperature and fab lighting optimization. In this technical paper, we studied various possible solutions and recommendation was offered for complete elimination of the defect.

A family of new non-orthogonal multiple access techniques named Scrambled Coded Multiple Access (SCMA) is described. It allows different users of a communications system to efficiently share the same bandwidth by using specific scramblers to separate user bursts. SCMA technology is designed for satellite communications with extremely small antenna terminals for potential applications such as asset tracking, sensor networking, and smart grid; and also for latency reduction. The latest variation, Asynchronous Scrambled Coded Multiple Access (A-SCMA) on the other hand, not only removes the need for timeslot synchronization and precise power control, but also offers an unprecedented system throughput of more than 4 bits/symbol at a modest per user signal to noise ratio (SNR).

The combination of Non-orthogonal Multiple Access (NOMA) and Multi-antenna Multi-input Multi-output (MIMO) approach has been considered as an assuring multiple access for fifth generation (5G) technology to improve the system network capacity and the efficiency of the spectrum. In this paper, we investigated a MIMO-NOMA based network code, exclusive-OR (XOR) scheme which applies physical layer network coding (PNC) in two-way relay networks to reduce the transmission time slots. The Quadrature Phase Shift Keying (QPSK) modulation technique is selected to apply to enable the data source transmitting with the carrier wave. In this paper, convolutional codes and the Soft Decision Viterbi algorithm is used to encode and decode the information bits. The coded MIMO-NOMA gives the less bit error rate compared to coded NOMA alone. Moreover, we have analyzed the spectral efficiency and energy efficiency in NOMA and MIMO-NOMA. The sum capacity and transmission rate are also compared in NOMA and MIMO-NOMA scheme. The simulation results show that MIMO-NOMA has higher spectral efficiency and faster transmission rate than NOMA scheme. Finally, the user fairness of MIMO-NOMA has been evaluated.

Orthogonal Frequency Division Multiplexing (OFDM) is one of the popular methods that used to transmit the data in high speed rate in cellular communication. However, OFDM has a high Peak to Average Power Ratio (PAPR) which caused by the non-linearity of the High Power Amplifier (HPA) that provides transmission power to the OFDM signal. The high PAPR caused the OFDM signal to be low in efficiency and giving a high bit error rate. Thus, in this paper, a method for PAPR reduction that caused by the HPA is proposed. Precoding technique is one of the popular way to compensate the bad impact of the HPA by reducing the PAPR of the signal. In this paper, the bad impact caused by the HPA model on the OFDM signal will be analyzed. Moreover, the PAPR of the OFDM signal by using precoding techniques will be compared with the PAPR of the OFDM signal without precoding technique. From the simulation results, it was found that the ZCT-OFDM has a better PAPR reduction compared to the DCT and WALSH technique. It was also found that the lower the order of modulation scheme is, the better the PAPR reduction performance for OFDM signal compared to higher order modulation scheme.

The common problem vespel collet; the eutectic die bonding process pick up tool; is a short lifespan. The collet tip dimension is similar to shrink chip (size

Cracked die is the occurrence of fractures in the die and it is one of the common defects in semiconductor device due to the brittle nature of the silicon. There are various factors that can cause the occurrence of cracked die, and it is most commonly originated from wafer sawing or mechanical stress from the die attach process. Due to the advancement of technology and the market demand the die size and die thickness has been continuously made into smaller and thinner, however this makes the die more prone to cracking issue. Apart from this, surface defects can be created on the wafer from the different steps of wafer thinning and the micro-cracks can be further induced through wafer sawing. As a result of this the mechanical stress induced by the assembly process will amplify the micro-cracks in silicon which eventually leads to die fracture. In order to minimize the cracked die occurrences, certain critical parameters in die attach process such as bonding force, ejector height, head landing sensor and also the collet and needle condition are regularly checked while performing machine health check, as these are some of the common parameters and machine parts that have direct contact on the die and potentially contribute to the cracked die issue. However these measurements cannot help to determine whether the machine mechanical parts movement is in good condition which can potentially cause abnormal mechanical vibration and eventually leading to the cracked die issue. In this study the cracked die occurring frequency of a few die attach machines have been compiled and some machines are observed to have higher cracked die occurrences than the others. While all the 6 machines have the critical parameters within the specification, a new approach of impact force measurement was done on all the machines to check and confirm whether there is any abnormal mechanical part movement or vibration, and the data show that there is significant co-relationship between the high and irregular impact force with the cracked die occurring frequency. Measures have been taken on the machines that have high and irregular impact force trend by changing and fine-tuning the bond head and ejector mechanical parts and as a result the impact force has been rectified into normal trend. The rectified machines have been monitored for more than one year and there is zero occurrence of cracked die. The results of the study show that the impact force trend can help to reflect the mechanical parts movement condition of the die attach machine, which can help to reduce the occurrence of machine-related cracked die issue, and also become a more scientific, quantitative and effective method for the engineers and technicians to verify the mechanical parts abnormalities in the die attach machine in the future.

In this paper, MIMO system is used to improve the performance of transmitted signal by reducing the signal fading rate. MIMO VBLAST technique is implemented to enhance the spectral efficiency and bandwidth with spatial diversity. Multiple antennas allowed the system to transmit multiple same data signal at the same time in which the data will not fade simultaneously, hence, improving the Signal to Noise Ratio (SNR) of the system. Zero Forcing (ZF) and Minimum Mean Square Error (MMSE) detection scheme are used to minimize the effect of channel estimation errors. The simulation was carried out in MATLAB software. Based on the simulation result, the proposed scheme for MIMO VBLAST has better network throughput in term of successful message delivery performance compared to the traditional scheme.

In reasoning about games, we can understand players' behaviors according to their belief, action, and preference. While modal logic can be easily used to represent and reason about agents' beliefs and knowledge if we adopt an epistemic reading of modal operators, reasoning about action requires the extension of modalities. Dynamic logic is one of the earliest attempt along this direction. The original motivation of dynamic logic is to reason about program. However, it can be applied to any structural set of actions. In this paper, we propose a graded propositional dynamic logic (gPDL) for possibilistic reasoning about regular program.

Cooperative non-orthogonal multiple access (NOMA) assisted by a dedicated relay is an effective way to improve the spectral efficiency and reliability of a cognitive radio network. Two-stage relay selection is useful in cognitive radio scenario that consists of primary user and secondary user. However, the performance of existing decode-and-forward (DF) relay selection is limited by the decoding requirements of DF relay. This paper proposes an amplify-and-forward (AF) relay selection for cooperative NOMA in cognitive radio network. A relay is selected from a group of AF relays to forward messages to a primary user and a secondary user. The spatial randomness of relays are captured in this paper by applying stochastic geometry tools. In addition, the impact of power allocation coefficient is also investigated in this paper. Simulation results demonstrate the superior performance gains of the proposed scheme over the existing scheme assisted by DF relays in terms of outage probability and ergodic rate.

In this paper wireless communications network is considered to transmit source information to destination. N number sources transmitting data to a destination receiver, through M relays nodes by employing random linear network coding (RLNC). Two methods of failure decoding probability on performance of wireless communication are explained and discussed in this paper. The major parameter that will impact on the decoding probability of RLNC are quantity of relays, a finite field size, and erasure probability. The comparison of two old bounds method (Seong's approach) and new bounds method (from Khan's approach) is explored in this research. Simulation results, shows that new bounds method on probability of decoding failure having better simulation results compared to old bound method. It is also concluded that calculating the bounds of the decoding probability is an easy method to analyze failure decoding probability compared to Chiasserini's approach method which is based on network coded matrix.

In recent years, due to the rapid development of e-commerce, personalized recommendation systems have prevailed in product marketing. However, recommendation systems rely heavily on big data, creating a difficult situation for businesses at initial stages of development. We design several methods — including a traditional classifier, heuristic scoring, and machine learning — to build a recommendation system and integrate content-based collaborative filtering for a hybrid recommendation system using Co-Clustering with Augmented Matrices (CCAM). The source, which include users' persona from action taken in the app & Facebook as well as product information derived from the web. For this particular app, more than 50% users have clicks less than 10 times in 1.5 year leading to insufficient data. Thus, we face the challenge of a cold-start problem in analyzing user information. In order to obtain sufficient purchasing records, we analyzed frequent users and used web crawlers to enhance our item-based data, resulting in F-scores from 0.756 to 0.802. Heuristic scoring greatly enhances the efficiency of our recommendation system.

In this study we provide an economic analysis of the energy consumption savings obtained by introducing research results of the European project CONVINcE for Video On Demand and Video “over the top” (OTT) services. This should allow positioning quantitatively possible savings in the end-to-end chain, including service suppliers (head-ends), networks and customers (terminals). The key performance indicator is the average consumption of energy for one hour of Video. The energy savings that could be achieved implementing the results of the project in head-end, networks and terminals parts are evaluated economically for the European market. We obtain an order of magnitude of some hundred millions Euros savings per year. It remains small at the European level. However, at the level of the service, it represents between 10 to 12% savings.

The SuperKEKB collider has started the beam commissioning this year. The BEAST II project is particularly designed to measure the beam background around the interaction point of Belle II experiment. We developed a system using BGO (Bismuth Germanium Oxide, Bi 4 (GeO 4 ) 3 ) crystals and one FPGA based DAQ board to monitor the background in real time. In this presentation, we will introduce the design of our system and some preliminary results from the BEAST II phase 1 commissioning.

To measure the variation of device Vt requires long test for conventional WAT test structures. This paper presents a framework that can efficiently and effectively obtain the mean and variance of Vt for a large number of DUTs. The proposed framework applies the model-based random forest as its core model-fitting technique to learn a model that can predict the mean and variance of Vt based on only the combined Id measured from parallel connected DUTs. The experimental results based on the SPICE simulation of a UMC 28nm technology demonstrate that the proposed model-fitting framework can achieve a more than 99% R-squared for predicting both of Vt mean and variance. Compared to conventional WAT test structures using binary search, our proposed framework can achieve 42.9X speedup in turn of the required iterations of Id measurement per DUT.

Relaying is a promising technique to improve wireless physical-layer security. Existing literature shows that a full-duplex relay can further improve the secrecy capacity and secrecy outage probability compared to conventional half-duplex relay, but this comes at a price of sophisticated implementation. For sake of easy implementation, two-path successive relaying has been proposed to emulate the full-duplex relay by scheduling a pair of half-duplex relays to assist the source transmission alternately. However, the performance of two-path successive relaying in secrecy communication remains unexplored. This paper proposes a secrecy two-path successive relaying protocol for a scenario with one source, one destination and two half-duplex relays. The relays operate alternately in a time division mode to forward messages continuously from source to destination in the presence of an eavesdropper. To further confuse the eavesdropper, the source and relay are scheduled to transmit jamming signals at appropriate intervals. Analytical results on the secrecy outage probability reveals that the proposed protocol is able to deliver the target secrecy rate when the SNR of the eavesdropping channels are high. In addition, the secrecy outage probability of the proposed protocol is the joint secrecy outage probability of the relay pair. Numerical simulations show that the proposed protocol achieves the highest ergodic secrecy capacity and lowest secrecy outage probability compared to the existing half duplex relaying, full duplex relaying and full duplex jamming schemes.