This research proposes a model for presenting email to Artificial Neural Network (ANN) to classify spam and legitimate emails. The proposed model based on selecting wise 13 fixed features relevant to spam emails combined with text features.
The experiment tests many scenarios to find out the best-suited combination of features representation. These scenarios show the effect of using term frequency (tf), term frequency-inverse document frequency (tf*idf), Level two (L2) normalization, and principal component analysis (PCA) for dimension reduction. Text features vectors are represented in the principal component space as a reduced form of the original features vectors. PCA reduction effect on ANN performance is also studied.
Among these tests, best-suited model that improves ANN classification and speeds up training is concluded and suggested. An idea of integrating ANN anti-spam filter into score-based anti-spam systems is also explained in this paper. XEAMS email gateway, the commercial anti-spam, already uses Na?ve Bayes (NB) filter as one of its many techniques to identify spam email. The proposed approach influences filtering results by 7.5% closer to XEAMS anti-spam system results than NB filter does on real-life emails of Arabic and English messages.

Because of the tremendous growth in the network based services as well as the sharing of sensitive data, the network security becomes a challenging task. The major risk in the network is the intrusion. Among various hardening system, intrusion detection system (IDS) plays a significant role in providing network security. Several traditional techniques are utilized for network security but still they lack in providing security. The major drawbacks of these network security algorithms are inaccurate classification results, increased false alarm rate, etc. to avoid these issues, an Integrated Perceptron Kernel Classifier is proposed in this work. The input raw data are preprocessed initially for the purpose of removing the noisy data as well as irrelevant data. Then the features form the preprocessed data are extracted by clustering it depending up on the Fuzzy C-Mean Clustering. Then the clustered features are extracted by employing the Density based Distance Maximization approach. After this the best features are selected using Modified Ant Colony Optimization by improving the convergence time. Finally the extracted best features are classified for identifying the network traffic as normal and abnormal by introducing an Integrated Perceptron Kernel Classifier. The performance of this framework is evaluated and compared with the existing classifiers such as SVM and PNN. The results prove the superiority of this framework with better classification accuracy.

Mobile wallet is a payment platform that stores money as a value in a digital account on mobile device which can then be used for payments with or without the need for the use credit/debit cards. The cases of cyber-attacks are on the rise, posing threats to the confidentiality, integrity and availability of information systems including the mobile wallet transactions. Due to the adverse impacts of cyber-attacks on the mobile payment service providers and the users, as well as the risks associated with the use of information systems, performing risk management becomes imperative for business organizations. This research work focuses on the assessment of the vulnerabilities associated with mobile wallet transactions and performs an empirical risk management in order to derive the security priority level needed to ensure the security and privacy of the users of mobile wallet platforms. Based on the extensive literature review, a structured questionnaire was designed and administered to the mobile wallet users who are Paga student customers via the internet. A total number of 52 respondents participated in the research and their responses were analyzed using descriptive statistics. The results of the analysis show that mobile wallet Login details are the most important part of customer information that need to be highly protected as their compromise is likely to affect others. Also, customers’ information such as Mobile Wallet Account Number, Registered Phone Number, Linked ATM Card details, and Linked ATM Card PIN among others are also plausible to attacks. Hence, different security priority levels were derived to safeguard each of the components and possible security tools and mechanisms are recommended. The study also revealed that there are vulnerabilities from the mobile wallet users end that also pose threat to the security of the payment system and customers’ transaction which need to be properly addressed. This research work will enable the mobile payment service providers focus on their services and prioritize the security solutions for each user’s information types or components base on the risks associated with their system and help in taking an inform security related decisions.

Human presence detection is a continuously sought of an issue by the scientific community. Visual camera-based technologies have emerged recently with low cost and easy usage. However, these technologies have been increased the user privacy issues. Hence it is highly essential to design a human detection system without compromising the user privacy, comfort, cost and easy deployment. The pyroelectric infrared (PIR) based sensor systems are introduced however this technology is incapable to detect the presence of stationary human because it can detect the fluctuating signals only. In this paper, we have proposed a novel infrared (IR) based sensor system to detect the human presence either mobile or immobile in targeted locations with high accuracy. The proposed infrared (IR) sensor is designed to sense the heat radiation emitted by the human body, it detects the human presence accurately in targeted locations. The proposed IR based sensor system has successfully deployed in a targeted location and tested successfully for detecting the human presence and also other objects.

The underwater acoustic environment is a promising technology which explores the real-time data collection for various applications. However, these channels are prone to errors, and characterized by propagation delay, half duplex communication. At low frequencies, the fading phenomenon extensively affect the behavior of the channel and hence the effect the design of reliable communication system. The underwater acoustic channels to perform appreciably reliable communication, an attempt are made by various modulation and coding techniques. Simulation results for the combination of BPSK modulation with Reed Solomon code (BPSK-RS) having various interleavers Random Interleaver, Matrix Interleaver, have been investigated. To improve the Bit Error Rate performance various modulation techniques such as BPSK, QPSK, and QAM were combined with coding algorithms like RS code, Turbo code and different Interleavers. The investigation of the above combination reveals that IDMA-OFDM-MIMO with BPSK modulation, Turbo code with Random Interleaver technique improves significantly Bit Error Rate performance.

The deployment of sensor nodes in underwater environment is constrained by some resources of sensor node like: energy, processing speed, cost and memory and also affected by dynamic nature of water. The main purpose of node deployment is to get the sensed data from the underwater environment. One of the major tasks is to cover the whole area in underwater and also there must be a full connectivity in the network so that each sensor nodes are able to send their data to the other sensor node. Some researchers use the concept of node mobility for better coverage and connectivity. This work proposes an efficient node deployment technique for enhancing the coverage and connectivity in underwater sensor network. Simulation results show good performance in terms of area coverage and connectivity.