Search

Machine learning (ML) techniques are part of artificial intelligence. ML involves imitating human behavior in solving different problems, such as object detection, text handwriting recognition, and image classification. Several techniques can be used in machine learning, such as Neural Networks (NN). The expansion in information technology enables researchers to collect large amounts of various data types. The challenging issue is to uncover neural network parameters suitable for object detection problems. Therefore, this paper presents a literature review of the latest proposed and developed components in neural network techniques to cope with different sizes and data types. A brief discussion is also introduced to demonstrate the different types of neural network parameters, such as activation functions, loss functions, and regularization methods. Moreover, this paper also uncovers parameter optimization methods and hyperparameters of the model, such as weight, the learning rate, and the number of iterations. From the literature, it is notable that choosing the activation function, loss function, number of neural network layers, and data size is the major factor affecting NN performance. Additionally, utilizing deep learning NN resulted in a significant improvement in model performance for a variety of issues, which became the researcher's attention.

Object classification, such as handwritten Arabic character recognition, is a computer vision application. Deep learning techniques such as convolutional neural networks (CNNs) are employed in character recognition to overcome the processing complexity with traditional methods. Usually, a CNN is followed by an activation function such as a rectified linear unit (ReLU) or leaky ReLU to filter the extracted features. Most handwritten character recognition endures an imbalanced number of positive and negative vectors. This issue decreases CNN performance when adopting ReLU and leaky ReLU for the next deep layers in the architecture. Hence, this study proposed an optimized leaky ReLU to retain more negative vectors using a CNN architecture with a batch normalization layer to address this weakness. To evaluate the proposed method, four datasets are used: Arabic Handwritten Characters Dataset (AHCD), self-collected, Modified National Institute of Standards and Technology (MNIST), and AlexU Isolated Alphabet (AIA9K). The proposed method shows significant performance in terms of accuracy, precision, and recall measures compared to the state-of-art methods. The results showed outstanding improvement over the known leaky ReLU as follows: 99% for AHCD, 95.4% for self-collected data, 90% for HIJJA dataset and 99% for Digit MNIST. The proposed CNN architecture with the proposed optimized leaky ReLU showed a stable accuracy performance and error rates between the training, validation, and testing phases. This indicates that most samples are trained and classified correctly.

Lack of secure codes implemented in the web apps will lead to cyber-attack because of vulnerabilities. The statistic shows that highest record on the data theft related cyber-attacks are through the SQL injection technique. Hence, an effective SQL injection detection is needed in any web system to combat this threat. In this research, machine learning technique is used where training is provided to the SQL injection detector using a training data and then is evaluated against a testing data. The research relies on the preparation of the training and testing datasets. Training sets are used by the detector to establish the knowledge base and the test set is used to evaluate the performance of the detector. The result of the detection shows that the proposed technique produces high accuracy in recognizing malicious and benign web requests.

This review describes a cloud-based intelligent power management system that uses analytics as a control signal and processes balance achievement pointer, and describes operator acknowledgments that must be shared quickly, accurately, and safely. The current study aims to introduce a conceptual and systematic structure with three main components: demand power (direct current (DC)-device), power mix between renewable energy (RE) and other power sources, and a cloud-based power optimization intelligent system. These methods and techniques monitor demand power (DC-device), load, and power mix between RE and other power sources. Cloud-based power optimization intelligent systems lead to an optimal power distribution solution that reduces power consumption or costs. Data has been collected from reliable sources such as Science Direct, IEEE Xplore, Scopus, Web of Science, Google Scholar, and PubMed. The overall findings of these studies are visually explained in the proposed conceptual framework through the literature that are considered to be cloud computing based on storing and running the intelligent systems of power management and mixing.

In the last two decades, the field of DNA-based steganography has emerged as a promising domain to provide security for sensitive information transmitted over an untrusted channel. DNA is strongly nominated by researchers in this field to exceed other data covering mediums like video, image, and text due to its structural characteristics. Features like enormous hiding capacity, high computational power, and the randomness of its building contents, all sustained to prove DNA supremacy. There are mainly three types of DNA-based algorithms. These are insertion, substitution, and complementary rule-based algorithms. In the last few years, a new generation of DNA-based steganography approaches has been proposed by researchers. These modern algorithms overpass the performance of the old ones either by exploiting a biological factor that exists in the DNA itself or by using a suitable technique available in another field of computer science like artificial intelligence, data structure, networking, etc. The main goal of this paper is to thoroughly analyze these modern DNA-based steganography approaches. This will be achieved by explaining their working mechanisms, stating their pros and cons, and proposing suggestions to improve these methods. Additionally, a biological background about DNA structure, the main security parameters, and classical concealing approaches will be illustrated to give a comprehensive picture of the field.

Since the number of internet users is increasing and sensitive information is exchanging continuously, data security has become a problem. Image steganography is one of the ways to exchange secret data securely using images. However, several issues need to be mitigated, especially in the imperceptibility (security) aspect, which is the process of embedding secret data in the images that can be vulnerable to attacks. This paper focuses on developing a secure method for hiding secret messages in an image, based on the standard Least Significant Bit (LSB). Before proceeding with the embedding stage, the secret message's size is reduced by compression using the Huffman algorithm, followed by two operations, which are the Boolean operation Exclusive-NOR (XNOR) operation and the Fibonacci algorithm when selecting pixels to embed the secret message. As a result of these processes, a stego-image is created with two secret keys. We obtained promising results against standard images with higher Peak Signal-to-Noise Ratio (PSNR) values of 66.6170, 65.8928, and 65.9386 dB for Lena.bmp, Baboon.bmp, and Pepper.bmp, respectively, as compared to other state-of-the-art schemes. The evaluation stage proves the increasing level of security as well as imperceptibility.

In the era of Industry 4.0 (IR 4.0), information leakage has become a critical issue for information security. The basic approach to addressing information leakage threats is to implement an information security policy (ISP) that defines the standards, boundaries, and responsibilities of users of information and technology of an organization. ISPs are one of the most commonly used methods for controlling internal user security behaviours, which include, but not limited to, computer usage ethics; organizational system usage policies; Internet and email usage policies; and the use of social media. Human error is the main security threat to information security, resulting from negligence, ignorance, and failure to adhere to organizational information security policies. Information security incidents are a problem related to human behaviour because technology is designed and operated by humans, presenting the opportunities and spaces for human error. In addition to the factor of human error as the main source of information leakage, this study aims to systematically analyse the fundamental issues of information security policy compliance. An analysis of these papers identifies and categories critical factor that effect an employee’s attitude toward compliance with ISP. The human, process, technology element and information governance should be thought as a significant scope for more efficiency of information security policy compliance and in any further extensive studies to improve on information security policy compliance. Therefore, to ensure these are properly understood, further study is needed to identity the information governance that needs to be included in organizations and current best practices for developing an information security policy compliance within organizations.

From the digital forensics point of view, image forgery is considered as evidence that could provide a major breakthrough in the investigation process. Additionally, the development of storage device technologies has increased storage space significantly. Thus a digital investigator can be overwhelmed by the amount of data on storage devices that needs to be analysed. In this paper, we propose a model for classifying bulk JPEG images produced by the data carving process or other means into three different classes to solve the problem of identifying forgery quickly and effectively. The first class is JPEG images that contain errors or corrupted data, the second class is JPEG images that contain forged regions, and the third is JPEG images that have no signs of corruption or forgery. To test the proposed model, some experiments were conducted on our own dataset in addition to CASIA V2 image forgery dataset. The experiments covered different types of forgery technique. The results yielded around 88% accuracy rate in the classification process using five different machine learning methods on CASIA V2 dataset. It can be concluded that the proposed model can help investigators to automatically classify JPEG images, which reduce the time needed in the overall digital investigation process.

There are various approaches to detect malware. Among them is via dynamic analysis which is a very essential technique capable of detecting unknown malware. The dynamic analysis monitors the behaviour of the executable by providing its execution behaviour information. Since the complexity of the malware is increasing, it is important to monitor the malware and study how malware behaves to help in detecting it. In this paper, we highlighted the instrumentation technique to observe behaviour of Portable Executable execution. We briefly explored some of the related works. We discussed about dynamic analysis and Windows API Calls. We discussed on our realtime behaviour monitor. The concept of n-gram was explained and before concluding, several challenges were highlighted.

The demand of cloud robotics makes data encryp-tion essential for peer robot communications. Certain types of data such as odometry, action controller and perception data need to be secured to prevent attacks. However, the introduction of data encryption caused increment of overhead for data stream communication. This paper presents an evaluation of CryptoROS architecture on Robot Operating System (ROS) which focused on peer-to-peer conversations between nodes with confidentiality and integrity violation. OpenSSL is used to create a private key and generate a Certificate Signing Request (CSR) that contains public key and a signature. The CSR is submitted to a Certificate Authority (CA) to chain the root CA certificate and encryption of RSA private key with AES-256 and a passphrase. The protected private key are securely backed up, transported, and stored. Experiments were carried out multiple times with and without the proposed protocol intervention to assess the performance impact of the Manager. The results for different number of messages transmitted each time increased from 100, 250 to 500 with performance impact 1.7%, 0.5% and 0.2%, respectively. It is concluded that CryptoROS capable of protecting messages and service requests from unauthorized intentional alteration with authenticity verification in all components.

Malware sophistication is on the rise and continue being a serious threat against privacy, availability and integrity of information. This paper proposed an architecture which consist of two detectors. The first detector observes for malware in general. It will classify an executable if whether it is benign or malign and updates its Knowledge Base dynamically. The second detector is a specialised one, aiming to detect ransomware via its deep scan capability. This detector will be activated if the first detector senses the presence of ransomware pre-execution activities. The detector will decide whether a program is a ransomware or not and dynamically update its Knowledge Base dynamically.

Text steganography is regarded as the most challenging type to hide secret data to a text file because it is not enough unnecessary information compared with other carrier files. This study aimed to deal with the capacity (how much data can be hidden in the cover carrier) and security (the inability of disclosing the data by an unauthorized party) issues of text steganography. Generally, the data hiding capacity of text steganography is limited, and imperceptibility is very poor. Therefore, a new scheme was suggested to improve the two-letter word technique by using the Lempel-Ziv-Welch algorithm. This scheme can hide 4 bits in each position of a two-letter word in the cover text by inserting a nonprinting Unicode symbol. Each two-letter word can have four different locations in the text. Some experiments were conducted on the proposed method (enhancement) and our previous method (two-letter word) to compare their performance applying twelve secret message samples in terms of capacity and Jaro-Winkler. On the other hand, the performance of our proposed method was compared with other related studies in terms of capacity. The results show that the proposed method not only had a high embedding capacity but also reduced the growing size ratio between the original cover and stego cover. In addition, the security of the proposed approach was improved through improving imperceptibility and by using stego-key.

Implementing cryptography on Q-Geographic Information System (QGIS) software is a crucial step to provide the software with an extra layer of protection on digital mapping against duplication from outsiders. With the growth of advancement in technology, copyright can be easily violated by irresponsible people to gain profit illegally. Therefore, this study implements Advance Encryption Standard (AES) algorithm into digital mapping in order to prevent duplication from happening. This is performed by encrypting the original map (called shapefile), and allowing the users to save only the additional information that they added on that particular file. Hence, if the users want to sell the duplication map with some additional information, they have to consult with the owner.

Phishing is an attempt to obtain private/confidential information such as usernames, passwords, and financial details. It is often for malicious reasons by disguising as a trustworthy entity in an electronic communication such as email. The chances of obtaining confidential or personal information are higher when website medium combined with email medium in launching phishing attacks. Universiti Kebangsaan Malaysia (UKM) has experienced phishing emails attacks in 2016. Besides technology that focuses on email security, the safety awareness program that meant to provide education to the users especially UKM staffs needs to be enhanced to reduce the risk of thievery on personal data, university confidential information and research data. The simulation approach in a real environment can provide a true picture to the staffs about the serious impact of phishing attacks. The objectives of the simulation are to measure and to educate UKM staffs on the security awareness. We designed a spear phishing simulation procedure with collaboration between the Faculty of Information Science and Technology (FTSM), Information Technology Center, Bursary Department and Department of Registrar, UKM. The simulation was conducted from 11–13 January 2017 with 553 email addresses were identified from five different faculties. There were 209 respondents (38%) who have entered their official ids (captured) and password (not captured). The differences in the number of respondents between science and technology (S&T) faculties and non-S&T faculties indicated the security awareness is in the worrying level. A high percentage of responses among the management and professional group can also be classified as being in an alarming rate. This simulation is the first practice in UKM and it helps to increase awareness and provide education about cyber security.

In a learning environment, a low student-lecturer ratio is considered a practical solution by many educational institutions. However, the number of students in information technology is increasing every year. This could lead to a significant increase in the workload of lecturers, who need to evaluate assignments, quizzes and projects. Hence, it is desirable that an automated assessment tool is used to lessen their workload. In the era where mobile devices are getting popularity, the high demand to execute suitable quality code is there and the cost is on the processing power of the CPU which has a direct implication on the power source or the battery used. With various implementations of cryptography algorithms available, many of them could satisfy different level of needs. In this research, we introduce the architecture for a multi-layered security of automated assessment of programming code. First, we review the existing research studies in the area. We describe the features of the tool, as part of a complex e-learning environment. We also discuss the implementation of security, to protect data transmission and storage used by the tool. Challenges the system might face and the potential solutions, are also described.

The Trusted Computing Group (TCG) has introduced the Trusted Platform Module (TPM) as a solution to assure end users of their privacy and confidentiality. Although the TPM is designed to prevent software attacks, the TPM itself is vulnerable to physical attacks that could enable intruders to gain access to confidential data. In general, the TPM provides an ID and implements a password identification technique to prevent unauthorized users from gaining access to the TPM. The TPM user authentication is carried out by the TPM itself, which exposes the TPM to direct risk as highly skilled intruders can break the authentication line of defence and gain access to the TPM. The process of encrypting and decrypting information, especially when asymmetric algorithms are used, is viewed as a process that consumes time and resources, which decreases the speed of the computer. In order to solve the problems, a TPM User Authentication Model (TPM-UAM) that can provide the TPM with a higher level of security and resistance against physical attacks has been proposed as we proposed in our previous research paper (Alshar’e et al., 2014). The technique is based on biometric authentication to prove the identity of the users and to allow the process of authentication to happen at an independent platform using virtualization that will keep the TPM out of reach until a user is completely verified and approved. The TPM-UAM is able to provide a more satisfactory level of confidence for data and processes that can be rated as highly confidential and private. The model was successfully developed and tested and the results confirmed the model efficiency and ability to secure TPM and all functions have been confirmed to be working perfectly according to what they were designed for. This paper describes the design and implementation of TPM-UAM system based on the proposed authentication model, virtualization has been implemented to create authentication platform to prevent direct interaction with TPM and biometrics has been implemented to verify identities and supervise running TPM, the system testing results in confirming the system functionality and ability to secure and protect TPM.

Information security presents a huge challenge for both individuals and organizations. The Trusted Computing Group (TCG) has introduced the Trusted Platform Module (TPM) as a solution to end-users to ensure their privacy and confidentiality. TPM has the role of being the root of trust for systems and users by providing protected storage that is accessible only within TPM and thus, protects computers against unwanted access. TPM is designed to prevent software attacks with minimal consideration being given toward physical attacks. Therefore, TPM focus on PIN password identification to control the physical presence of a user. The PIN Password method is not the ideal user verification method. Evil Maid is one of the attacking methods where a piece of code can be loaded and hidden in the boot loader before loading TPM. The code will then collects confidential information at the next boot and store it or send it to attackers via the network. In order to solve this problem, a number of solutions have been proposed. However, most of these solutions does not provide sufficient level of protection to TPM. In this study we introduce the TPM User Authentication Model (TPM-UAM) that could assist in protecting TPM against physical attack and thus increase the security of the computer system. The proposed model has been evaluated through a focus group discussion consisting of a number of experts. The expert panel has confirmed that the proposed model is sufficient to provide expected level of protection to the TPM and to assist in preventing physical attack against TPM.

In this study we propose a security architecture for mobile Health (mHealth) communications. mHealth is a term used for medical-related services or communication, delivered using mobile devices such as mobile phones, tablet computers and PDAs. Communication in the health-related field often involve sensitive information (e.g., an email about a patient's illness is sent between doctors), which is transmitted over the Internet. However, although the Internet greatly facilitates the communication, it is undeniable that the threats to the Internet are becoming more prevalent. A multi-agent security architecture is presented in this study to provide a secure environment for mHealth communication. Agents are skilled in order to handle the communication processes at both sender's and recipient's side. This includes the security processes, which make use of cryptography protocols to secure data at both sides.

Problem statement: In this study, we address the security needs in on line communications, specifically in the e-health domain. We focus on ho w to provide different security strengths to differ ent types of communications in e-health, where each communication transmits different types of information with different levels of sensitivity. Approach: The Multi-Agent System (MAS) approach is used to develop an agent-based system that can cater for di stributed processes. We use the agents' characteris tics such as autonomous , interactive , extendible and mobile to handle the security processes for users in different environments and devices. We integrate di fferent types of encryption algorithms with differe nt security strengths in order to provide different se curity needs. Results: We present our security model called MAgSeM that consists of eight agents, which are skilled to complete its goal as well as the ove rall system goals autonomously. Conclusion: We conclude that MAgSeM security model is suitable not only for the e-health domain, but also other domains tha t practices online communications.

Problem statement: Delivering services online is important in e-health. Services that are delivered through online communications between engaging parties, often involve sensitive information transmitted over the Internet. However, while the Internet successfully facilitates these services, significant threats also come in parallel. Network attacks, information breaches and malicious software on a computer system are common threats to the Internet. These threats can cause severe damage to computer systems and also the information. As we study current security technologies particularly that provide security to online communications, we found out that these technologies do not cater for different kinds of security needs because of the rigid way the security mechanisms are constructed. Therefore, we are interested in developing a security model that facilitates these needs, specifically in e-health. Approach: First, the area where different security requirements are needed are explored, such as the information classification found in ISO17799. This classification is based on the sensitivity levels of the information, where the more sensitive information requires higher security measures compared to the less sensitive information. Then, the information classification is applied to the e-health environment, so that our security model can handle the security processes for each classification. Results: The multilayer communication approach or MLC is the proposed security model. MLC classifies communications in e-health into five categories: Layer 1 to Layer 5 representing extremely sensitive, highly sensitive, medium sensitive, low sensitive and no sensitive data. This classification refers to the different sensitivity of the information exchanged during communications. For example, Extremely Sensitive communication involves exchanging extremely sensitive information, which requires highest security mechanisms, while Low Sensitive communication requires lower security mechanism. Conclusion: MLC provides five different types of security needs, where users can flexibly choose their own security preferences for their online communications, which the current technologies are lacking.

A web service is defined as the method of communication between the web applications and the clients. Web services are very flexible and scalable as they are independent of both the hardware and software infrastructure. The lack of security protection offered by web services creates a gap which attackers can make use of. Web services are offered on the HyperText Transfer Protocol (HTTP) with Simple Object Access Protocol (SOAP) as the underlying infrastructure. Web services rely heavily on the Extended Mark-up Language (XML). Hence, web services are most vulnerable to attacks which use XML as the attack parameter. Recently, a new type of XML-based Denial-of-Service (XDoS) attacks has surfaced, which targets the web services. The purpose of these attacks is to consume the system resources by sending SOAP requests that contain malicious XML content. Unfortunately, these malicious requests go undetected underneath the network or transportation layers of the Transfer Control Protocol/Internet Protocol (TCP/IP), as they appear to be legitimate packets.In this paper, a middleware tool is proposed to provide real time detection and prevention of XDoS and HTTP flooding attacks in web service. This tool focuses on the attacks on the two layers of the Open System Interconnection (OSI) model, which are to detect and prevent XDoS attacks on the application layer and prevent flooding attacks at the Network layer.The rule-based approach is used to classify requests either as normal or malicious,in order to detect the XDoS attacks. The experimental results from the middleware tool have demonstrated that the rule-based technique has efficiently detected and prevented theattacks of XDoS and HTTP flooding attacks such as the oversized payload, coercive parsing and XML external entities close to real-time such as 0.006s over the web services. The middleware tool provides close to 100% service availability to normal request, hence protecting the web service against the attacks of XDoS and distributed XDoS (DXDoS).\

Users are concerned about the protection of personal information they share with mobile applications. Researchers have previously explored security threats to mobile applications through wireless network access, including the disclosure of personal information through unencrypted traffic, excessive information disclosure to service providers, and flaws in TLS security. This study replicates these security threats and performs an assessment of the potential privacy impact for a sample of 30 Android applications. The results show that disclosure of personal information through unencrypted traffic is a significant risk. Individual applications were found which disclosed a user's identity and application usage, and persistent device identifiers were leaked allowing user information to be linked across applications and wireless sessions. A small number of applications disclosed inappropriate amounts of personal information to service providers which could allow user tracking. TLS issues continue to pose a risk, with one application exhibiting a previously identified TLS certificate validation issue, and a potentially new encryption protocol downgrade flaw was identified triggered by an invalid certificate. Insecure authentication techniques were used by 30% of applications tested and pose a privacy risk even when applications use TLS.

This paper established a new security mechanism for online communications using mobile agents. E-health has been widely used as a communication system and information changes to deliver health services to the users through the Internet. However, the Internet imposes threats that will risk the information in transit. Therefore it is important to ensure that the communication is safe and the user’s privacy is protected. This paper extends our previous research results, Multilayer Communication (MLC) approach [1], and introduced mobile agents to MLC in e-health. The cryptographic protocols such as encryption/decryption, digital signature, and hash code are used as protection mechanisms. We focus on how Sender can securely transfer sensitive information to Recipient while still maintaining control over it. Sender keeps the key for decryption at his/her side until the agent needs it. A token is carried by the agent, which is used as a mean to call home platform to obtain the key for decryption processes. The significant of this new mechanism is that it gives Sender control over the transmitted data. Recipient can also avoid burden with the encryption details as long as he/she knows that the message is indeed comes from Recipient.

This paper presents a multi-agent security architecture, which utilizes the agent characteristics to cater for security processes in online communications. The Multilayer Communication approach (MLC) [1] is used to determine the security processes, which uses cryptography protocols to secure data and communication channel. Agents are skilled to perform certain tasks. At the Sender’s host, agents interact with each other to secure a message to be sent to the Recipient, including encryption, digital signature, and hash code. A mobile agent is used to carry the encrypted messages as well as the agent’s code to the Recipient’s host. Our approach also provides mechanisms to verify the authenticity, confidentiality and the integrity of the code and data that arrived at the Recipient’s host. The message and the code are authenticated, the code is executed to perform tasks to recover the plaintext.

Nowadays, information and communication technologies are widely used in e-health to deliver services. The most popular communication choice is the Internet, as it offers cheap and worldwide coverage. However, as the technologies grow, the threats also grow in parallel. The threats put the patient's privacy at risk. Therefore, it is important to make sure that security technologies can cater for the privacy and security needs, whenever communications occur through the Internet. This paper presents a multi-agent based security framework for e-health services. The actors and the type of communications are described and a multi-level communications for the users is introduced. The security requirements and the multi-agent based security approach are modeled and presented in the framework.

Nowadays, auction marketplace are conducted online and executed via electronic channels. Taking one step further, online auction e-marketplace adopted software agent technology in the bidding process. A mobile agent-based e-auction marketplace must fit the concept of halal trade and following Shari’ah law in order to fulfill muslim community needs and must offer secure and trustworthy trading environment. Nevertheless, mobile agent raises issues concerning the protection of platform/host and malicious attacks from a variety of intervening that might alter the information its carries. This paper investigate the requirement of trusted and secure online auction in line with Shari’ah transaction rules using the mobile agent technology by proposing a security architecture for mobile-agent based e-auction marketplace. This paper also investigates possible online auction frauds that can affect bidding process and focused on bid shielding and shill bidding. At the end of this the paper, security architecture for mobile agent-based e-auction marketplace in Shari’ah compliant environment is presented.

Data carving is a very important topic in digital investigation and computer forensic. Researches are needed to focus on improving data carving techniques to enable digital investigators to retrieve important data and evidences from damaged or corrupted data resources. This paper is the result of a systematic literature review which answer three main questions in data carving filed. The Results fall into four main directions. First it shows the need of realistic data sets for tools testing. Secondly, it points to the need of object validation under fragmented data storage. Thirdly, investigating content based validation and its benefits in digital investigation field. Finally, it points to a new direction for using semantic validation to reduce false positive rates.

Augmented reality is a human-machine interaction tool that presents information generated by computer on the real world using a camera. Augmented reality technology has the potential to draw students’ attention to visualize a layer of information on real objects using handheld devices such as tablets and smartphones. The objective of this study was to assess students’ perception of the use of augmented reality in learning microprocessor. Therefore, a mobile application was designed and developed by integrating elements of augmented reality for teaching and learning this topic. This study was conducted based on the ADDIE model that consists of five phases: Analysis, Design, Development, Implementation and Evaluation. The study found that students’ perception of the use of augmented reality is positive based on the average mean score. However, the disadvantage of this application is the use of images as compared to the use of real microprocessor as the learning object. An area for future study is to assess the students’ perception of the use of real objects as the learning object in using mobile augmented reality application.