Department of Software Engineering

Permanent URI for this collectionhttps://cris.hit.ac.zw/handle/123456789/21

Browse

Filters

2012 - 20177

Settings

Start date: 2012End date: 2019

Search Results

Now showing 1 - 7 of 7
  • Thumbnail Image
    Item
    Scalable malware identification and classification using deep neural network
    (2017-01-01) Nyamugudza, Tendai; Raja, Kumaravelu
    Malware presents a challenge to organizations as they threaten smooth functioning of both physical and virtual system. Timely identification of malware is critical as it allows organizations to eliminate the threat before damage has been done. This paper proposes a scalable deep learning framework for classifying portable executable files as benign or malicious using file header information. The proposed method relies on the representational power of deep neural networks which allows them to learn complex characteristics found in the file header information. A deep neural network is trained using header information extracted from sample of benign and malicious files binaries. An accuracy of 0.98 and false positive rate of 0.019 were obtained.
  • Thumbnail Image
    Item
    Network traffic intelligence using a low interaction honeypot
    (Proceedings of International Conference On Science, Engineering and Technology, 2017-05-02) Nyamugudza, Tendai; Rajaseka, Venktesh; Sen, Prasad; Viswanathan, Madhu
    Advancements in networking technology have seen more and more devices becoming connected day by day. This has given organizations capacity to extend their networks beyond their boundaries to remote offices and remote employees. However as the network grows security becomes a major challenge since the attack surface also increases. There is need to guard the network against different types of attacks like intrusion and malware through using different tools at different networking levels. This paper describes how network intelligence can be acquired through implementing a lowinteraction honeypot which detects and track network intrusion. Honeypot allows an organization to interact and gather information about an attack earlier before it compromises the network. This process is important because it allows the organization to learn about future attacks of the same nature and allows them to develop counter measures. The paper further shows how honeypot- honey net based model for interruption detection system (IDS) can be used to get the best valuable information about the attacker and prevent unexpected harm to the networ
  • Thumbnail Image
    Item
    Multi-Processor Based Intelligent Industrial Monitoring and Control System Based on µCOS-II and Wireless Sensor Networks
    (International Journal of Science and Research (IJSR), 2014-07-01) Zvarevashe, Tinotenda; Vasumuthi, D
    This paper is based on my M Tech project which shares the same title as this paper. It presents a model which illustrates how we can incorporate a Real Time Operating System (RTOS) into an Industrial setup whereby a sensor node resides in the field where processing is carried out and a Master or Control station resides in a control room and the two communicate using a wireless protocol. The RTOS is meant to provide predictability, faster time response and high performance among other wide provisions. The RTOS used to implement this model is µCOS-II (Microcontroller Operating System) from J Labrosse. It is a pre-emptive kernel where the highest priority task in the ready queue is executed first. Thus faster processing of control commands and real time logging of data channeled by sensor node to control station is expected as a result of incorporating the operating system rather than use of a super loop.
  • Thumbnail Image
    Item
    Digital Currency:The Emergence of Bitcoins
    (International Journal of Science and Research (IJSR), 2014-06-06) Mpofu, Thabiso Peter; Masaiti, Budwell; Mukosera, Macdonald
    : Bitcoins are a crypto currency whose concept was developed in 2009 by Satoshi Nakamoto. Bitcoins are digital currencies which operate on a peer to peer system. The system is decentralized as there is no central regulatory authority as with fiat currency. For an individual to transact you need a bitcoin wallet which has one or more private and public keys associated with it. Unlike fiat currency and electronic payment methods such as Visa and MasterCard which are based on trust, Bitcoin usage is based on cryptographic proof. Bitcoin usage has been on the increase and they can be converted into fiat currency through bitcoin exchange
  • Thumbnail Image
    Item
    Designing of Android Mobile Based System Using QR Code
    (INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT, 2014-11-01) Muradzikwa, Gresham; Sarai, Noreen; Sibanda, Dumisani; Govere, Weston D.
    This paper explores a solution to create a cashless mobile payment system. The aim is to provide the most cost efficient and secure alternative to current systems. Current systems use SMS and USSD to process payments. These are not cost effective methods of communication. There is also no current method of processing credit payment on a mobile phone without the need for a specialized piece of hardware. The system is broken up into three parts, a visual QR code, Qpay Android application and a payment server. The identification of mobile phone is encoded in a QR Code allowing the built in camera on a mobile phone to scan a card. This was improved on by using a HTTPS connection between mobile phone and server. HTTPS provides an encrypted communication channel. This paper shows that a mobile phone is capable of processing QR code payments on a mobile phone. Time taken to process a payment was within an acceptable limit.
  • Thumbnail Image
    Item
    Analysis of NSL-KDD Dataset for Fuzzy Based Intrusion Detection System
    (International Journal of Science and Research (IJSR), 2012-01-08) Mukosera, Macdonald; Mpofu, Thabiso Peter; Masaiti, Budwell
    In a bid to provide useful information for intrusion detection, we focused on analyzing the NSL-KDD dataset. In this analysis, we seek to simplify the process of mining fuzzy rules by reducing the features and categorizing the dataset into various smaller clusters as smaller units of the dataset are easier to work with than the whole single large dataset. It is less complex to observe and discover sound fuzzy rules from a smaller dataset and this work serves as a foundation to a fuzzy logic based intrusion detection system. This paper presents a methodology for data preprocessing towards an intrusion detection system and Microsoft excel was used in the process.
  • Thumbnail Image
    Item
    A Survey of the Security Use Cases in Big Data
    (International Journal of Innovative Research in Computer and Communication Engineering, 2014-05-01) Zvarevashe, Kudakwashe; Mutandavari, Mainford; Gotora, Trust
    Big data is the collection of large and complex data sets that are difficult to process using on-hand database management tools or traditional data processing applications. The invention of online social networks, smart phones, fine tuning of ubiquitous computing and many other technological advancements have led to the generation of multiple petabytes of both structured, unstructured and semi-structured data. These massive data sets have lead to the birth of some distributed data processing and storage technologies like Apache Hadoop and MongoDB. To tackle the security issues in Hadoop, the Kerberos protocol has been introduced in its second edition. However, this technological movement has created some security loopholes in the processing and storage of the data sets. This paper tries to list some of the directions research on Big Data challenges has taken for the past five years together with their corresponding Use cases.