MSIT Theses and Dissertations (2017)

Permanent URI for this collection

Browse

Recent Submissions

Now showing 1 - 5 of 34
  • Item
    A Sandboxing based security model to contain malicious traffic in smart homes
    (Strathmore University, 2017) Nkinyili, Tiberius Tabulu
    The Internet of Things (lOT) is a developing Next Generation Network (NGN) paradigm that aims to have more devices connected to the Internet and the possibility of these devices to autonomously communicate with each other. These devices mainly use wireless links to communicate, with little or no flow control, error checking or security monitoring. While this helps support mobility and optimize performance, the compromise in flow control and security monitoring, renders them more vulnerable to potential attacks from malicious users. This poses security threats to data exchanged between devices especially in a smart home environment. This necessitates having mechanisms to provide security against malicious messages and unauthorized modification of information to limit potential attacks on integrity and confidentiality of data. Isolation mechanisms would be ideal to cushion devices and the entire lOT network. Sandboxing involves isolating suspect data, processes, applications or devices from the rest of the system. This restricts access to more system resources hence ensuring continuity and availability of the entire system. This research work thus proposed a model to ensure comprehensive data security in a smart home by using sandboxing. The model proposed mechanisms to provide an isolating environment to contain malicious traffic by evaluating levels of authorization, and restricting communication nodes to what they were allowed to. This thus ensured a proactive data security approach in lOT networks within a smart home environment. Linux security Module implementations were used to provide a custom sandbox from the Kernel level. Instant Contiki, a virtual version of the lOT operating system Contiki, was used to emulate lOT communication with Cooja as the emulating module.
  • Item
    Multi-sensor fire detection system using an Arduino Uno microcontroller
    (Strathmore University, 2017) Obanda, Zephaniah Shiwalo
    Untimely response, constrained navigation due to poor urban planning and traffic jams, highly flammable construction materials, insufficient capacity by the fire department and lack of access to automated fire detection systems by residents due to purchasing costs are among the factors that affect fire-fighting services in Kenya and across the African continent. The aftermath of a fire outbreak could very acute leading to widespread loss of property and loss of lives. Residential areas contain numerous flammable materials such as clothing, books, wooden cabinets, beddings and plastics while also housing sources of ignition that include cooking gas and electronic devices thus are prone to severe fire accidents. Fire outbreaks have an inception period of about 3 to 5 minutes which is the optimal time to detect it and put it out after which it might get out of control.This implies that timely identification of a potential fire outbreak is crucial to managing it.Currently, most residential establishments as well as business premises are not fitted with fire detection systems owing to lack of awareness, high purchasing costs and inefficiency of the devices given the high false alarm rates which have a cost attached to them such as the unnecessary deployment of fire-fighting personnel. The fire detection devices are highly susceptible to false alarms because reliance on one sensor that reads only one percept from the environment for instance smoke or heat. However, the advancement of the Internet-of-Things has led to the development of ‘smart’ technologies where multiple sensors can be incorporated into objects like fire detectors additionally enabling them to communicate wirelessly with other objects and carry out programmed tasks. This research aimed at proposing a prototype of a fire detection system using a multi-sensor approach. This research applied rapid prototyping methodology for development of the prototype. Data was collected from secondary sources and experimentation.The prototype used an MQ2 gas sensor, a Grove temperature sensor, a Grove light sensor and an Arduino microcontroller, a GSM and GPS shield. In the event of a fire outbreak, the device will be able to send an SMS alert to the home owner as well as the firefighting department with GPS coordinates of the residence. The prototype recorded 83% success rate and 17% false alarm rate based on 6 test cases of which only one failed.
  • Item
    A Model for early detection of potato late blight disease: a case Study in Nakuru County
    (Strathmore University, 2017) Toroitich, Patrick Kiplimo
    The agricultural sector has been a key backbone to Kenya’s economy. Agriculture has played a key role in the economy through agricultural farm produce exports and job creation hence improving and maintaining good farming practices is critical in ensuring agricultural yields. Potato (Solanum tuberosum L.) is a major food and cash crop for the country, widely grown by small-scale farmers in the Kenyan highlands. However, early detection of potato diseases such as potato late blight still remains a challenge for both farmers and agricultural extension officers.Consequently agricultural extension officers who play a critical role in training and creating awareness on sound agricultural practices are few and often lack sufficient knowledge and tools.Current techniques used for determining and detecting of crop diseases have heavily relied upon use human vision systems that try to examine physical and phenotypic characteristics such as leaf and stem color. This technique is indeed important for diagnosis of crop diseases, however the use of this technique is not efficient in supporting early detection of crop diseases. This study proposed use of sensors and back propagation algorithm for the prediction of potato late blight disease. Temperature and humidity sensor probes placed on the potato farms were instrumental in monitoring conditions for potato late blight disease. These parameters constituted abiotic factors that favor the development and growth of Phytophthora infestants. Back propagation neural network model was suitable for the prediction of potato late blight disease. In designing the potato late blight prediction model, historical weather data, potato variety tolerance on late blight disease was used to build an artificial neural network disease prediction model.Incoming data streams from the sensors was used to determine level and risk of blight. This study focused on a moderate susceptible cultivator of potato in developing the model. The algorithm was preferred due to its strengths in adaptive learning. The developed model achieved an accuracy of 93.89% while the precision obtained was 0.949. The recall ratio from the neural network was 0.968 and an F-measure of 0.964.
  • Item
    A HIV/AIDS viral load prediction system using artificial neural networks
    (Strathmore University, 2017) Tunduny, Titus Kipkosgei
    Human Immunodeficiency Virus (HIV) has been affecting people since it was first discovered in 1986. This is as a result of the HIV virus being present in the patient bloodstream for the remainder of their normal life, as there is no cure that exists as of now. HIV, if left unmanaged would end up developing into Acquired Immune Deficiency Syndrome (AIDS), a syndrome that weakens a patient’s immune system and leaves them susceptible to other opportunistic infections. Antiretroviral therapy (ART) has been successfully used in managing the progression of the HIV virus in the human body. However, poor adherence attributable to ignorance, adverse drug effects, and age have derailed the attainment of viral load suppression amongst the HIV positive people. The progression of the virus is tracked by counting Cluster of Differentiation 4 positive cells, and the amount of virus in the blood (viral load) every 6 months. This research introduces the use of multi-layer artificial neural networks with backpropagation to predict the HIV/AIDS viral load levels over a given period of time (in weeks). The Data-driven Modelling methodology was used in the development of the model. This methodology was ideal since the model relied solely on pre-existing data, and supports artificial neural networks. The model developed performed at an accuracy level of 93.76% and a mean square error of 0.0323. The results showed that the neural network can be used as a suitable algorithm for HIV/AIDS viral load level prediction. The learning rate used in the study was 0.005 and the momentum was 0.9. The iterations for the training, testing and validation varied.
  • Item
    A Mobile application for HIV education and stigma level measure: a case of Nairobi
    (Strathmore University, 2017) Thumbi, Cameline Mukami
    The world wants to reduce HIV and AIDS spread by the year 2030 as part of the sustainable development goals (SGDs). HIV and AIDS has become the world’s most devastating epidemic especially in developing countries like Kenya. Many people have died because of HIV and AIDS related illnesses since it was first reported in Kenya in 1984. To be able to achieve the sustainable development goals, HIV education and stigma reduction would be essential. The problem being studied is people living with HIV not being able to access service due to discrimination and stigma. HIV education and awareness programs have been implemented by various governmental and private bodies for which have improved the HIV prevalence rate. While the prevalence rate has improved, more needs to be done to advance access to the right information and reduce the stigma associated with HIV. The general objective of this research is to develop a mobile based application for PLHIV that will assist in information access as a means of education. The stigma is expected to go down by using anti-stigma messages and tagging of the same. The secondary objective is to reduce stigma associated with HIV through education on HIV and anti –stigma messages geo tagging. Among the main reasons for not seeking treatment include discrimination, stigma, wrong information, lack of support and drugs. Data collection involved both primary and secondary methods which included use of questionnaire, observation and literature review. The data was analysed using thematic analysis. This research sought to find how best to increase access to the right information as a means of education on HIV through use of the mobile phone. The software development life cycle was used for the development of the application. An android based mobile application was developed as a proof of concept for access to information on HIV care and geo tagging of anti-stigma messages.