Shubi Kaijage's research while affiliated with The Nelson Mandela African Institute of Science and Technology and other places

A microstructure alcohol biosensor has been proposed to operate in the wavelength range of 0.8 to 2.0 µm for the sensing of propanol, butanol, and pentanol, unveiling impressive results of relative sensitivity and confinement loss. The results are achieved by implementing closely arranged cladding air holes of 3 rings with a single elliptical core hole for analyte infiltration. Performance evaluation of the sensor was conducted using COMSOL Multiphysics software and yields relative sensitivity of 96.75%, 89.60%, and 82.02% for propanol, butanol, and pentanol, respectively, and confinement losses of 5.49 × 10−12 dB/m for propanol, 1.98 × 10−12 dB/m for butanol, and 9.36 × 10−13 dB/m for pentanol. Other optical parameters have also been analysed that recorded effective refractive index, high power fraction, low birefringence, small effective area, and large nonlinear coefficients. The proposed biosensor is eligible for practical application in alcohol sensing with these results. Moreover, this proposed biosensor is suitable as a supercontinuum source in optical communication systems because of the high nonlinear coefficients.

Personalized learning tailors material and strategy to student requirements, interests, and goals in e-learning. These developments help educational institutions and other organizations to keep up with the fast pace of information technology, communications, and computing power. Studies show that self-adaptive learning and relevant learning information improve study efficiency. Compared to traditional teaching methods, the practice of online education is well in its infancy. On the other hand, the pedagogy and evaluation of students in online courses have a large gap that has to be filled, necessitating significant improvements in e-learning. We call this approach to education "personalized learning," which is a central focus of today's leading online education platforms. Several studies have been conducted on e-learning and personalized learning, but few investigated the development trend of personalized learning technologies and applications. Therefore this study examines the literature to close the gap and promote the development trend for personalized learning technologies and applications in higher education from 2010 to 2021 by analyzing related journal articles. The pivotal studies used inclusion criteria after a search generated 372 complete research articles and reduced them to 146 publications based on their proposed learning domains and research themes. Through carefully reviewing current trends and successes in numerous aspects of personalized learning, this discussion analyzes prospective future research directions in the field of personalized learning.

The rapid growth in mobile and wireless devices has led to an exponential demand for data traffic and exacerbated the burden on conventional wireless networks. Fifth generation (5G) and beyond networks are expected to not only accommodate this growth in data demand but also provide additional services beyond the capability of existing wireless networks while maintaining a high quality of experience (QoE) for users. The need for several orders of magnitude increases in system capacity has necessitated the use of millimetre wave (mm-wave) frequencies as well as the proliferation of low-power small cells overlaying the existing macro-cell layer. These approaches offer a potential increase in throughput in magnitudes of several gigabits per second and a reduction in transmission latency, but they also present new challenges. For example, mm-wave frequencies have higher propagation losses and a limited coverage area, thereby escalating mobility challenges such as more frequent handovers (HOs). In addition, the advent of low-power small cells with smaller footprints also causes signal fluctuations across the network, resulting in repeated HOs (ping-pong) from one small cell (SC) to another. Therefore, efficient HO management is very critical in future cellular networks since frequent HOs pose multiple threats to the quality-of-service (QoS), such as a reduction in the system throughput as well as service interruptions, which results in a poor QoE for the user. However, HO management is a significant challenge in 5G networks due to the use of mm-wave frequencies with much smaller footprints. To address these challenges, this work investigates the HO performance of 5G mm-wave networks and proposes a novel method for achieving seamless user mobility in dense networks. The proposed model is based on a double deep reinforcement learning (DDRL) algorithm. To test the performance of the model, a comparative study was made between the proposed approach and benchmark solutions, including a benchmark developed as part of this thesis. The evaluation metrics considered include system throughput, execution time, ping-pong, and the scalability of the solutions. The results reveal that the developed DDRL-based solution vastly outperforms not only conventional methods but also other machine-learning-based benchmark techniques. The main contribution of this thesis is to provide an intelligent framework for mobility management in the connected state (i.e. HO management) in 5G. Though primarily developed for mm-wave links between UEs and BSs in ultra-dense heterogeneous networks (UDHNs), the proposed framework can also be applied to sub-6 GHz frequencies.

A heptagonal core and hexagonal cladding Porous Core Photonic Crystal Fiber (PC-PCF) is proposed for terahertz wave guidance. Important transmission properties including confinement loss, effective material loss (EML), and core power fraction are investigated by using the Finite Element Method (FEM) in COMSOL Multiphysics simulation software. At an operating frequency of 1 THz and a carefully selected optimum parameter set, the proposed fiber shows confinement loss in the order of 10−4 dB/m, relative material absorption loss of 0.04 cm−1, and core power fraction of 45%. Guidance properties such as effective area, bending loss, and dispersion are also investigated in detail. Due to the simple design of circular air holes in both core and cladding, the PC-PCF is envisaged to be fabricated using the available stack and draw method. The proposed fiber is envisaged serve as an efficient waveguide for transmitting terahertz wave, particularly in communication applications due to its small loss and dispersion.

Tanzanian Government, through its national audit office in March 2020, reported the prevalence decline of two animal diseases, namely; Foot and Mouth Disease (FMD) and Contagious Bovine Pleuropneumonia (CBPP). Similarly, an increase in three animal diseases, which are African Swine Fever (ASV), Contagious CaprinePleuropneumonia (CCPP), and Lumpy Skin Disease (LSD). The national audit office mentioned inadequate animal disease surveillance system in the country was among the challenges that hinder diseases control. Therefore, this study reviews the existing animal diseases surveillance systems global and suggests measures to enhance animal diseases surveillance systems in Tanzania. This review focuses on the possibility of sharing surveillance data among livestock stakeholders (including livestock keepers) in Tanzania, considering available resources such as animal diseases existing prediction models and mobile-based surveillance systems. Also, the availability of mobile technologies such as Short Message Service (SMS), Unstructured Supplementary Service Data (USSD) andautomatic voice calls (Robocalls). Reviews synthesize the previous studies to explore strengths, opportunities, weaknesses and challenges for better future interventions through proper and timely information sharing. This study selected 46 records from the 147 identified for review. The selected records include 24 from bibliographic databases, 14 from full-text journals and other non-bibliographic databases, and 8 from the open search on websites

The study analysed the current state of the adoption of social media analytics (SMA) technology and tools to tourism small and medium enterprises (SME) in Tanzania and to understand the factors which influence the adoption of SMA tools to tourism SMEs. SMA became the basic technology in understanding and getting insight into social media data. Tourism SMEs are among the users of social media platforms. The researcher carried out the survey in Arusha and Kilimanjaro regions to determine the current state of the adoption of SMA to tourism SMEs. The technology, organization and environment (TOE) framework has been used to determine the factors that contribute to the adoption of SMA. The study finding shows that many tourism SMEs are using social media in their business activities however the adoption of SMA is very low, majority of the SMA are not adopting and using SMA. Technology, organization and environmental factors seem to allow the adoption of SMA to tourism SMEs. The study contributes to the research body of knowledge on the current status of adoption of SMA to tourism SMEs and managerial contribution on the factors to determine when tourism SMEs wants to adopt SMA.

This study delivers the results of the survey conducted to investigate the tourism SMEs' usage of social media analytics as their business intelligence tool. The study uses the unified theory of acceptance and use of technology (UTAUT) model to investigate the usage of SMA to tourism SMEs. 71 tourism SMEs were interviewed in Arusha and Kilimanjaro regions by filling the semi-structured questionnaires, then followed by the data analysis using MS excel and Python. The results of the findings show that the usage of social media analytics as a business intelligence tool for tourism SMEs is beneficial to them, however, findings show very few tourism SMEs are conducting social media analytics on their social media platforms. Most tourism SMEs use built-in social media analytics and they are performing simple metrics like counting the number of likes, comments, and shares. Further results explain that the majority of the tourism SMEs don't have more information and knowledge about social media analytics and tools used to perform social media analytics as well as they are performing social media analytics without following any implementation framework to guide the process. The study recommends that social media stakeholders increase awareness of social media analytics to tourism SMEs so that they could use it and get more advantages of using social media. Furthermore, researchers information systems analysts and developers develop social media analytics tools specifically for tourism SMEs and provide them step-by-step procedures that will help them in using and managing the social media analytics activities.

Ethanol, methanol and water are polar solvents with similar physical properties albeit contrasting chemical properties. Therefore, it is essential to provide accurate and reliable methods for detecting these liquids. In this paper, a novel liquid infiltrated photonic crystal fibre for ethanol, methanol and water sensing is introduced. The novel structure is modelled, simulated and analysed in the terahertz (THz) region using a full vectorial finite element method. It is shown that the THz light, which is guided using modified total internal reflection, is confined within the infiltrated analytes with negligible losses. For the detection of infiltrated liquids at 1.6 THz operating frequency, the proposed fibre demonstrates high sensitivities up to 99.73% and confinement losses in the order of 10 ⁻⁴ dB/m. Manufacturing of the proposed fibre is feasible using existing fabrication technologies and it is envisaged that the fibre may provide a solution to existing challenges in detecting common polar solvents.

A highly sensitive non-complex cored photonic crystal fiber sensor for hazardous chemical sensing with water, ethanol, and benzene analytes has been proposed and is numerically analyzed using a full-vector finite element method. The proposed fiber consists of a hexagonal core hole and two cladding air hole rings, operating in the lower operating wavelength of 0.8 to 2.6 µm. It has been shown that the structure has high relative sensitivity of 94.47% for water, 96.32% for ethanol and 99.63% for benzene, and low confinement losses of 7.31 × 10−9 dB/m for water, 3.70 × 10−10 dB/m ethanol and 1.76 × 10−13 dB/m benzene. It also displays a high power fraction and almost flattened chromatic dispersion. The results demonstrate the applicability of the proposed fiber design for chemical sensing applications.