https://dr.ur.ac.rw/handle/123456789/27 Research works from students of the College of Science and Technology 2026-04-13T10:52:54Z https://dr.ur.ac.rw/handle/123456789/2608 Designing smart system for smallholder farmers to monitor environmental conditions for potato crop using IoT and machine learning KIPTOO, Fancy Because of the rising urbanization and strong demand for farm products, the agriculture sector has expanded rapidly, particularly in Africa. Agriculture sector in Kenya is one of the main economic activities that provide employment roughly to two-thirds of the working population, contributing 33 percent of the country's GDP on average. Potato farming in Kenya comes as a second food crop after maize, and a source of income that helps to improve many livelihoods. However, smallholder potato farmers continue to face environmental challenges, which threaten the country's food security. In Kenya, most smallholder potato farmers use traditional methods (visual observation) to monitor the environmental parameters of potato crop, due to the high cost of monitoring tools available on the market. In addition, the available tools in the market such as rain gauge are limited to a specific use. Therefore, this study proposes the use of smart potato system to monitor environmental parameters which employ the use of IoT and ML technologies to address environmental challenges that smallholder potato farmers face which has huge negative effect on potato crop; high or low temperatures interferes with the growth of potato tubers and damage potato leaves, high humidity causes late blight, slow tuber development and high moisture leads to tuber rot, high or low soil pH results to interference of the nutrients absorption. This study developed a prototype system to monitor the environmental parameters. We deployed the prototype system in the potato field to monitor environmental parameters, and the data collection process was completed successfully. The study designed a web application for potato farmers to store environmental data collected from the field and a prediction system that utilizes machine learning (ML) algorithms to recommend a course of action when the environmental parameters under observation drop or rise above the ecological recommended threshold. This system will help farmers in better decision-making, improve efficiency, minimize resource wastage and lower production costs. Master's Dissertation 2023-12-04T00:00:00Z https://dr.ur.ac.rw/handle/123456789/2607 IoT Based sound governor MUNYANA, Raphael In Rwanda, the government of Rwanda tried to avoid the noise disturbance from public and private institutions mainly caused by bars, night clubs and religious churches by fixing soundproof, moving religious churches in the non-residential area but the outcomes were not fully successful. In this project, a feasibility study is presented on a new monitoring system in which an acoustic pattern classification algorithm running in circuit actuators is used to automatically assign the measured sound level to the different noise sources. For monitoring the variation of parameters like sound pollutionlevels from their normal levels; in this case, the actuator devices are connected to the embedded computing system and build a system called IoT-based sound governor. IoTbased Sound governor will help the substandard soundproof to monitor the level of sound in bars, churches and nightclubs according to the regulations of government. The actuator called sound cuter which is a sound compressor will be connected to the audio mixer which generates sound and make an automatic regulation of sound to be not greater than normal level and when it will get closer to 90 the sound compressor will automatically reregulate to a normal level. This case study is going to give to the concerned institutions, a sustainable solution of sound pollutionmonitoring in those different areas using Internet of things innovations. IoT-based sound governor of an efficient building requires monitoring and measures the condition in case of exceeding the established level of noise. Master's Dissertation 2023-11-01T00:00:00Z https://dr.ur.ac.rw/handle/123456789/2606 Designing an FPGA-based system-on-chip for X-ray image analysis and tuberculosis detection MWANG'ONDA, Arthur Nathaniel This thesis investigates the utilization of FPGA technology in tuberculosis (TB) detection through X-ray image analysis, focusing on designing an efficient system within the Kria KV260 FPGAbased System-on-Chip architecture. The methodology integrates hardware development emphasizing the Kria KV260 FPGA-based SoC and deep learning model training, transitioning from conventional computing to specific optimization within Vitis AI for deployment on the Kria KV260 platform. Comparative analysis assesses the FPGA-based SoC against CPU-based platforms, evaluating power efficiency, throughput, and latency. Results highlight the FPGA's superiority, notably the Kria KV260, demonstrating significantly lower power consumption during inference, emphasizing FPGAs inherent energy efficiency. Additionally, the Kria KV260 exhibits superior throughput and lower latency, crucial for efficient TB detection, underscoring FPGAs speed and responsiveness. In conclusion, the study extensively demonstrates FPGA-based architectures' prowess, particularly the Kria KV260, in enhancing energy efficiency and performance for TB detection from X-ray images. These findings emphasize substantial improvements in power consumption, throughput, and latency compared to conventional CPUbased platforms, particularly vital in healthcare applications reliant on deep learning inference. This research lays a foundation for future advancements leveraging FPGA technology for efficient and accurate TB diagnosis in healthcare settings. Master's Dissertation 2023-11-01T00:00:00Z https://dr.ur.ac.rw/handle/123456789/2605 Integration of IoT and Geospatial Techniques for enhanced nutrients distribution mapping in maize farming THOTHO, Doreen; Thotho, Doreen In the domain of agriculture, effective nutrient management practices serve as a fundamental pillar for achieving optimal crop yields while minimizing the adverse environmental impact. General fertilizer recommendations often lead to over or under fertilization and traditional soil testing methods fail to capture the spatial variability of nutrients in the field. This research aims to enhance maize farming practices through the integration of Internet of Things (IoT) and geospatial techniques for nutrient distribution mapping. It involves the design and implementation of an IoT-enabled handheld device for spatially referenced macronutrient measurements in maize fields and employing an interpolation technique, Inverse Distance Weighting (IDW) for data analysis and nutrient distribution map generation. These maps empower farmers with the invaluable insights to make informed decisions regarding the application of fertilizers allowing for the optimization of maize crop growth while minimizing resource wastage. The efficacy of this system is validated through field tests conducted on a maize farm. These nutrient distribution map has an accuracy percentage of 94.92, 92.99, 94.04 for nitrogen, phosphorus and potassium prediction for un sampled locations respectively. This study establishes the foundation for implementing more sustainable, efficient, and environmentally conscious maize farming practices by harnessing the power of IoT and geospatial integration, thereby contributing to global food security and responsible resource management. Master's Dissertation 2023-12-08T00:00:00Z