Browsing by Author "Dr. Muhammad Farooq-i-Azam"

Now showing 1 - 3 of 3

Handoff Management in Light Fidelity Networks
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2023) Waseem Ahmad,; SP21-RCS-005; Dr. Muhammad Farooq-i-Azam; LHR TP 8057
Handoff Management in Light Fidelity Networks Many technologies have been considered to handle the speedy developing range of cell phones and their extended requests for service providers. Among those technologies, Li-Fi is the one that can provide better communication and cover the spectrum shortage. However, Li-Fi has a short range and different service outages. Therefore, Li-Fi is integrated with Wi-Fi to provide better communication in the form of a hybrid Wi-Fi Li-Fi system. This thesis investigates handoff in hybrid light fidelity (Li-Fi) and wireless fidelity (Wi-Fi) networks, i.e., HLWNets. The hybrid network has more challenges to overcome the issue of the small spectrum regions of Li-Fi access points and the interference issue of two different networks. Li-Fi has a short coverage range, while Wi-Fi covers a great area. Due to this, to provide reliable coverage in HLWNets, we integrated Li-Fi networks with Wi-Fi networks. In HLWNets, we face the issue of frequent handoffs. Different methods have been introduced to resolve this issue. In this research, our proposed method is presented based on a signal-to-interference ratio (SINR) scheme instead of reference signal received power (RSRP) because it is a standard scheme but requires additional signals during transmission. The simulation results show that the proposed technique reduces frequent handoffs between Wi-Fi and Li-Fi. During the proposed method, we introduced the AWGN channel while transmitting data from the transmitter to provide reliable data on the receiver. This method also offers better throughput than a standalone Li-Fi system. We have developed a simulation network with handoff between fully working access points of hybrid Li-Fi and Wi-Fi. The simulation results work on the throughput and diminish the packet delay and drop ratio. The algorithm used for load balancing is based on evolutionary game theory for hybrid Li-Fi/Wi-Fi networks. The proposed greedy algorithm also overcomes the frequent handoffs in a hybrid Li-Fi/Wi-Fi network and provides rapid information transmission that improves the quality of service (QoS). We used vertical handoff and horizontal handoff based on the threshold. The greedy algorithm is used to decrease the data loss and improve throughput. The user is connected to a Li-Fi access point (AP) mostly. When light intensity decreases, we communicate to other AP based on the threshold value. If no high-intensity Li-Fi AP is available, the user is associated with Wi-Fi AP. The RSSI method moves the user from VIII Wi-Fi AP to Li-Fi AP. This process decreases the data loss rate and improves throughput and fast data transmission.
Landslide Monitoring using Solar Energy Powered Wireless Sensor Network and Machine Learning Techniques
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2024) Malka Arooj; FA22-REE-003; Dr. Muhammad Farooq-i-Azam; LHR TP 9372
This research examines the case of landslide monitoring and how energy-autonomous Wireless Sensor Networks (WSN) integrated with Machine Learning (ML) technologies enhanced achievement of implementation attributes of accuracy and efficiency gains. These include the ability of constantly generating power through solar energy, and the ability of a ML algorithm to predict since this is the basic building block of this more encompassing architecture. Hence, the main objective is to explore the traditional and the new models of landslide monitoring, their problems that might be encountered and the ways of early recognition and prediction. In the context of the study, a high-level self-operating autonomous system is proposed to for efficient monitoring of landslides using solar energy powered WSNs and ML techniques. In proposed approach passive networking technique is used to enhance the operation and longevity of the nodes. Passive networks have led to a limited energy consumption which has a figure of 69.76% as compared to the 100 % that is expected with traditional methodologies. Different types of ML models were employed to obtain more realistic outcomes, and Gradient Boosting Machines (GBM) was more effective than the other ML models used in this thesis. To enhance the results of accuracy, precision, recall, F1 score and Matthews Correlation Coefficient (MCC) a series of techniques like ensemble learning, feature scaling, hyper parameter tuning were applied on the GBM. However, it was only the Chi-square that significantly enhanced the findings; the percentage stood at 86. 42% accuracy, 89. 23% precision, 85. 93% recall, F1 score of 87.55% and MCC value of 72.69%. In further development of the integrated defining methodologies, more attempts at ML models along with optimization strategies should be researched. In enhancing the practical use of the system this proposed model needs to take place in varied geographical and environmental contexts to increase its robustness as well as its applications.
Landslide Monitoring Using State of the Art Machine Learning Techniques with Image Processing of Space Borne Remote Sensing Imagery Data
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2024) Mohammad Mazan; FA21-REE-003; Dr. Muhammad Farooq-i-Azam; LHR TP 9366
Landslides present significant worldwide difficulties, leading to profound repercussions such as property destruction and human death. As a result, the methods we currently have, including light detection and ranging (LiDAR) and on-site assessments by experts to look at things like topography, ground fractures, slope stability, and changes in vegetation, are not good enough to accurately predict and prevent these events. The rising frequency of landslides worldwide, intensified by alterations in weather patterns and human actions, underscores the necessity for sophisticated alert and surveillance tools. Landslides can result in devastating outcomes, such as fatalities, displacement of individuals, and significant economic repercussions. Technological advancements, namely in machine learning, remote sensing, and image processing, provide promising options to improve landslide prediction and mitigation efforts. This work aims to address the critical issue of evaluating and predicting landslides, with the objective of creating a dependable approach. We extract landslide causative factors from remote sensing imagery data through image processing, compile them into a single dataset, and then use this dataset to train machine learning algorithms. The landslide causative factors included in our dataset are the Topographic Wetness Index (TWI), Peak Ground Acceleration (PGA), Stream Power Index (SPI), Terrain Ruggedness Index (TRI), Normalized Difference Vegetation Index (NDVI), curvature, elevation, faults, rainfall, geology, solar radiation, aspect, landcover, slope, streams, and roads. We tested how well different machine learning algorithms could predict the risk of landslides. These included Logistic Regression (LR), Random Forest (RF), XGBoost (XGB), AdaBoost, Long Short-Term Memory (LSTM) networks, Deep Neural Networks (DNN), NGboost, and 1D Convolutional Neural Networks 1D (CNN 1D). The results demonstrate significant accuracy scores for each model, with the LSTM network emerging as the most efficient, showcasing an accuracy of 0.91 and an amazing Area Under the Curve (AUC) score of 94%. The success of the LSTM can be due to its proficiency in managing sequential and time-dependent data, which is essential for comprehending and predicting the dynamic characteristics of landslides. The LSTM’s ability to utilize temporal patterns improves the effectiveness of early warning systems, representing a notable progress in landslide prediction approaches.