@article{CRSIJ26000133,
                  author = {Obieze O. E., Onuigbo C. M., Omeche Akaemeuwa Ambrose},
                  title = {An IoT-Driven Smart Surveillance Framework for Security Enhancement in South-East Nigeria Using Wireless Sensor Networks},
                  journal = {Cosmo Research and Science International Journal},
                  year = {2025},
                  volume = {1},
                  number = {6},
                  pages = {587-603},
                  issn = {3108-1584},
                  url = {https://cosmorsij.com/published/CRSIJ26000133.pdf},
                  abstract = {Insecurity in South-East Nigeria continues to demand intelligent, proactive, and scalable surveillance solutions capable of addressing dynamic threats. This study presents the design and performance evaluation of an Internet of Things (IoT) and Wireless Sensor Network (WSN)-based hybrid security system for real-time intrusion detection and rapid response. The proposed framework integrates distributed multi-sensor nodes – comprising motion, acoustic, and vibration sensors – with edge computing and cloud-based monitoring platforms. A hybrid detection model combining rule-based logic and lightweight machine learning algorithms is implemented to enhance detection accuracy while minimizing false alarms. Simulation results indicate that the system achieves a high detection accuracy of 96.8% with a significantly reduced false alarm rate of 3.2%, alongside an average response time of 1.8 seconds, enabling near real-time threat identification. Energy optimization techniques, particularly duty cycling and event-driven communication, resulted in a 28% reduction in power consumption, making the system suitable for deployment in resource-constrained environments. Further validation using a confusion matrix shows high True Positive and True Negative rates (968 each), with minimal False Positives and False Negatives (32 each), confirming balanced classification performance. The ROC curve demonstrates excellent discriminative capability with an Area under the Curve (AUC) approaching unity, while FAR vs FRR analysis reveals a low Equal Error Rate (EER), indicating optimal threshold selection. Overall, the system provides a reliable and efficient framework for enhancing security infrastructure.},
                  keywords = {Internet of Things (IoT), Wireless Sensor Networks (WSN), Intrusion Detection, Smart Surveillance, Machine Learning, Edge Computing},
                  month = {May}
        }