Hybrid Approach For Anomaly Detection Using Clustering Mechanism
Ruchika Rami, Zakiyabanu S. Malek · 2024
This paper explores the design and implementation of an IoT-based homeautomation system using the ESP32 microcontroller, integrated with DHT11,LDR, and gas sensors. The primary objective is to collect environmental data suchas temperature, humidity, ambient light levels, and air quality, and transmit thisdata to the ThingSpeak cloud platform for real-time monitoring and analysis. Byleveraging Wireless Sensor Networks (WSN), the data is fetched from ThingSpeakand analyzed in MATLAB using advanced clustering algorithms, specificallyfocusing on fuzzy clustering, k-medoids, and k-means, to detect anomalies withhigh accuracy and superior detection rates. The ESP32 microcontroller, known forits powerful processing capabilities and integrated Wi-Fi, serves as the system'score. The DHT11 sensor monitors temperature and humidity, the gas sensordetects various gases to ensure safety, and the LDR sensor measures ambient lightlevels for energy-efficient lighting control. Data transmitted to ThingSpeak isvisualized in real-time and retrieved for further analysis in MATLAB. Fuzzyclustering is emphasized for its ability to handle uncertainties and providenuanced anomaly detection by assigning membership levels to data points fordifferent clusters. K-medoids, robust to noise and outliers, uses actual data pointsas cluster centers, while k-means, although sensitive to noise, is also employed forpartitioning data into clusters. The system's performance is evaluated based onthroughput, latency, and detection rate. High throughput ensures efficient dataprocessing, low latency allows near real-time insights, and a high detection rateminimizes false positives and negatives. this project demonstrates significantimprovements over existing home automation systems, highlighting the potentialof IoT and advanced data analysis techniques in enhancing the functionality,reliability, and safety of smart homes.