Agriculture is one of the pillars to the social, economical and political well-being of Ethiopia. Weather, farm inputs, farm management practices and soil health status are among the significant parameters hindering farm productivity of farmers. Extensive cropping and grazing practices have also affected the country’s soil health which caused poor water holding capacity, nutrient depletion, reduced soil depth and poor productivity. In countries like Ethiopia, where agriculture has a significant economical role, it is imperative to monitor the status of the soil regularly to make logical decisions regarding farming practices, input use and resource management, among others using Wireless Sensor Networks. In recent years, the automation in agriculture has become a significant issue. The parameters like soil pH, soil moisture, Nitrogen, Phosphorus, Potassium, temperature, and humidity play a crucial role to increase crop yield. In this paper we have developed an architecture that is deployed using sensors to monitor the health status of soil. In this work, Soil Moisture, DHT22 / AM2302, PH and Nitrogen Phosphorus Potassium sensors were used to collect real time data of soil in the farm land. Here the deployed sensors along with the Arduino Microcontroller circuit is made to detect the different component of the soil. In this work, when a soil moisture sensor is outside of anything and immersed in very dry soil, it gives 1024 reading. As the water is poured to soil increases, the reading is less than 370. On this study k-nearest neighbor which uses proximity to make classifications or predictions about the grouping of an individual data point, were used to classify the soil health status. For this classification of the soil health, we used K nearest neighbor (KNN) supervised machine learning algorithm on the data collected by deployed sensors. And, the evaluation is performed on KNN using accuracy as evaluation metrics. In this thesis, the KNN algorithm is affected by the value of K that we choose and the accuracy rate is better at k= 3 i.e. 87.42%. Keywords: Wireless sensor networks, soil health monitoring, Nutrient detection, Environment monitoring, Sensors.