Denoised Dataset of Aquaponic Fish Pond Water Quality Measurement using Internet of Things devices

The original raw dataset used in this study is publicly available on Mendeley Data at https://data.mendeley.com/datasets/yd36bx6f8f/2. contains measurements of pH, TDS, and water temperature using internet of things devices and sensors. The dataset is collected using an IoT sensor with ESP8266 as the microcontroller. Measurements were made on aquaculture consisting of a 1 m × 1 m × 1 m pond media with a water volume of 1 m × 1 m × 70 cm and hydroponic media with the Nutrient Film Technique (NFT) system. Measurements were conducted for three months from January 2023 to March 2023 with data dimensions of 118,286 rows and 5 columns. The data provided has been filtered according to the optimal recommended value: pH data is in the range of 6.5–8.5, TDS data ≤ 500 mg/L (ppm), and water temperature data ranges from 24–27 °C. In this dataset, the raw time-series data were further processed using several denoising techniques to suppress sensor noise and impulsive spikes while preserving the intrinsic temporal relationships among environmental parameters. Specifically, three denoising approaches were applied: a rolling median filter, a Hampel-based statistical filter, and an Adaptive Kalman Filter (AKF). These methods were used to generate denoised datasets that were subsequently evaluated to assess how noise reduction influences the structural predictability and inter-parameter relationships among pH, TDS, and water temperature.