Blockchain and AI enabled communications for indoor environmental quality and health monitoring

Purpose The purpose of this paper is to develop a blockchain-enabled, Internet of Things (IoT)-based intelligent system for indoor environmental quality and health monitoring process. Design/methodology/approach In the IoT system, the DHT11 temperature and humidity sensor, MQ135 air quality gas sensor and MAX30100 pulse oximeter sensor were connected to an Arduino Uno micro-controller to obtain readings which were pushed onto a remote client, i.e. the Raspberry Pi 3B+. The data was transmitted through a secure blockchain channel and retrieved on a local server whereby machine learning algorithms such as multi-layer perceptron (MLP) and long short-term memory (LSTM) were used for prediction. Findings The results of the blockchain analysis demonstrate that for an increasing array size, the number of blocks decreases and, in turn, time complexity increases, latency increases and throughput decreases. Data readings transmitted in numerous smaller blocks are more secure than data transmitted in few larger blocks. LSTM predicted SpO2 levels with an average accuracy of 99.00% while MLP predicted SpO2 levels with an average accuracy of 98.87%. It was observed that with a long exposure in high AQI environment, the poor air quality reduced the oxygen exchange with a lowest SpO2 level of 93% in the bloodstream. Conversely, in low AQI environments, SpO2 levels above 95% were recorded. Originality/value The main originality of this paper a secured IoT system which uses AI to determine how AQI and heat index affect SpO2 levels in the blood stream.