Estimating the probability of messages delivery in an IoT system

Objective . The aim of the study is to develop a model for estimating the probability of message delivery in a medical Internet of Things (IoMT) system and to study the dependence of this value on the number of retransmissions used to compensate for distorted or lost telemetry data. Method . The transmission of telemetry messages is carried out in accordance with the MQTT-SN protocol. Message delivery is performed using the server as an intermediary device to which client devices connect, i.e. sensor devices that transmit measured data, and wireless devices for healthcare workers that receive this data. To develop a message delivery model, it is proposed to use the apparatus of probabilistic graphs. The adequacy of the model was verified based on computational experiments. Result . A mathematical model of the process of delivering telemetry messages in an IoT system has been developed, which adequately reflects the dependence of the probability of message delivery on the characteristics of wireless channels, parameters of transmitted packets and acknowledgments, as well as the allowed number of retransmissions. Conclusion . The model can be used to estimate the probability of message delivery in the medical Internet of Things system. The use of the model makes it possible to select theoretically justified values of the permissible number of retransmissions, established to achieve the required values of message delivery probability under the current level of bit error rate in wireless channels.