Search for a command to run...
The article presents an approach to constructing a mathematical model for predicting the throughput capacity of unmanned aerial vehicle communication channels based on the analysis of SITL telemetry logs. The proposed methodology takes into account the dynamic nature of the information flow, the unevenness of the intervals between packets, the probability of losses, and the fluctuations in channel parameters under the influence of external and internal factors. The paper identifies the key characteristics that affect channel stability: packet arrival rate, amount of useful information, loss rate, and throughput degradation rate. The structure of information flows in the UAV control system is considered and the relationships between channel parameters are formalised, which allows predicting its future state. The purpose develop a mathematical model for predicting the throughput capacity of the UAV communication channel using SITL telemetry logs to ensure early detection of channel degradation and improve data exchange reliability. Results. Based on the analysis of packet protocols and channel behaviour in different load modes, a mathematical model was developed to describe the temporal dynamics of throughput and the relationship between current parameters and predicted values. The model includes data normalisation, time series smoothing, estimation of the local rate of change in throughput, and determination of the expected effective throughput, taking into account packet loss. Modelling was performed in Mission Planner, which implemented various load scenarios: changes in telemetry frequency, artificial interference, SNR fluctuations, and increased traffic intensity. The results of the experiments showed that the forecast data corresponded to the actual values and confirmed that the model is capable of detecting the approach of the channel to a critical state in advance. The derivative throughput indicator, which decreases sharply before the onset of peak losses, proved to be particularly informative. Conclusions. The results obtained show that mathematical forecasting of the UAV communication channel bandwidth based on telemetry logs allows for early identification of critical channel conditions and increases the efficiency of information system resource utilisation. The proposed model can be integrated into real-time systems to adapt telemetry parameters, increase the stability of communication channels, and ensure the reliability of information transmission even in unstable environments and under variable loads. Keywords: unmanned aerial vehicle, mathematical model, data flows, bandwidth, telemetry, SITL, information flows, communication channel, channel degradation, data transmission stability.