The paper examines the issue of interference impact on the control channels of FPV (First Person View) unmanned aerial vehicles. An approach combining hardware-in-the-loop and neural network modeling is proposed for assessing and predicting the influence of deliberate interference on remote control protocols. Hardware-in-the-loop modeling allows for reproducing complex electromagnetic interference environments using equipment (such as jammers) combined with software simulation, ensuring high reliability of results. The neural network model, based on a multilayer perceptron (MLP), is trained on data obtained from hardware-in-the-loop experiments and can quickly predict the effectiveness of interference impacts across various scenarios. Descriptions of the models, mathematical relationships, and calculation examples are provided, along with an implementation example of a simple neural network. The conclusion highlights the advantages of the integrated use of hardware-in-the-loop and neural network modeling for selecting interference parameters that disrupt FPV control protocols, and ensuring the reliability of drone control system tests under interference conditions.