Increasing the Reliability of Information in the Control Circuits of Flight Apparatus Having Discrete Communication Channels

Adaptive management of processes, in particular the flight of aircraft (especially long- range ballistic missiles), requires the availability of operational and reliable information about the state of the process and its parameters. The initial information about the state of the process is the current values of the object's parameters, which are measured using primary measuring transducers - sensors. These sensors are the beginning of the information and measurement channels in the control circuit of the aircraft (LA), through which information about the parameters of the aircraft is sent to the on-board digital computer (BCOM), the reliability of which depends on the accuracy of achieving the goal of the flight of the LA. A method of increasing the reliability of information transmission has been developed based on the use of a group of coding systems, its adaptive selection as the best in the process of operational assessment of the state of a discrete channel on a real-time scale, while the synthesis of the initial adaptive predictive model (at the stage of identification in the "learning" mode) is carried out and adjusted this model in operational identification mode. An analysis of methods and algorithms for detecting errors in discrete communication channels in the presence of synchronization errors was carried out, as well as selected software tools for the study of error statistics in stationary and dynamic modes. It is noted that for practical purposes, various variants of suboptimal filter structures with adjustable parameters are used, which are adjusted, like regulators, for specific conditions of their application. A moving average filter and a filter implementing the exponential smoothing method are usually used. It should be noted that the software implementation of the exponential smoothing algorithm requires the performance of a smaller number of arithmetic operations and, accordingly, a smaller volume of RAM of the BCM than for the moving average algorithm. The choice of the type of filter and its amplitude-phase characteristic are determined from the condition of minimum root mean square deviation. The most effective of them are Kalman-Busy filters.