Adaptively tunable coding procedure for transmission of position information over the limited-band communication channel
Alexander Fradkov, Boris Andrievsky
The problem is considered of transmission object position
estimates using band limited communication channels. Due to
advances in technology, it is feasible to have a tracking filter attached to the sensor, and transmit only the condensed information as a tradeoff between computation and communication requirements. This paper proposes a scheme, which minimizes transmission data rate by sending adaptively encrypted innovations. The system considered consists of sensor node, which contains a tracking filter and encoder, bandwidth limited communication channel, and receiver node with a decoder. It is assumed that the sensor uses linear Kalman filter for tracking, and that the receiver knows all the sensor parameters. The source(sensor) node performs a simple Kalman filter based tracking. The channel is assumed bandwidth limited, but otherwise noiseless. It is assumed that the second derivative of the object position is bounded above by the known constant, and the sample rate is the design parameter. Adaptive tuning procedure for the coder range
parameter is proposed and numerically studied.