The document describes the derivation of a simple Kalman filter. It begins by introducing a process state x and measurement z, related by equations with additive noise terms w and v. An a priori state estimate x^ is updated using a measurement z to give an a posteriori estimate x+. The blending factor K is derived by minimizing the a posteriori error variance, yielding K = s^/ (s^ + sv), where s^ is the a priori variance and sv is the measurement noise variance. This optimal K balances the a priori estimate and measurement based on their relative uncertainties. The Kalman filter thus combines estimates and measurements in a way that minimizes the estimated error.