Title :
Conditions for the equivalence of ARMAX and ARX systems
Author :
McGraw, G.A. ; Gustafson, C.L. ; Gillis, J.T.
Author_Institution :
Aerospace Corp., Los Angeles, CA, USA
fDate :
4/1/1993 12:00:00 AM
Abstract :
It is shown that an autoregressive moving average with exogenous input (ARMAX) system can be represented as an autoregressive with exogenous input (ARX) model if and only if the polynomial matrix corresponding to the moving average (MA) part of the system does not drop rank. A construction using the matrix fractional description of the system is used to prove this result. This construction shows that accurate ARX parameter estimates of systems driven by unmeasured disturbances can often be obtained by proper addition of sensor measurements and extending the order of the ARX model
Keywords :
estimation theory; statistical analysis; time series; ARMAX system; ARX systems; matrix fractional description; parameter estimates; polynomial matrix; unmeasured disturbances; Autoregressive processes; Gaussian noise; Linear systems; Measurement standards; Noise measurement; Parameter estimation; Polynomials; Sensor systems; Sufficient conditions; Yttrium;
Journal_Title :
Automatic Control, IEEE Transactions on
