A two step Burg algorithm [spectral analysis]

The problem of estimating the parameters of a real-valued, stationary, nondeterministic, autoregressive process of order p from a time series of finite length is discussed. Burg´s algorithm estimates these parameters indirectly by sequentially estimating one reflection coefficient at a time. The proposed approach is to sequentially estimate the reflection coefficients in pairs. The new algorithm has the same order of computational complexity as Burg´s. It is guaranteed to generate parameter estimates that correspond to a stationary process (as does Burg´s), and it produces estimates of the power spectral density that do not appear to suffer from spectral line splitting, in contrast to Burg´s algorithm