I am dealing with a sort of "variance-covariance" matrix that surges from the multiplication of the Hankel matrix of a complex-valued time series and it's conjugate transpose. $$R_X=\mathcal{H}\mathcal{H}^*$$ (where * represents the conjugate transpose).


This part is not important for the question but may offer some context.

I estimate an eigendecomposition of the $R_X$ matrix such that I end up with

$$R_X=U\cdot (\Lambda_X +\sigma^2\cdot I)\cdot U^T$$

where $\Lambda_X$ and $\sigma^2\cdot I$ are the eigenvalues of the variance-covariance matrices of two independent sources.

By some means I manage to estimate $\sigma^2\cdot I$ such that I could obtain a new matrix $R_{X2}=U\cdot (\Lambda_X)\cdot U^T$.


Theoretically $R_{X2}$ is also the result of multiplying the Hankel matrix of a complex-valued time series and it's conjugate transpose.

$$R_{X2}=\mathcal{H_2}\mathcal{H_2}^*$$

But I don't know $\mathcal{H_2}$.

So this lead me to the original question: Is there any way of obtaining $\mathcal{H_2}$ knowing $R_{X2}$? Where the latter is a complex-valued positive semidefinite matrix?

Such a decomposition of a matrix into the product of a Hankel matrix and it's conjugate transpose is unique?

Any hint or suggestion will be greatly appreciated.

Your Answer

 

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.