Estimation with a helper who knows the interference

We consider the problem of estimating a signal corrupted by independent interference in which the estimator (decoder) is aided by a helper (encoder) with a limited power budget that has knowledge of the interfering signal noncausally. In the Gaussian case, this problem is equivalent to the problem of Assisted Interference Suppression considered by Grover and Sahai and we obtain an improved lower bound for this problem using Verdú´s relationship between mismatched estimation and relative entropy. We also extend our analysis to consider the case when the signal to be estimated, in addition to the interference, is known to the helper. We establish a lower bound that improves on those recently derived by Huang and Narayanan.