Tirza Routtenberg

In this paper, a new class of lower bounds on the outage error probability in Bayesian vector parameter estimation is proposed. The minima of the h-outage error probability is obtained by the maximum windowed a-posteriori probability. However, computation of this estimator and its corresponding performance is usually not tractable and thus, lower bounds on the outage error probability can be very useful for performance analysis. The proposed class of lower bounds on the outage error probability is derived using Hölder's inequality. The proposed bounds were implemented for the linear Gaussian model. It was shown that in this case the tightest proposed bound attained the minimum probability of outage error.