Instead of approximating the independent Q-factor, the approximate rollout action (widetilde {a}^{RL}(m)) is obtained by maximizing the approximation of the Q-factor difference (Q_{m}(a -Q_{m}(a -Q_{mcal {H}})), widetilde{a^{mathcal = argmaxlimits_{a in {matH}}ak{{A}}}}left{widetilde{a}_{m}(a)-widetilde{Q}_{m}(a^{RL}hcal {H}})right} (54).
This means that we calculate the dictionary matrix D maximizing the approximation performance of all y i (i = 1,2,…,N) in terms of the SSIM index under the constraint that the number of the nonzero elements in x i (i = 1,2,…,N) is T or less.
The information needed to resolve the phase ambiguity is contained in the preamble and, by a proper selection of the analyzed message, we can maximize the approximation impact to the key metrics such as BER or MSE of the phase estimation.
In the above equation, we try to maximize the approximation performance of y i (i = {1,2,…,N|x i (j) ≠ 0}) by x i (j d j, i.e., by the target atom d j and its corresponding representation coefficient x i (j).
Although we cannot achieve this goal directly, we can maximize the approximation of given by S x 0), p) (see (14)), where we require that x1 0) = x2 0) = and || ||2 = β.
The data covariance was used to estimate the Multiple Sparse Priors maximizing the free-energy approximation to the model-evidence (using automatic relevance determination) [31].
Specifically, for each y i (i = 1,2,…,N), we first search one atom which provides its optimal approximation, maximizing the SSIM index.
A close approximation to maximizing the minimum SNR for ZF receiver is also provided under the assumption of M
24 Estimations were made by maximizing the likelihood of the data, using the stochastic approximation expectation maximization algorithm followed by importance sampling to obtain the objective function value for hypothesis testing.
In this approach, we introduce the SSIM index as the approximation performance, and then, the optimal reconstruction results maximizing the SSIM index can be obtained.
It is worth noting that beside maximizing the sum-rate, the problem of maximizing the individual rate using the approximation in (16) has the same solution obtained in (22).
