Figure 9 The IMAT for detecting the number, location, and values of sparsity.
We study the performance of different transmission scenarios by repeating our simulations for different values of sparsity factor, k n, and near-sparsity parameter, ε k.
The numbers of zero-slip and nearly-zero-slip subfaults were approximately similar to the true numbers of subfaults, which indicates that the selected values of sparsity hyperparameters λ and v are valid.
Furthermore, increasing the value of the sparsity control parameter may decrease the number of prior connections by enforcing more sparsity and eliminating connections that are least consistent with the experimental data.
When the SNR is higher than 15dB, the mean value of the sparsity estimate is two.
Hence, the actual value of the sparsity is K = 2, and the simulation result is in agreement with the simulation setting.
When the neighbourhood size s is fixed, from Figure 9, the smaller value of the sparsity level K 0 will give the greater accuracy of the classification.
The optimal clustering solution was obtained by examining the proportion of deviance (POD) score for different pre-specified numbers of clusters k (k = 2, 3, 4, 5, 6) and for different values of the sparsity parameter λ (λ = 0, 0.05, 0.15, 0.15 0.15).
The results are evaluated quantitatively for different values of the sparsity-controlling parameter α.
To measure the probability of exact reconstruction, we performed 500 trials for a single sparsity value of k.
