ivaBSS (version 1.0.0)

avg_ISI: Average Intersymbol Inference

Description

Calculates the average intersymbol inference for two sets of matrices.

Usage

avg_ISI(W, A)

Arguments

W

Array of unmixing matrices with dimension [P, P, D].

A

Array of true mixing matrices with dimension [P, P, D].

Value

Numeric value between 0 and 1, where 0 is the optimal result indicating that the sources are separated perfectly in each dataset.

Details

The function returns the average intersymbol inference for the set of estimated unmixing matrices and the set of true mixing matrices. The average ISI gets the value between 0 and 1, where 0 is the optimal result. The average ISI is calculated as the mean ISI over each dataset separately. The average ISI does not take the permutation of the estimated sources into account.

References

Anderson, M. (2013). Independent vector analysis: Theory, algorithms, and applications. PhD dissertation, University of Maryland, Baltimore County.

See Also

joint_ISI, jbss_achieved

Examples

# NOT RUN {
# Mixing matrices and unmixing matrices generated
# from standard normal distribution
P <- 4; D <- 4;
W <- array(rnorm(P * P * D), c(P, P, D))
A <- array(rnorm(P * P * D), c(P, P, D))

avg_ISI(W, A)

if (require("LaplacesDemon")) {
  # Generate sources from multivariate Laplace distribution
  P <- 4; N <- 1000; D <- 4;
  S <- array(NA, c(P, N, D))

  for (i in 1:P) {
    U <- array(rnorm(D * D), c(D, D))
    Sigma <- crossprod(U)
    S[i, , ] <- rmvl(N, rep(0, D), Sigma)
  }

  # Generate mixing matrices from standard normal distribution
  A <- array(rnorm(P * P * D), c(P, P, D))

  # Generate mixtures
  X <- array(NaN, c(P, N, D))
  for (d in 1:D) {
    X[, , d] <- A[, , d] %*% S[, , d]
  }

  # Estimate sources and unmixing matrices
  res_G <- NewtonIVA(X, source_density = "gaussian")
  avg_ISI(coef(res_G), A)
}
# }