Skip to Main Content
The authors consider the link of parallel profiles with linear dependencies (PARALIND) model with multiple-input multiple-output orthogonal frequency division multiplexing receiver. The authors derive a PARALIND-based blind receiver algorithm that has much close performance to non-blind minimum mean-squared error and it has better performance than constant modulus method. Neither the channel state information nor statistical characteristic is a necessity for the presented PARALIND algorithm, whereas the proposed algorithm distinctively supports small sample sizes. Numerical results verify its useful behaviour.