Enhanced Tensor-Based Joint Channel and Symbol Estimation in Dual-Hop MIMO Relaying Systems

In this letter, an enhanced alternating least squares (EALS) algorithm for joint channel and symbol estimation is proposed in a dual-hop multiple input multiple output (MIMO) relaying communication system based on a simplified Khatri-Rao space-time (KRST) coding at the source node. The proposed method exploits the parallel factor (PARAFAC) and PARATUCK2 models at the relay node and the destination node, respectively. Exploiting the structure of these tensor models, we present an iterative alternating estimation scheme using optimized step size factors for accelerated convergence and lower overall complexity compared to conventional approaches. Identifiability conditions and computational complexity are analyzed, and simulation results are provided to corroborate the effectiveness of the proposed algorithm.