Tensor space–time (TST) coding for MIMO wireless communication systems

In this paper, we propose a tensor space–time (TST) coding for multiple-input multiple-output (MIMO) wireless communication systems. The originality of TST coding is that it allows spreading and multiplexing the transmitted symbols, belonging to R data streams, in both space (antennas) and time (chips and blocks) domains, owing the use of two (stream- and antenna-to-block) allocation matrices. This TST coding is defined in terms of a third-order code tensor admitting transmit antenna, data stream and chip as modes. Assuming flat Rayleigh fading propagation channels, the signals received by K receive antennas during P time blocks, composed of N symbol periods each, with J chips per symbol, form a fourth-order tensor that satisfies a new constrained tensor model, called a PARATUCK-(2,4) model. Conditions for identifiability and uniqueness of this model are established, and a performance analysis of TST coding is made, before presenting a blind receiver for joint channel estimation and symbol recovery. Finally, some simulation results are provided to evaluate the performance of this receiver. ► Presenting in a unified way several tensor-based MIMO wireless communication systems. ► Introducing a new tensor model, the so-called PARATUCK-( N 1 , N ) model. ► Proposing a new tensor space–time coding for MIMO wireless communication systems, with an associated blind receiver. ► Analyzing the performance of the proposed solution in terms of maximum diversity gain, and tensor model identifiability and uniqueness. ► Evaluating the performance of the proposed blind receiver by means of numerical simulations.