Dynamic metasurface antennas (DMAs) is a new antenna technology comprising multiple microstrips, each encompassing several reconfigurable metamaterial elements acting as radiating and receiving antennas. The whole unit is usually connected via the input/output port of each microstrip to the radio frequency (RF) chains of the baseband unit. Point-to-point (P2P) multiple-input multiple-output (MIMO) system is one of the most prominent wireless communication systems, with its use cases recently increasing with the booming of machine-to-machine communication. However, it is yet to be explored with DMAs. This work thus studies the achievable rate maximization of a DMA-based P2P MIMO system whose transmitter or receiver is equipped with DMA instead of traditional antennas. Noteworthy, this is the first work to analyze the deployment where both the BS and the user are equipped with DMA. Additionally, we numerically demonstrate how uplink and downlink communication can be achieved in each DMA deployment strategy. Finally, through the simulation results, the performance of our proposed designs is compared with various benchmark systems, such as conventional P2P and IRS-aided MIMO systems, where in both cases, it is shown that the proposed designs provide the best tradeoff in achieving high performance with significant RF chains reduction.

In this work, we study the dynamic metasurface antennas (DMA) for use in the transmission of wireless communication signals instead of conventional metallic antennas. DMA belongs to a family of metamaterials, which allow their properties to be reconfigured in real-time. The reconfiguration of DMA is mostly hindered by the associated Lorentzian constrain. In this work, we propose a technique of splitting the Lorentzian constraints into two parts, which allows the avoidance of conventional relaxation techniques. The splitting technique eventually led to performance improvement in both single-user and multiuser MISO systems.