Active RIS Assisted Rate-Splitting Multiple Access Network: Spectral and Energy Efficiency Tradeoff

Abstract

With the increasing demand of high data rate and massive access in both ultra-dense and industrial Internet-ofthings networks, spectral efficiency (SE) and energy efficiency (EE) are regarded as two important and inter-related performance metrics for future networks. In this paper, we investigate a novel integration of rate-splitting multiple access (RSMA) and reconfigurable intelligent surface (RIS) into cellular systems to achieve a desirable tradeoff between SE and EE. Different from the commonly used passive RIS, we adopt reflection elements with active load to improve a newly defined metric, called resource efficiency (RE), which is capable of striking a balance between SE and EE. This paper focuses on the RE optimization by jointly designing the base station (BS) transmit precoding and RIS beamforming (BF) while guaranteeing the transmit and forward power budgets of the BS and RIS, respectively. To efficiently tackle the challenges for solving the RE maximization problem due to its fractional objective function, coupled optimization variables, and discrete coefficient constraint, the formulated nonconvex problem is solved by proposing a two-stage optimization framework. For the outer stage problem, a quadratic transformation is used to recast the fractional objective into a linear form, and a closed-form solution is obtained by using auxiliary variables. For the inner stage problem, the system sum rate is approximated into a linear function. Then, an alternating optimization (AO) algorithm is proposed to optimize the BS precoding and RIS BF iteratively, by utilizing the penalty dual decomposition (PDD) method. Simulation results demonstrate the superiority of the proposed design compared to other benchmarks.

Metasurface-Assisted Terahertz Sensing

“…The function of active metasufaces that have high flexibility for manipulating electromagnetic waves is tunable and even reconfigurable. The concept of reconfigurable intelligent surfaces (RISs) composed of versatile active metasurfaces has been proposed and researched in many areas, such as secure satellite transmission networks [79][80][81][82]. Espe-cially, graphene and MEMS are suitable for metasurface-based THz sensing due to their tunable characteristics, effectively functioning on the THz band, leading to a research hotspot in the biodetection field during the decade.…”

Secure transmission for reconfigurable intelligent surface assisted communication with CV‐DDPG

“…The other two target networks are just used to calculate the action a and replace the estimated values ŷt as Equation (16). They can avoid the instability caused by excessive parameter fluctuations, while also reducing the correlation between the training network and the input samples.…”

“…Compared to traditional relaying, it can significantly enhance the degrees of freedom within wireless channels and improve channel quality [15]. In addition, the active RIS can provide greater flexibility and adaptability, allowing real-time adjustments based on specific requirements [16]. While the passive RIS allows large-scale antennas to achieve massive-MIMO gains without high power consumption [17].…”

Research on the Current Situation of Korean Literature Teaching in Qingdao National University, Shandong Province, China

“…The sum-rate maximization (SRM) problem in ( 2) is a basic problem in the active RIS system. It resembles many system designs including secrecy rate maximization [17], energy efficiency maximization [18], and sub-connected array architecture [12,14]. However, solving problem (2) is difficult.…”

“…Problem ( 2) is a non-convex program as the objective function and the constraint (2d) are nonconvex; plus, the variables w and φ are coupled in both the objective function and constraint (2d). In the existing literature, the most popular way is to apply block coordinate descent (BCD) and general purpose optimizers, such as CVX [11]; see, e.g., [8,12,14,17,18]. This can incur prohibitively high computational complexity when M and N are large, which happens in massive MIMO and large-scale RIS systems.…”