arXiv subjects

Aparna V C

Publications and source records attributed to Aparna V C.

2 recordsLinked to original sources

arXiv2025

RIS, Active RIS or RDARS: A Comparative Insight Through the Lens of Energy Efficiency

Multiplicative fading is a major limitation of reconfigurable intelligent surfaces (RIS), restricting their effective coverage in both existing sub-6GHz systems and future mmWave networks. Although active RIS architectures mitigate this issue, they require high power consumption and introduce practical challenges due to the need for integrated amplifiers. Recently, reconfigurable distributed antenna and reflecting surfaces (RDARS) have been proposed to alleviate multiplicative fading through connected modes. In this work, we compare RIS, active RIS, and RDARS in terms of coverage and energy efficiency (EE) in both sub-6GHz and mmWave bands, and we investigate the impact of placement and the number of elements of reconfigurable surface (RS) on EE and coverage. The simulation results show that RDARS offers a highly energy-efficient alternative of enhancing coverage in sub-6GHz systems, while active RIS is significantly more energy-efficient in mmWave systems. Additionally, for a lower number of RS elements and for near UEs, RIS remains considerably more energy-efficient than both active RIS and RDARS.

cs.IT↗

arXiv2025

Geometry Aware Meta-Learning Neural Network for Joint Phase and Precoder Optimization in RIS

In reconfigurable intelligent surface (RIS) aided systems, the joint optimization of the precoder matrix at the base station and the phase shifts of the RIS elements involves significant complexity. In this paper, we propose a complex-valued, geometry aware meta-learning neural network that maximizes the weighted sum rate in a multi-user multiple input single output system. By leveraging the complex circle geometry for phase shifts and spherical geometry for the precoder, the optimization occurs on Riemannian manifolds, leading to faster convergence. We use a complex-valued neural network for phase shifts and an Euler inspired update for the precoder network. Our approach outperforms existing neural network-based algorithms, offering higher weighted sum rates, lower power consumption, and significantly faster convergence. Specifically, it converges faster by nearly 100 epochs, with a 0.7 bps improvement in weighted sum rate and a 1.8 dB power gain when compared with existing work. Further it outperforms the state-of-the-art alternating optimization algorithm by 0.86 bps with a 2.6 dB power gain.

cs.LG↗