کنفرانس مهندسی برق ایران

صفحه اصلی / سی و چهارمین کنفرانس بین المللی مهندسی برق

Efficient Data Detection in Massive Multiuser MIMO Systems Using Manifold Optimization

نویسندگان :

Ali Ghanbarpour¹ Mojtaba Amiri² Ali Olfat³

1- School of Electrical and Computer Engineering, University of Tehran 2- School of Electrical and Computer Engineering, University of Tehran 3- School of Electrical and Computer Engineering, University of Tehran

کلمات کلیدی :

Data detection،MMSE،massive MIMO systems،Riemannian manifold optimization

چکیده :

A major challenge for uplink multiuser massive multiple-input and multiple-output (MIMO) systems is the data detection problem at the receiver. Riemannian manifold optimization based methods have been developed for detection in communication systems. Manifold-based methods exploit the geometry of Riemannian manifold to convert the constrained problem into unconstrained. They not only have a better performance than classical MMSE detectors, but also, do not suffer performance loss when the number of user terminals (UT) is comparable to the number of base station (BS) antennas. The proposed algorithm is developed within the Riemannian manifold framework and follows a generalized Newton approach; we refer to this method as the Generalized Newton Method (GNM). The GNM is an appriximation of Newton method for the case where the cost function has norm form and simplifies the equation in Newton method by approximating the Riemannian gradient and Hessian. As rough first and second derivative are used in Newton equation, GNM has faster convergence. Simulation results show that the proposed detector significantly enhances performance in different test scenarios.