Research output: Contribution to Journal/Magazine › Journal article › peer-review
Research output: Contribution to Journal/Magazine › Journal article › peer-review
}
TY - JOUR
T1 - A performance comparative study on the implementation methods for OFDMA cross-layer optimization
AU - C. Zarakovitis, Charilaos
AU - Ni, Qiang
PY - 2012/6
Y1 - 2012/6
N2 - One key issue in cross-layer optimization techniques for next-generation multi-user orthogonal frequency division multiple access (OFDMA)-based broadband wireless network systems lies in the implementation methods of optimal resource scheduling. In the literature the optimal solutions are derived either through dynamic programming (referred to as a purely numerical approach) or via mathematical analysis (referred to as an analytical approach). When the latter approach is adopted, an additional iterative process is usually needed for the final optimal solution to be obtained. This paper presents a first in-depth comparative study on the implementation performance between the analytical and the numerical techniques. For this purpose, various popular iterative methods and numerical methods are investigated in our study. Several performance metrics (e.g., achieved overall data rate, absolute approximation error, and computational time) are utilized for comparison. Our simulation results demonstrate clearly that the analytical approaches indeed outperform the numerical ones. Furthermore, regarding different iterative methods, it is shown that the semi-implicit root (SIR) mechanism performs best in terms of the convergence rate, the root-finding accuracy, and the computational time. (C) 2011 Elsevier B.V. All rights reserved.
AB - One key issue in cross-layer optimization techniques for next-generation multi-user orthogonal frequency division multiple access (OFDMA)-based broadband wireless network systems lies in the implementation methods of optimal resource scheduling. In the literature the optimal solutions are derived either through dynamic programming (referred to as a purely numerical approach) or via mathematical analysis (referred to as an analytical approach). When the latter approach is adopted, an additional iterative process is usually needed for the final optimal solution to be obtained. This paper presents a first in-depth comparative study on the implementation performance between the analytical and the numerical techniques. For this purpose, various popular iterative methods and numerical methods are investigated in our study. Several performance metrics (e.g., achieved overall data rate, absolute approximation error, and computational time) are utilized for comparison. Our simulation results demonstrate clearly that the analytical approaches indeed outperform the numerical ones. Furthermore, regarding different iterative methods, it is shown that the semi-implicit root (SIR) mechanism performs best in terms of the convergence rate, the root-finding accuracy, and the computational time. (C) 2011 Elsevier B.V. All rights reserved.
KW - Cross-layer scheduling
KW - Semi-implicit root (SIR)
KW - Iterative method
KW - Multiple access
KW - Orthogonal Frequency Division Multiple Access (OFDMA)
KW - Optimization techniques
KW - Resource allocation
KW - Wireless networks
U2 - 10.1016/j.future.2011.10.008
DO - 10.1016/j.future.2011.10.008
M3 - Journal article
VL - 28
SP - 923
EP - 929
JO - Future Generation Computer Systems
JF - Future Generation Computer Systems
IS - 6
ER -