Energy Saving for Cognitive Multicast OFDM Systems: A Time-Frequency Two-Dimensional Method

Wenjun XU  Shengyu LI  Zhihui LIU  Jiaru LIN  

IEICE TRANSACTIONS on Communications   Vol.E98-B   No.6   pp.974-983
Publication Date: 2015/06/01
Online ISSN: 1745-1345
DOI: 10.1587/transcom.E98.B.974
Type of Manuscript: PAPER
Category: Energy in Electronics Communications
energy saving,  cognitive radio,  time-frequency,  multicast,  OFDM,  

Full Text: PDF>>
Buy this Article

This paper studies the energy-saving problem in cognitive multicast orthogonal frequency-division multiplexing (OFDM) systems, for which a time-frequency two-dimensional model is established to enable the system energy conservation through joint temporal and spectral adaptations. The formulated two-dimensional problem, minimizing the total power consumption whilst guaranteeing the minimal-rate requirement for each multicast session and constraining the maximal perceived interference in each timeslot for the active primary user, is categorized as mixed integer non-convex programming, whose optimal solution is intractable in general. However, based on the time-sharing property, an asymptotically optimal algorithm is proposed by jointly iterating spectrum element (SE) assignment and power allocation. Moreover, a suboptimal algorithm, which carries out SE assignment and power allocation sequentially, is presented as well to reduce the computation complexity. Simulation results show the proposed joint algorithm can achieve the near-optimal solution, and the proposed sequential algorithm approximates to the joint one very well with a gap of less than 3%. Compared with the existing slot-by-slot energy-saving algorithms, the total power consumption is considerably decreased due to the combined exploitation of time and frequency dimensions.