We formulate the subcarrier and power allocation problem in cognitive radio networks employing orthogonal frequency division multiplexing （OFDM） as a non-linear optimization problem with the objective of maximizing ...We formulate the subcarrier and power allocation problem in cognitive radio networks employing orthogonal frequency division multiplexing （OFDM） as a non-linear optimization problem with the objective of maximizing sum capacity under constraints of available subcarriers, interference temperature, power budget, etc. A close-to-optimal solution with much reduced complexity is proposed to separate the problem into two steps, which also considers fairness among secondary users. A fair al- gorithm for subcarrier allocation （FA_SA） is firstly presented. Secondly, a fast iterative water-filling algorithm for power allocation （FIWFA_PA） is also proposed to maximize the sum capacity. Exten- sive simulation results show that sum capacity performance of our low-complexity solution is very close to the optimal one, while significantly improving fairness and reducing computation complexity compared with the existing solutions.展开更多
基金Supported by the National High Technology Research and Development Programme of China（ No. 2007AA01Z221, No. 2009AA01Z246） , and the National Natural Science Foundation of China（ No. 60672124, 60832009）.
文摘We formulate the subcarrier and power allocation problem in cognitive radio networks employing orthogonal frequency division multiplexing （OFDM） as a non-linear optimization problem with the objective of maximizing sum capacity under constraints of available subcarriers, interference temperature, power budget, etc. A close-to-optimal solution with much reduced complexity is proposed to separate the problem into two steps, which also considers fairness among secondary users. A fair al- gorithm for subcarrier allocation （FA_SA） is firstly presented. Secondly, a fast iterative water-filling algorithm for power allocation （FIWFA_PA） is also proposed to maximize the sum capacity. Exten- sive simulation results show that sum capacity performance of our low-complexity solution is very close to the optimal one, while significantly improving fairness and reducing computation complexity compared with the existing solutions.
