东南大学学报(自然科学版)
東南大學學報(自然科學版)
동남대학학보(자연과학판)
JOURNAL OF SOUTHEAST UNIVERSITY
2014年
1期
7-11
,共5页
黄毅%胡爱群%成月良%谢胜东
黃毅%鬍愛群%成月良%謝勝東
황의%호애군%성월량%사성동
差分互相关法%时延估计%OFDM
差分互相關法%時延估計%OFDM
차분호상관법%시연고계%OFDM
differential cross-correlation%time delay estimation%orthogonal frequency division mul-tiplexing (OFDM)
提出了一种适用于OFDM系统的基于训练序列的精确时延估计方法.该方法利用频域差分互相关法,精确提取小数时延;在频域补偿小数时延后,再将时域相关法与上升沿法相结合,得到整数时延.由复杂度分析结果可知,当搜索法中获取小数时延的搜索次数大于2时,该方法的小数时延估计计算复杂度与搜索法相比低很多.仿真结果显示,当采样频率为10 MHz、多普勒频移为222.22 Hz、子载波个数为1024的情况下,根据该方法所实现的时延估计精度能够达到分米级,平均比搜索法高出10 dB;同时,在低信噪比时,该方法比传统的时域相关法具有更健壮的整数时延估计性能.
提齣瞭一種適用于OFDM繫統的基于訓練序列的精確時延估計方法.該方法利用頻域差分互相關法,精確提取小數時延;在頻域補償小數時延後,再將時域相關法與上升沿法相結閤,得到整數時延.由複雜度分析結果可知,噹搜索法中穫取小數時延的搜索次數大于2時,該方法的小數時延估計計算複雜度與搜索法相比低很多.倣真結果顯示,噹採樣頻率為10 MHz、多普勒頻移為222.22 Hz、子載波箇數為1024的情況下,根據該方法所實現的時延估計精度能夠達到分米級,平均比搜索法高齣10 dB;同時,在低信譟比時,該方法比傳統的時域相關法具有更健壯的整數時延估計性能.
제출료일충괄용우OFDM계통적기우훈련서렬적정학시연고계방법.해방법이용빈역차분호상관법,정학제취소수시연;재빈역보상소수시연후,재장시역상관법여상승연법상결합,득도정수시연.유복잡도분석결과가지,당수색법중획취소수시연적수색차수대우2시,해방법적소수시연고계계산복잡도여수색법상비저흔다.방진결과현시,당채양빈솔위10 MHz、다보륵빈이위222.22 Hz、자재파개수위1024적정황하,근거해방법소실현적시연고계정도능구체도분미급,평균비수색법고출10 dB;동시,재저신조비시,해방법비전통적시역상관법구유경건장적정수시연고계성능.
An accurate time delay estimation scheme based on training sequence in orthogonal fre-quency division multiplexing (OFDM)systems is proposed.The scheme employs differential cross-correlation technology to extract the fractional delay in the frequency domain.Then,correlation technology and leading edge method are used to obtain the integer delay after compensation for the fractional delay in the frequency domain.The complexity analysis results show that if the number of searches for estimating the fractional delay is greater than two in the searching method,the computa-tional complexity in the fractional delay estimation of the proposed algorithm is much lower than that of the searching method.The simulation results show that the proposed scheme can achieve an accu-racy of decimeter level in the sampling frequency of 10 MHz,the Doppler shift of 222.22 Hz and the number of subcarriers of 1 024,which is about 10 dB higher than that of the searching method on average.Moreover,the proposed scheme has a more robust integer delay estimation performance than the traditional correlation technology even at low signal-to-noise ratio (SNR).