arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1305.0870cs.DScs.ITcs.MMmath.IT
收藏

最多使用4k个样本计算k稀疏n长度离散傅立叶变换,复杂度为O(k log k)

Computing a k-sparse n-length Discrete Fourier Transform using at most 4k samples and O(k log k) complexity

Sameer Pawar, Kannan Ramchandran

摘要:

给定$n$长度的输入信号$\mbf{x}$,众所周知,其离散傅立叶变换(DFT)$\mbf{X}$可以使用快速傅立叶变换(FFT)以$O(n \log n)$复杂度计算.如果谱$\mbf{X}$恰好是$k$ -稀疏的(其中$k<<n$),我们能做得更好吗?当$k$在$n$中是次线性时,我们证明了在$k$和$n$中是渐近的(精确地,$k \propto n^δ$,其中$0 < δ<1$),并且非零DFT系数的支持是均匀随机的,我们可以以两种基本方式利用这种稀疏性(i){ \bf {样本复杂度}}:我们仅需要$M=rk$确定性地选择的输入信号$\mbf{x}$的样本(其中当$0 < δ< 0.99$时为$r < 4$);以及(ii){ \bf {计算复杂度}}:我们可以使用$O(k \log k)$运算来可靠地计算DFT $\mbf{X}$,其中大Oh中的常数很小,并且与计算长度近似等于稀疏参数$k$的少量DFT中所涉及的常数有关。我们的算法成功的概率很高,与失败的概率渐近消失到零的样本数量,$M$。

英文摘要:

Given an $n$-length input signal $\mbf{x}$, it is well known that its Discrete Fourier Transform (DFT), $\mbf{X}$, can be computed in $O(n \log n)$ complexity using a Fast Fourier Transform (FFT). If the spectrum $\mbf{X}$ is exactly $k$-sparse (where $k<<n$), can we do better? We show that asymptotically in $k$ and $n$, when $k$ is sub-linear in $n$ (precisely, $k \propto n^δ$ where $0 < δ<1$), and the support of the non-zero DFT coefficients is uniformly random, we can exploit this sparsity in two fundamental ways (i) {\bf {sample complexity}}: we need only $M=rk$ deterministically chosen samples of the input signal $\mbf{x}$ (where $r < 4$ when $0 < δ< 0.99$); and (ii) {\bf {computational complexity}}: we can reliably compute the DFT $\mbf{X}$ using $O(k \log k)$ operations, where the constants in the big Oh are small and are related to the constants involved in computing a small number of DFTs of length approximately equal to the sparsity parameter $k$. Our algorithm succeeds with high probability, with the probability of failure vanishing to zero asymptotically in the number of samples acquired, $M$.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑