部分LLL减少
Partial LLL Reduction
摘要:
Lenstra-Lenstra-Lovasz(LLL)约化在数字通信中有着广泛的应用。它可以大大提高球形译码(SD)算法求解整数最小二乘(ILS)问题的速度,并改善ILS问题的次优解Babai整数点的性能。最近Ling和Howgrave Graham提出了所谓的有效LLL(ELLL)降低。它的计算复杂性比LLL小,而对Babai整数点的性能影响与LLL相同。在本文中,我们提出了一个部分LLL(PLLL)减少。PLLL避免了ELLL的数值稳定性问题,该问题可能导致Babai整数点的性能非常差。此外,数值模拟表明,它比ELLL更快。我们还表明,在理论上,PLLL和ELLL有一个典型的SD算法作为LLL的搜索速度相同的效果。
英文摘要:
The Lenstra-Lenstra-Lovasz (LLL) reduction has wide applications in digital communications. It can greatly improve the speed of the sphere decoding (SD) algorithms for solving an integer least squares (ILS) problem and the performance of the Babai integer point, a suboptimal solution to the ILS problem. Recently Ling and Howgrave-Graham proposed the so-called effective LLL (ELLL) reduction. It has less computational complexity than LLL, while it has the same effect on the performance of the Babai integer point as LLL. In this paper we propose a partial LLL (PLLL) reduction. PLLL avoids the numerical stability problem with ELLL, which may result in very poor performance of the Babai integer point. Furthermore, numerical simulations indicated that it is faster than ELLL. We also show that in theory PLLL and ELLL have the same effect on the search speed of a typical SD algorithm as LLL.
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。