超越最小距离：AWGN球码高信噪比误差概率展开中的最优首项系数
Beyond Minimum Distance: The Optimal Leading Coefficient in the High-SNR Error-Probability Expansion for AWGN Spherical Codes
浏览论文内容
中文总结 AI 辅助
本文研究AWGN球码高SNR误差概率展开的最优首项系数，证明逐SNR重新优化码本可得到更小首项系数，推导了不同参数下的系数值，指出优化可改善渐近误差。
中文摘要 AI 辅助
填充最优的$(M,n)$球码在高信噪比（SNR）下可实现最大可达最小距离，因此拥有最优误差指数。在随SNR增长保持固定的码本中，最小首项系数$K_{M,n}^{\text{fix}}$等于有序最近对的最小数量。本文证明，在每个SNR处重新优化码本可得到更小的首项系数$K_{M,n}^\boldsymbol{\times}$。具体而言，SNR意义下的最小精确最大似然误差概率为$P_e^\boldsymbol{\times}=B_{\boldsymbol{\times}}^\boldsymbol{\times}(K_{M,n}^\boldsymbol{\times}+o(1))$，而填充最优码本上的最小精确误差为$P_e^{\text{fix}}=B_{\boldsymbol{\times}}^\boldsymbol{\times}(K_{M,n}^{\text{fix}}+o(1))$，其中$B_{\boldsymbol{\times}}^\boldsymbol{\times}$为公共因子。本文证明$K_{M,n}^\boldsymbol{\times}\boldsymbol{\times}K_{M,n}^{\text{fix}}$，严格不等式$K_{M,n}^\boldsymbol{\times}\boldsymbol{\times}K_{M,n}^{\text{fix}}$因此在所有足够高的SNR下产生更小的精确误差，即$P_e^\boldsymbol{\times}\boldsymbol{\times}P_e^{\text{fix}}$。本文还将$K_{M,n}^\boldsymbol{\times}$表征为最小高斯软填充能量的极限。正交形界构造解释了该严格不等式：依赖SNR的扰动使一类极限最近对略微更近，而不改变最优指数及其对构造族首项系数的单位贡献，同时将其余最近对在更大但仍消失的尺度上推得更远，导致其贡献消失。对于$2\boldsymbol{\times}M\boldsymbol{\times}n+1$，本文证明$K_{M,n}^\boldsymbol{\times}=K_{M,n}^{\text{fix}}=M(M-1)$；对于$M=n+k$（$2\boldsymbol{\times}k\boldsymbol{\times}n$），本文证明$K_{n+k,n}^{\text{fix}}=4n(k-1)$且$K_{n+k,n}^\boldsymbol{\times}\boldsymbol{\times}4k(k-1)$，特别地，当$n\boldsymbol{\times}3$时$K_{n+2,n}^\boldsymbol{\times}=8\boldsymbol{\times}4n=K_{n+2,n}^{\text{fix}}$，因此SNR意义下的优化将渐近误差改善了$n/2$倍；当$k=n$时，两个系数均等于$4n(n-1)$；对于$3\boldsymbol{\times}k\boldsymbol{\times}n-1$，本文猜想$K_{n+k,n}^\boldsymbol{\times}=4k(k-1)$。
英文摘要
Packing-optimal $(M,n)$ spherical codes attain the largest achievable minimum distance and hence the optimal error exponent at high SNR. Among these codebooks held fixed as SNR grows, the smallest leading coefficient, $K_{M,n}^{\mathrm{fix}}$, equals the smallest number of ordered closest pairs. We show that reoptimizing the codebook at every SNR can yield a smaller leading coefficient $K_{M,n}^\ast$. Specifically, we show that the SNR-wise minimum exact maximum-likelihood error probability is $P_e^\ast=B_γ^\ast(K_{M,n}^\ast+o(1))$, while the minimum exact error over packing-optimal codebooks is $P_e^{\mathrm{fix}}=B_γ^\ast(K_{M,n}^{\mathrm{fix}}+o(1))$, where $B_γ^\ast$ is a common factor. We prove that $K_{M,n}^\ast\leq K_{M,n}^{\mathrm{fix}}$. Strict inequality, $K_{M,n}^\ast < K_{M,n}^{\mathrm{fix}}$ therefore gives a smaller exact error at all sufficiently high SNRs, i.e., $P_e^\ast<P_e^{\mathrm{fix}}$. We also characterize $K_{M,n}^\ast$ as the limit of the minimum Gaussian soft-packing energy. An orthoplex-bound construction explains strict inequality: an SNR-dependent perturbation makes one class of limiting closest pairs slightly closer, without changing the optimal exponent or their unit contributions to the constructed family's leading coefficient, while moving the remaining closest pairs farther apart on a larger but still vanishing scale, causing their contributions to vanish. For $2\leq M\leq n+1$, we prove $K_{M,n}^\ast=K_{M,n}^{\mathrm{fix}}=M(M-1)$. For $M=n+k$, $2\leq k\leq n$, we prove $K_{n+k,n}^{\mathrm{fix}}=4n(k-1)$ and $K_{n+k,n}^\ast\leq 4k(k-1)$. In particular, $K_{n+2,n}^\ast=8<4n=K_{n+2,n}^{\mathrm{fix}}$ for $n\geq3$, so SNR-wise optimization improves the asymptotic error by the factor $n/2$. At $k=n$, both coefficients equal $4n(n-1)$. For $3\leq k\leq n-1$, we conjecture $K_{n+k,n}^\ast=4k(k-1)$.
