\[\Large \text{The Proof of Fibonacci GCD}\]
\[\text{By Sangber}\]


\(\text{Fibonacci Sequence}\)

\(\quad \quad \text{If we record sequence}\ \{F_n\}\ \text{as Fibonacci Sequence, then }\{F_n\} \text{ will have these properties:}\)

\[F_i = \begin{cases}1, & i\leq 2 \\ F_{i-1} + F_{i-2}, & \mathrm{otherwise}\end{cases}\]

\(\quad \quad \text{So we can see that } \{F_n\} \text{ is like “1,1,2,3,5,8,13,21...”.}\)


\(\text{GCD}\)

\(\quad \quad \text{“GCD” is the abbreviation of “Greatest Common Divisor”.}\)
\(\quad \quad \text{And we often record } (a, b) \text{ as the GCD of } a \text{ and } b.\)
\(\quad \quad \text{For example, } (12, 15) = 3, (48, 36) = 12.\)


\(\text{Two Theorems about GCD}\)

\(\text{Euclidean theorem}\)

\[(a, b) = (b, a \bmod b), a, b \in N_{+}\]
\(\text{The proof: }\)

  • \(\text{If } a = b, (a, b) = a = b, \text{we can surely tell that the theorem is established.}\)
  • \(\text{If } a < b, (b, a \bmod b) = (b, a) = (a, b), \text{the theorem is established.}\)
  • \(\text{If } a > b, \text{we can assume that } a = k \times b + r, (k \in N_{+}, 0 \le r < b), \text{ so } a \bmod b = r. \\ \text{As for } \forall d, \text{meeting that } d|a \land d|b, d|(a - k \times b), \text{ actually it means } d | r. \\ \text{So we can tell that } (a, b) = (b, r), \text{ that is, } (a, b) = (b, a \bmod b).\)

\(\text{Stein's Algorithm}\)

\[(a, b) = (a, b - a),b > a\]
\(\text{The proof:}\)
\(\text{Actually, the proof of it is similar to that of Euclidean theorem, you can finish it yourselves.}\)


\(\text{One Lemma}\)

\[(F_n,F_{n - 1})=1, n \in N_{+}\]
\(\text{The Proof}\):

  • \(\text{In the case of } n = 1 \text{ and } n = 2,\text{ the lemma is established obviously.}\)
  • \(\text{In the case of } n \ge 2 :\)
    \(\text{Obviously:}\)
    \[F_n > F_{n - 1} > F_{n - 2}, F_n = F_{n - 1} + F_{n - 2}\]
    \[\therefore F_n\ \bmod F_{n - 1} = F_{n - 2}\]
    \(\quad \ \text{According to Euclidean theorem and the lemma above:}\)
    \[(F_n,F_{n-1})=(F_{n-1},F_n\ \bmod\ F_{n-1})=(F_{n-1},F_{n-2})\]
    \[\therefore (F_n,F_{n-1})=(F_{n-1},F_{n-2})=(F_{n-2},F_{n-3})= \cdots =(F_1,F_2)=1\]

\(\text{Fibonacci GCD}\)

\(\quad \quad \text{Here is the expression of the Fibonacci GCD:}\)

\[\forall n,m \in \text{Z}^{+},(F_n, F_m) = F_{(n, m)}\]


\(\text{The Proof of Fibonacci GCD}\)

\(\text{We assume that } n<m.\)
\(\text{Then we use } F_n \text{ and }F_{n+1} \text{ to express } F_{n + 2},F_{n + 3},F_{n + 4},F_{n + 5}\cdots\)

\[
\begin{aligned}
& F_{n + 2} = 1 F_n + 1 F_{n + 1} \\
& F_{n + 3} = 1 F_n + 2 F_{n + 1} \\
& F_{n + 4} = 2 F_n + 3 F_{n + 1} \\
& F_{n + 5} = 3 F_n + 5 F_{n + 1}
\end{aligned} \\
\cdots \cdots
\]

\(\text{We can see that in the expressions above, the coefficients of } F_n \text{ and } F_{n + 1} \text{ meet the properties of the Fibonacci Sequence.}\)
\[\therefore F_m=F_{m - n - 1} \times F_n + F_{m - n} \times F_{n + 1}\]
\[\therefore (F_n, F_m) = (F_n, F_{m - n - 1} \times F_n + F_{m - n} \times F_{n + 1})\]
\(\text{And obviously we can see that :}\)
\[F_n|F_{m-n-1}\times F_n\]
\[\therefore (F_n,F_m)=(F_n,F_{m-n}\times F_{n+1})\]

\(\text{According to the lemma above:}\)
\[(F_n, F_{n + 1}) = 1\]
\[\therefore (F_n, F_m) = (F_n, F_{m - n})\]

\(\text{All in all:}\)
\[\text{If }n<m,(F_n,F_m)=(F_n,F_{m-n})\]

\(\text{We can see that }, \text{the change rule of } n \text{ and } m \text{ meet Stein's Algorithm,so we finally discovered that }\)
\[(F_n,F_m)=F_{(n,m)}\]


\[\Large \text{That's all, thanks!}\]
\[\Huge Q\omega Q\]

The Proof of Fibonacci GCD的更多相关文章

  1. 【前端】Util.js-ES6实现的常用100多个javaScript简短函数封装合集(持续更新中)

    Util.js (持续更新中...) 项目地址: https://github.com/dragonir/Util.js 项目描述 Util.js 是对常用函数的封装,方便在实际项目中使用,主要内容包 ...

  2. 收集有用的 Javascript 片段

    内容目录 数组 arrayMax arrayMin chunk compact countOccurrences deepFlatten difference distinctValuesOfArra ...

  3. 关于斐波拉契数列(Fibonacci)

    斐波那契数列指的是这样一个数列 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233,377,610,987,1597,2584,4181,6765,10 ...

  4. 使用并行的方法计算斐波那契数列 (Fibonacci)

    更新:我的同事Terry告诉我有一种矩阵运算的方式计算斐波那契数列,更适于并行.他还提供了利用TBB的parallel_reduce模板计算斐波那契数列的代码(在TBB示例代码的基础上修改得来,比原始 ...

  5. (转)Fibonacci Tilings

    Fibonacci numbers {Fn, n ≥ 0} satisfy the recurrence relation (1) Fn+2 = Fn+1 + Fn, along with the i ...

  6. fibonacci数列的性质和实现方法

    fibonacci数列的性质和实现方法 1.gcd(fib(n),fib(m))=fib(gcd(n,m)) 证明:可以通过反证法先证fibonacci数列的任意相邻两项一定互素,然后可证n>m ...

  7. BZOJ 2813: 奇妙的Fibonacci

    2813: 奇妙的Fibonacci Time Limit: 20 Sec  Memory Limit: 512 MBSubmit: 497  Solved: 134[Submit][Status][ ...

  8. Codeforces 902D/901B - GCD of Polynomials

    传送门:http://codeforces.com/contest/902/problem/D 本题是一个数学问题——多项式整除. 对于两个整数a.b,求最大公约数gcd(a,b)的辗转相除法的函数如 ...

  9. 【学习笔记】关于最大公约数(gcd)的定理

    手动博客搬家: 本文发表于20181004 00:21:28, 原地址https://blog.csdn.net/suncongbo/article/details/82935140 结论1 \[\g ...

随机推荐

  1. 22 严格模式&this关键词&let&const

    严格模式: ECMA5后的新指令:"use strict" 它不算一条语句,而是一段文字表达式,更早版本的JavaScript会忽略它. 严格模式无法使用未声明的变量. 严格模式的 ...

  2. ANSYS 瞬态热分析---零件在水中冷却

    目录 1. 案例 2. APDL分析 1. 案例 一个温度为300℃的铜环和一个温度为200℃的铁环,放置到22℃的水中进行淬火.水桶为铁质的圆形.分析中忽略水的流动. 材料参数 热性能 铜 铁 水 ...

  3. C语言-数组与指针 字符与字符串

    1 字符与字符串:char c='a'而不能写出char c="a" //字符变量用单引号'',而字符串用双引号. 2 字符数组与字符指针的初始化: char s[10]={0}, ...

  4. 「NOI2009」植物大战僵尸

    「NOI2009」植物大战僵尸 传送门 这是一道经典的最大权闭合子图问题,可以用最小割解决(不会的可以先自学一下) 具体来说,对于这道题,我们对于两个位置的植物 \(i\) 和 \(j\) ,如果 \ ...

  5. How2J学习java-1、环境配置

    JDK环境变量配置分下载,配置,验证三个步骤. 一.首先需要到JDK下载网站下载所需的JDK版本可根据更新来定.主流的开发工具Idear下载. 1.首先看配置成功后的效果 点WIN键->运行(或 ...

  6. 解决fastjson反序列化日期0000-00-00失败的方案

    解决fastjson反序列化日期0000-00-00失败的方案 22 Jul 2016 一.案例场景复原 示例场景里涉及两个class:TestDemo.java, DateBeanDemo.java ...

  7. GlusterFS分布式文件系统概述

    一.GlusterFS概述 GlusterFS是一个开源的分布式文件系统,同时也是Scale-Out存储解决方案Gluster的核心,在存储数据方面有强大的横向扩展能力,通过扩展不同的节点可以支持PB ...

  8. Mybatis笔记一

    课程安排: mybatis和springmvc通过订单商品 案例驱动 第一天:基础知识(重点,内容量多) 对原生态jdbc程序(单独使用jdbc开发)问题总结 mybatis框架原理 (掌握) myb ...

  9. Legal High

    不让任何人承担责任,不想看的东西就回避, 但是,如果想夺回值得夸耀的生存方式,就必须看那些不愿意看的现实,必须带着身负重伤的觉悟前进,这才叫做战斗. 有怨言的话去坟墓里说,钱不是全部,钱就是你们向对手 ...

  10. MCM(矩阵链乘法)

    这是<算法导论>动态规划中的一个问题.问题简述如下:我们在求解矩阵相乘时通常会有一个最优括号方案来对矩阵进行顺序相乘,这样会减少大量的计算时间. 我们知道矩阵A.B相乘,只能是当矩阵A的列 ...