Fibonacci数列 Huffman树

「已注销」

于 2022-06-03 13:48:05 发布

阅读量640

点赞数 2

文章标签：概率论

本文链接：https://blog.csdn.net/int_main_Roland/article/details/125112173

版权

本文深入探讨了Fibonacci数列，包括其通项公式、求和通项，以及Fibonacci分布的Huffman树。通过Laplace变换和Z变换解析了Fibonacci数列，并讨论了Huffman树的期望编码长度，揭示了其极限行为。此外，还介绍了Fibonacci数列的其他恒等式。

摘要由CSDN通过智能技术生成

Fibonacci数列

for k = 1:6
    k = int16(k);
    fprintf('f%d = %d\n', k, Fibonacci(k));
end

function fk = Fibonacci(k)
    assert(isinteger(k) && numel(k)==1 && k>=1);
    if k > 2
        %fk = Fibonacci(k-1) + Fibonacci(k-2);
        dp = ones(k,1);
        for i = 3:k
            dp(i) = dp(i-1) + dp(i-2);
        end
        fk = dp(k);
    else
        fk = 1;
    end
end

f1 = 1
f2 = 1
f3 = 2
f4 = 3
f5 = 5
f6 = 8

Fibonacci通项

最简单的证明的方法是使用数学归纳法.
$f_n = \frac{\phi^n - \psi^n}{\phi - \psi}$
$f_{n} + f_{n+1} = \frac{\phi^{n+1} - \psi^{n+1} + \phi^{n} - \psi^{n}}{\phi - \psi} = \frac{(\phi+1)\phi^{n} - (\psi+1)\psi^{n}}{\phi - \psi} = \frac{\phi^2 \phi^{n} - \psi^2 \psi^{n}}{\phi - \psi} = f_{n+2}$
$\phi = \frac{1+\sqrt{5}}{2}$ , $\psi = \frac{1-\sqrt{5}}{2}$ , $\phi + 1 = \phi^2$ , $\psi + 1 = \psi^2$ , 原因是他们是 $x^2-x-1=0$ 的根.

使用初等数学方法求解Fibonacci通项主要通过化归到求解等比数列及其求和的通项.

$f_{n} = f_{n-1} + f_{n-2}$
$g_{n-1} = f_{n} + B f_{n-1} = A (f_{n-1} + B f_{n-2}) = A g_{n-2}$
$g_{n-1} = f_{n} + \frac{-1-\sqrt{5}}{2} f_{n-1} = \frac{1-\sqrt{5}}{2} \left( f_{n-1} + \frac{-1-\sqrt{5}}{2} f_{n-2} \right) = \frac{1-\sqrt{5}}{2} g_{n-1}$
$g_{n-1} = f_{n} + \frac{-1+\sqrt{5}}{2} f_{n-1} = \frac{1+\sqrt{5}}{2} \left( f_{n-1} + \frac{-1+\sqrt{5}}{2} f_{n-2} \right) = \frac{1+\sqrt{5}}{2} g_{n-1}$
${g_n = A^n g_0\}$ 呈等比数列, 容易求解通项.
$f_n = (-B)^n \left( f_0 + \sum\limits_{k=1}^{n} \frac{g_{n-k}}{(-B)^{n-k+1}} \right) = (-B)^n \left( f_0 + \sum\limits_{k=1}^{n} \frac{A^{n-k} g_0}{(-B)^{n-k+1}} \right) = (-B)^n \left( f_0 + \frac{g_0}{-B} \sum\limits_{k=1}^{n} \left(\frac{A}{-B}\right)^{n-k} \right)$ 呈等比数列求和, 容易求解通项.

使用高等数学方法求解Fibonacci通项主要通过化归到频域的多项式真分式(proper polynomial fraction).

离散时间线性时不变系统
$x [n] = x [n - 1] + x [n - 2]$
连续化(上采样)
$\delta_T(t) = \sum\limits_{n=0}^{+\infty} x(t) \delta(t-nT) = \sum\limits_{n=0}^{+\infty} x(nT) \delta(t-nT)$
Laplace变换
$\mathcal{Z}\left[x[n]\right] = \mathcal{L}\left[x(t)\right] = \sum\limits_{n=0}^{+\infty} x(nT) e^{-snT} = \sum\limits_{n=0}^{+\infty} x[n] z^{-n}$
Z变换
$z^{-1} \left( X(z) - x[-1]z \right) + z^{-2} \left( X(z) - x[-1]z - x[-2]z^{2} \right) = \left( z^{-1} + z^{-2} \right) X(z) + A_0 + A_1 z^{-1}$