Simple application of generating function (Derivation of general term formula of Cattelan number and Fibonacci sequence)

Solving Recurrence Relations Using Generating Functions

simple example

Generating function is a kind of infinite sequence f 0 , f 1 , f 2 , f 3 , . . . f_{0},f_{1},f_{2},f_{3},...f0,f1,f2,f3,. . . Representation of coefficients expressed as a power sequence:

F ( x ) = f 0 + f 1 x + f 2 x 2 + f 3 x 3 + . . . F(x)=f_{0}+f_{1}x+f_{2}x^{2}+f_{3}x^{3}+... F(x)=f0+f1x+f2x2+f3x3+...

For example, the sequence 1 , 1 , 1 , . . . 1,1,1,...1,1,1,. . . , we use the generating function to express:

G ( x ) = 1 + x + x 2 + x 3 + . . . G(x)=1+x+x^{2}+x^{3}+... G(x)=1+x+x2+x3+...

We can calculate its simple expression:

G ( x ) = 1 + x + x 2 + x 3 + . . . − x G ( x ) = − x − x 2 − x 3 − . . . } ⇒ G ( x ) − x G ( x ) = 1 ⇒ G ( x ) = ∑ i = 0 ∞ x i = 1 1 − x \left.\begin{array}{rcl} G(x)=1+x+x^{2}+x^{3}+...\\ -xG(x)=-x-x^{2}-x^{3}-...\end{array}\right\}\Rarr G(x)-xG(x)=1\Rarr G(x)=\sum\limits^{\infty}_{i=0}x^{i}=\frac{1}{1-x} G(x)=1+x+x2+x3+...xG(x)=xx2x3...}G(x)xG(x)=1G(x)=i=0xi=1x1

Another example is the sequence of positive integers 1 , 2 , 3 , 4 , . . . 1,2,3,4,...1,2,3,4,. . . , expressed as a generating function:

N ( x ) = 1 + 2 x + 3 x 2 + 4 x 3 + . . . + ( n + 1 ) x n N(x)=1+2x+3x^{2}+4x^{3}+...+(n+1)x^{n} N(x)=1+2x _+3x _2+4x _3+...+(n+1)xn

We use the same method to obtain its simple expression:

N ( x ) − x N ( x ) = 1 + x + x 2 + x 3 + . . . = G ( x ) = 1 1 − x N(x)-xN(x)=1+x+x^{2}+x^{3}+...=G(x)=\frac{1}{1-x} N(x)xN(x)=1+x+x2+x3+...=G(x)=1x1

therefore:

N ( x ) = ∑ i = 0 ∞ ( i + 1 ) x i = 1 ( 1 − x ) 2 N(x)=\sum\limits^{\infty}_{i=0}(i+1)x^i=\frac{1}{(1-x)^{2}} N(x)=i=0(i+1)xi=(1x)21

Commonly used generating functions

1 1 − a x = ∑ i = 0 ∞ a i x i = 1 + a x + a 2 x 2 + a 3 x 3 + . . . \frac{1}{1-ax}=\sum\limits^{\infty}_{i=0}a^{i}x^{i}=1+ax+a^{2}x^{2}+a^{3}x^{3}+... 1ax1=i=0aixi=1+ax+a2x _2+a3x _3+...

1 1 − x r = ∑ i = 0 ∞ x r i = 1 + x r + x 2 r + x 3 r + . . . \frac{1}{1-x^{r}}=\sum\limits^{\infty}_{i=0}x^{ri}=1+x^{r}+x^{2r}+x^{3r}+... 1xr1=i=0xri=1+xr+x2 r+x3 r+...

1 ( 1 − x ) n = ∑ i = 0 ∞ ( n + i − 1 i ) x i = 1 + ( n 1 ) x + ( n + 1 2 ) x 2 + ( n + 2 3 ) x 3 + . . . \frac{1}{(1-x)^{n}}=\sum\limits^{\infty}_{i=0}\left(\begin{array}{c}n+i-1\\i\end{array}\right)x^{i}=1+\left(\begin{array}{c}n\\1\end{array}\right)x+\left(\begin{array}{c}n+1\\2\end{array}\right)x^{2}+\left(\begin{array}{c}n+2\\3\end{array}\right)x^{3}+... (1x)n1=i=0(n+i1i)xi=1+(n1)x+(n+12)x2+(n+23)x3+...

arithmetic progression

Generating functions can be used to find general term formulas for sequences of numbers. For example, a 0 a_{0}a0as the first item, ddd is the tolerance arithmetic sequencea 0 , a 1 , a 2 , a 3 , . . . a_{0},a_{1},a_{2},a_{3},...a0,a1,a2,a3,. . . . Let's first write its generating function:

A ( x ) = a 0 + a 1 x + a 2 x 2 + a 3 x 3 + . . . A(x)=a_{0}+a_{1}x+a_{2}x^{2}+a_{3}x^{3}+... A(x)=a0+a1x+a2x2+a3x3+...

Then misplaced and subtracted:

A ( x ) − x A ( x ) = a 0 + d ( x + x 2 + x 3 + . . . ) = a 0 + x d 1 − x = a 0 − d + d 1 − x A(x)-xA(x)=a_{0}+d(x+x^{2}+x^{3}+...)=a_{0}+\frac{xd}{1-x} = a_{0}-d+\frac{d}{1-x} A(x)x A ( x )=a0+d(x+x2+x3+...)=a0+1xxd=a0d+1xd

so:

A ( x ) = a 0 − d 1 − x + d ( 1 − x ) 2 = ∑ i = 0 ∞ ( ( a 0 − d ) + ( i + 1 ) d ) x i = ∑ i = 0 ∞ ( a 0 + i d ) x i A(x)=\frac{a_{0}-d}{1-x}+\frac{d}{(1-x)^{2}}=\sum\limits^{\infty}_{i=0}((a_{0}-d)+(i+1)d)x^i=\sum\limits^{\infty}_{i=0}(a_{0}+id)x^i A(x)=1xa0d+(1x)2d=i=0((a0d)+(i+1)d)xi=i=0(a0+id)xi

From this we can derive the general term formula of the arithmetic sequence as:

a n + 1 = a 0 + n d a_{n+1}=a_{0}+nd an+1=a0+nd

Fibonacci sequence

The general term formula of the Fibonacci sequence can also be obtained by this method. We know the recurrence relationship of the Fibonacci sequence:

f 0 = 0 f_{0}=0 f0=0

f 1 = 1 f_{1}=1 f1=1

f n = f n − 1 + f n − 2 , n ⩽ 2 f_{n} = f_{n-1}+f_{n-2}, n \leqslant 2 fn=fn1+fn2,n2

Its generating function is denoted as:

F ( x ) = f 0 + f 1 x + f 2 x 2 + f 3 x 3 + . . . + f n x n + . . . F(x)=f_{0}+f_{1}x+f_{2}x^{2}+f_{3}x^{3}+...+f_{n}x^{n}+... F(x)=f0+f1x+f2x2+f3x3+...+fnxn+...

Dislocation subtraction:

F ( x ) − x F ( x ) − x 2 F ( x ) F(x)-xF(x)-x^{2}F(x) F(x)xF(x)x2F(x)

= f 0 + ( f 1 − f 0 ) x + ( f 2 − f 1 − f 0 ) x 2 + . . . + ( f n − f n − 1 − f n − 2 ) x n + . . . =f_{0}+(f_{1}-f_{0})x+(f_{2}-f_{1}-f_{0})x^{2}+...+(f_{n}-f_{n-1}-f_{n-2})x^{n}+... =f0+(f1f0)x+(f2f1f0)x2+...+(fnfn1fn2)xn+...

= 0 + 1 x + 0 x 2 + . . . + 0 x n + . . . =0+1x+0x^{2}+...+0x^{n}+... =0+1x+0x _2+...+0x _n+...

= x =x =x

then:

F ( x ) = x 1 − x − x 2 F(x)=\frac{x}{1-x-x^{2}} F(x)=1xx2x

This is a partial fraction. We know that partial fractions can be decomposed. The theorem is as follows:

Let p ( x ) p(x)p ( x ) is the degree less thannnPolynomials of n , α 1 , α 2 , . . . , α n \alpha_{1},\alpha_{2},...,\alpha_{n}a1,a2,...,anare different non-zero numbers, then there are constants c 1 , c 2 , . . . , cn c_{1},c_{2},...,c_{n}c1,c2,...,cnmake

p ( x ) ( 1 − α 1 x ) ( 1 − α 2 x ) . . . ( 1 − α nx ) = c 1 1 − α 1 x + c 2 1 − α 2 x + . . . + cn 1 − α nx \frac{p(x)}{(1-\alpha_{1}x)(1-\alpha_{2}x)...(1-\alpha_{n}x)}= \frac{c_{1}}{1-\alpha_{1}x}+\frac{c_{2}}{1-\alpha_{2}x}+...+\frac{c_{n}} {1-\alpha_{n}x}( 1 a1x ) ( 1 a2x ) . . . ( 1 anx)p(x)=1 a1xc1+1 a2xc2+...+1 anxcn

For x 1 − x − x 2 \frac{x}{1-xx^{2}}1xx2x, we assume it can be decomposed into c 1 1 − α 1 x + c 2 1 − α 2 x \frac{c_{1}}{1-\alpha_{1}x}+\frac{c_{2}} {1-\alpha_{2}x}1 a1xc1+1 a2xc2。由1 − x − x 2 = ( 1 − α 1 x ) ( 1 − α 2 x ) 1-xx^{2}=(1-\alpha_{1}x)(1-\alpha_{2}x )1xx2=(1a1x)(1a2x ) , can solveα 1 = 1 + 5 2 , α 2 = 1 − 5 2 \alpha_{1}=\frac{1+\sqrt{5}}{2},\alpha_{2}=\frac {1-\sqrt{5}}{2}a1=21+5 ,a2=215 。又由 x 1 − x − x 2 = c 1 1 − α 1 x + c 2 1 − α 2 x \frac{x}{1-x-x^{2}}=\frac{c_{1}}{1-\alpha_{1}x}+\frac{c_{2}}{1-\alpha_{2}x} 1xx2x=1 a1xc1+1 a2xc2The general score is x = c 1 ( 1 − α 2 x ) + c 2 ( 1 − α 1 x ) x=c_{1}(1-\alpha_{2}x)+c_{2}(1-\alpha_ {1}x)x=c1(1a2x)+c2(1a1x ) . Substitute intoα 1 \alpha_{1}a1and α 2 \alpha_{2}a2Get c 1 = 1 5 , c 2 = − 1 5 c_{1}=\frac{1}{\sqrt{5}},c_{2}=-\frac{1}{\sqrt{5}}c1=5 1,c2=5 1。所以 x 1 − x − x 2 = 1 5 ( 1 1 − α 1 x − 1 1 − α 2 x ) \frac{x}{1-x-x^{2}}=\frac{1}{\sqrt{5}}(\frac{1}{1-\alpha_{1}x}-\frac{1}{1-\alpha_{2}x}) 1xx2x=5 1(1 a1x11 a2x1)

Root generalized generating function 1 1 − ax = ∑ i = 0 ∞ aixi = 1 + ax + a 2 x 2 + a 3 x 3 + . . . \frac{1}{1-ax}=\sum\limits^ {\infty}_{i=0}a^{i}x^{i}=1+ax+a^{2}x^{2}+a^{3}x^{3}+...1ax1=i=0aixi=1+ax+a2x _2+a3x _3+. . . , we can write the above fraction as:

x 1 − x − x 2 = 1 5 ( ( 1 + α 1 x + α 1 2 x 2 + α 1 3 x 3 + . . . ) − ( 1 + α 2 x + α 2 2 x 2 + α 2 3 x 3 + . . . ) ) = ∑ i = 0 ∞ α 1 i − α 2 i 5 x i \frac{x}{1-x-x^{2}}=\frac{1}{\sqrt{5}}((1+\alpha_{1}x+\alpha_{1}^{2}x^{2}+\alpha_{1}^{3}x^{3}+...)-(1+\alpha_{2}x+\alpha_{2}^{2}x^{2}+\alpha_{2}^{3}x^{3}+...))=\sum\limits^{\infty}_{i=0}\frac{\alpha_{1}^{i}-\alpha_{2}^{i}}{\sqrt{5}}x^i 1xx2x=5 1((1+a1x+a12x2+a13x3+...)(1+a2x+a22x2+a23x3+...))=i=05 a1ia2ixi

So we solve the general term formula of the Fibonacci sequence as:

f n = 1 5 ( ( 1 + 5 2 ) n − ( 1 − 5 2 ) n ) f_{n}=\frac{1}{\sqrt{5}}((\frac{1+\sqrt{5}}{2})^{n}-(\frac{1-\sqrt{5}}{2})^{n}) fn=5 1((21+5 )n(215 )n)

Other examples of linear recurrence with constant coefficients

Next, we use generating functions to solve a practical mathematical problem:

Topic: For numbers from 0 ∼ 9 0\sim90A code string composed of 9 , it must contain an even number of 0, the length isnnHow many code strings of n are there?

answer:

We set the length to be nThe code string of n hasan a_{n}anindivual.

when n = 1 n=1n=1 , the code string "1 11 ", code string "2 22 ", ..., code string "9 99 "are symbolic conditions. Soa 1 = 9 a_{1}=9a1=9

when n = kn=kn=When k , the legal code string hasak a_{k}ak, add a non-zero number behind it, and get a legal code string; the illegal code string has 1 0 k − ak 10^{k}-a_{k}10kak, add a 0 behind it to get a legal code string. Then we get the recursive relation:

a k + 1 = 9 a k + ( 1 0 k − a k ) = 8 a k + 1 0 k a_{k+1}=9a_{k}+(10^{k}-a_{k})=8a_{k}+10^{k} ak+1=9a _k+(10kak)=8ak+10k

Our order G ( x ) = ∑ i = 0 ∞ aixi G(x)=\sum\limits^{\infty}_{i=0}a_{i}x^{i}G(x)=i=0aixi generates a function for it, then there are:

G ( x ) − 1 G(x)-1 G(x)1

= ∑ i = 1 ∞ a i x i =\sum\limits^{\infty}_{i=1}a_{i}x^{i} =i=1aixi

= ∑ i = 0 ∞ a i + 1 x i + 1 =\sum\limits^{\infty}_{i=0}a_{i+1}x^{i+1} =i=0ai+1xi+1

= ∑ i = 0 ∞ 8 a i x i + 1 + ∑ i = 0 ∞ 1 0 i x i + 1 =\sum\limits^{\infty}_{i=0}8a_{i}x^{i+1}+\sum\limits^{\infty}_{i=0}10^{i}x^{i+1} =i=08aixi+1+i=010ixi+1

= 8 x ∑ i = 0 ∞ a i x i + x ∑ i = 0 ∞ 1 0 i x i =8x\sum\limits^{\infty}_{i=0}a_{i}x^{i}+x\sum\limits^{\infty}_{i=0}10^{i}x^{i} =8 xi=0aixi+xi=010ixi

= 8 x G ( x ) + x 1 − 10 x =8xG(x)+\frac{x}{1-10x} =8xG(x)+110xx

From this we get:

G ( x ) = 1 2 ( 1 1 − 8 x + 1 1 − 10 x ) G(x)=\frac{1}{2}(\frac{1}{1-8x}+\frac{1}{1-10x}) G(x)=21(1−8x _ _ _1+110x1)

so:

G ( x ) = ∑ i = 0 ∞ 8 i + 1 0 i 2 x i G(x)=\sum\limits^{\infty}_{i=0}\frac{8^{i}+10^{i}}{2}x^{i} G(x)=i=028i+10ixi

a n a_{n} anThe general formula for is:

a n = 1 2 ( 8 n + 1 0 n ) a_{n}=\frac{1}{2}(8^{n}+10^{n}) an=21(8n+10n)

Cattelan number

We can use generating functions to solve some problems with more complex recurrence relations. For example, the Catalan number we will discuss below. We introduce it from a practical problem.

Railway platforms are often designed as a last in first out (LIFO) stack structure. As shown below:

------------------------------------------------------------


<-出站               <------+     +---------         <-进站
                            |    |
                            |    | 
------------------------+   |    |   +----------------------
                        |   |    |   |
                        |   |    V   |
                        |            |
                        |            |
                        |     站     |
                        |            |
                        |     台     |
                        |            |

When the trains numbered 1, 2, 3, and 4 enter the station in sequence (only entering and leaving), the situation on the platform is as follows:

--------------------------------------------------------------


<-出站               <------+        +---------       <-进站
                            |       |
                            |       | 
------------------------+   |       |   +-------------------
                        |   |       |   |
                        |   |  站   V   |
                        |      台       |
                        |               |
                        | +-----------+ |
                        | |           | |
                        | |     4     | |
                        | |           | |
                        | +-----------+ |
                        |               |
                        | +-----------+ |
                        | |           | |
                        | |     3     | |
                        | |           | |
                        | +-----------+ |
                        |               |
                        | +-----------+ |
                        | |           | |
                        | |     2     | |
                        | |           | |
                        | +-----------+ |
                        |               |
                        | +-----------+ |
                        | |           | |
                        | |     1     | |
                        | |           | |
                        | +-----------+ |

Now these trains will leave the station, obviously only the train 4 closest to the outside of the platform can leave the station. Other trains can only leave the station after train 4 leaves the station. Because train 4 is the last to enter the station and must be the first to exit, the platform model is a last-in-first-out stack structure.

now have nnn trains, numbered1 , 2 , . . . , n 1,2,...,n1,2,...,n . They enter an empty platform in sequence (fori < j i<ji<j , trainiii must be greater than trainjjj advances to the station), and then leaves the station (as long as there is a train in the station, the train closest to the station can choose to leave the station, even at this timennn trains have not all entered the station). For example, when dispatching a train for the first time, the station is empty at this time, and no train can leave the station, so only train1 11 pit stop. When dispatching trains for the second time, you can choose train1 11 out of the station, you can also choose the train2 22 pit stops. If the second dispatch selects train2 22 enters the station, then when dispatching the train for the third time, you can choose train2 22 out of the station, you can also choose the train3 33 pit stops. And so on.

We define a sequence called "outbound sequence", which is obtained by arranging the numbers of outbound trains in the order of outbound trains. With n = 3 n=3n=3 o'clock for example. First let 3 trains enter the station in sequence, and then let the 3 trains in the station leave the station in sequence, then the order of the trains leaving the station is: train3 33 , train2 22 , train1 11 . The outbound sequence for this scheduling method is321 3213 2 1 . Another example is this scheduling: train1 11 into the station, train2 22 into the station, train2 22 exits, train3 33 into the station, train3 33 exits, train1 11 outbound. Its outbound sequence is231 2312 3 1 . In fact, each time the train is dispatched, either the train with the smallest number among the trains that have not entered the station is selected to enter the station, or the train with the largest number in the station is selected to leave the station. So we describe the scheduling process like this: inbound, inbound, outbound, inbound, outbound, outbound. This also determines the outbound sequence as231 231231

Now we ask the question: for nnHow many different outbound sequences are there for n incoming trains?

Answer: We remember that for nnThere are n trains to enter the station, a total ofcn c_{n}cndifferent outbound sequences.

Obviously c 1 = 1 c_{1}=1c1=1

For 1 ⩽ k ⩽ n 1\leqslant k\leqslant n1kn , we assume trainkkk is the last to leave the station. Because the train with the largest number in the station leaves the station every time it leaves the station, so in the trainkkBefore k enters the station, all trains with numbers smaller than it have left the station. For thisk − 1 k-1k1 car, withck − 1 c_{k-1}ck1different outbound sequences. And for numbers than kkFor trains with large k , because every time they enter the station, the train with the smallest number outside the station enters the station, so these trains will be in the train kkStop after k . And because they will be on the trainiii leaves the station before leaving the station, so these trains havecn − k c_{nk}cnkoutbound sequence. We might as well let c 0 = 1 c_{0}=1c0=1 , then there is a recursive formula:

c n = ∑ k = 1 n c k − 1 c n − k = ∑ k = 0 n − 1 c k c n − k − 1 c_{n}=\sum\limits_{k=1}^{n}c_{k-1}c_{n-k}=\sum\limits_{k=0}^{n-1}c_{k}c_{n-k-1} cn=k=1nck1cnk=k=0n1ckcnk1

This cn c_{n}cnIt is the Cattelan number. Now we want to use its recursive formula to find its general term formula.

We remember its generating function as C ( x ) = ∑ i = 0 ∞ cixi C(x)=\sum\limits_{i=0}^{\infty}c_{i}x^{i}C(x)=i=0cixi , and square it to get:

C 2 ( x ) C^{2}(x) C2(x)

= c 0 2 + ( c 0 c 1 + c 1 c 0 ) x + ( c 0 c 2 + c 1 c 1 + c 2 c 0 ) x 2 + . . . =c_{0}^{2}+(c_{0}c_{1}+c_{1}c_{0})x+(c_{0}c_{2}+c_{1}c_{1}+c_{2}c_{0})x^{2}+... =c02+(c0c1+c1c0)x+(c0c2+c1c1+c2c0)x2+...

= ∑ i = 0 ∞ ( ∑ k = 0 i c k c i − k ) x i =\sum\limits_{i=0}^{\infty}(\sum\limits_{k=0}^{i}c_{k}c_{i-k})x^{i} =i=0(k=0ickcik)xi

= ∑ i = 0 ∞ c i + 1 x i =\sum\limits_{i=0}^{\infty}c_{i+1}x^{i} =i=0ci+1xi

Dislocation subtraction:

C ( x ) − x C 2 ( x ) = c 0 = 1 C(x)-xC^{2}(x)=c_{0}=1 C(x)xC2(x)=c0=1

Solving the equation gives:

C ( x ) = 1 ± 1 − 4 x 2 x C(x)=\frac{1\pm\sqrt{1-4x}}{2x} C(x)=2x _1±1 4 x

We are now going to drop a root. From C ( x ) C(x)The definition of C ( x ) is lim ⁡ x → 0 C ( x ) = c 0 = 1 \lim\limits_{x\rarr0}C(x)=c_{0}=1x0limC(x)=c0=1

lim ⁡ x → 0 C ( x ) = lim ⁡ x → 0 1 ± 1 − 4 x 2 x = lim ⁡ x → 0 2 1 ∓ 1 − 4 x \lim\limits_{x\rarr0}C(x)=\lim\limits_{x\rarr0}\frac{1\pm\sqrt{1-4x}}{2x}=\lim\limits_{x\rarr0}\frac{2}{1\mp\sqrt{1-4x}} x0limC(x)=x0lim2x _1±1 4 x =x0lim11 4 x 2

So discard the solution with positive sign and get C ( x ) C(x)C(x)

C ( x ) = 1 − 1 − 4 x 2 x C(x)=\frac{1-\sqrt{1-4x}}{2x} C(x)=2x _11 4 x

Next continue to calculate C ( x ) C(x)C ( x ) , we need to use the binomial theorem:

( 1 + x ) α = 1 + ∑ i = 1 ∞ ( α i ) x i , α ∈ R (1+x)^{\alpha}=1+\sum\limits_{i=1}^{\infty}\left(\begin{array}{rcl}\alpha \\ i\end{array}\right)x^{i}, \alpha\in R (1+x)a=1+i=1(ai)xi,aR

Let α = 1 2 \alpha=\frac{1}{2}a=21, substitute into the above formula to calculate C ( x ) C(x)C(x)

C ( x ) C(x) C(x)

= 1 − 1 − 4 x 2 x =\frac{1-\sqrt{1-4x}}{2x} =2x _11 4 x

= 1 2 x ( 1 − ( 1 + ∑ i = 1 ∞ ( 1 2 i ) ( − 4 x ) i ) ) =\frac{1}{2x}(1-(1+\sum\limits_{i=1}^{\infty}\left(\begin{array}{c}\frac{1}{2} \\ i\end{array}\right)(-4x)^{i})) =2x _1(1(1+i=1(21i)( −4 x ) _i))

= 2 ∑ i = 1 ∞ ( 1 2 i ) ( − 4 x ) i − 1 =2\sum\limits_{i=1}^{\infty}\left(\begin{array}{c}\frac{1}{2} \\ i\end{array}\right)(-4x)^{i-1} =2i=1(21i)( −4 x ) _i1

= 2 ∑ i = 1 ∞ 1 2 ⋅ ( − 1 2 ) ⋅ ( − 3 2 ) ⋅ . . . ⋅ ( 3 2 − i ) i ! ( − 4 x ) i − 1 =2\sum\limits_{i=1}^{\infty}\frac{\frac{1}{2}\cdot(-\frac{1}{2})\cdot(-\frac{3}{2})\cdot...\cdot(\frac{3}{2}-i)}{i!}(-4x)^{i-1} =2i=1i!21(21)(23)...(23i)( −4 x ) _i1

= 1 + ∑ i = 2 ∞ 1 2 ⋅ 3 2 ⋅ . . . ⋅ ( i − 3 2 ) i ! ( 4 x ) i − 1 =1+\sum\limits_{i=2}^{\infty}\frac{\frac{1}{2}\cdot\frac{3}{2}\cdot...\cdot(i-\frac{3}{2})}{i!}(4x)^{i-1} =1+i=2i!2123...(i23)( 4x ) _i1

= 1 + ∑ i = 2 ∞ ( 2 i − 3 ) ! ! i ! ( 2 x ) i − 1 =1+\sum\limits_{i=2}^{\infty}\frac{(2i-3)!!}{i!}(2x)^{i-1} =1+i=2i!( 2 i 3 ) ! !( 2x ) _i1

= 1 + ∑ i = 2 ∞ ( 2 i − 3 ) ! ! i ! ⋅ ( 2 i − 2 ) ! ! ( i − 1 ) ! x i − 1 =1+\sum\limits_{i=2}^{\infty}\frac{(2i-3)!!}{i!}\cdot\frac{(2i-2)!!}{(i-1)!}x^{i-1} =1+i=2i!( 2 i 3 ) ! !(i1)!( 2 i 2 ) ! !xi1

= 1 + ∑ i = 2 ∞ ( 2 i − 2 ) ! i ! ( i − 1 ) ! x i − 1 =1+\sum\limits_{i=2}^{\infty}\frac{(2i-2)!}{i!(i-1)!}x^{i-1} =1+i=2i!(i1)!( 2 i 2 ) !xi1

= ∑ i = 1 ∞ ( 2 i − 2 ) ! i ! ( i − 1 ) ! x i − 1 =\sum\limits_{i=1}^{\infty}\frac{(2i-2)!}{i!(i-1)!}x^{i-1} =i=1i!(i1)!( 2 i 2 ) !xi1

= ∑ i = 0 ∞ ( 2 i ) ! ( i + 1 ) ! i ! x i =\sum\limits_{i=0}^{\infty}\frac{(2i)!}{(i+1)!i!}x^{i} =i=0(i+1)!i!( 2i ) ! _xi

= ∑ i = 0 ∞ ( 2 i i ) i + 1 x i =\sum\limits_{i=0}^{\infty}\frac{\left(\begin{array}{c}2i \\ i\end{array}\right)}{i+1}x^{i} =i=0i+1(2i _i)xi

所以 c n = ( 2 n n ) n + 1 c_{n}=\frac{\left(\begin{array}{c}2n \\ n\end{array}\right)}{n+1} cn=n+1(2 nn), the outbound sequence has ( 2 nn ) n + 1 \frac{\left(\begin{array}{c}2n \\ n\end{array}\right)}{n+1}n+1(2 nn)kind.

c n = ( 2 n n ) n + 1 c_{n}=\frac{\left(\begin{array}{c}2n \\ n\end{array}\right)}{n+1} cn=n+1(2 nn), which is the general term formula of Cattelan number.

reference article

Part of Cattelan numbers: Deriving the general term formula of Cattelan numbers with generating functions

Guess you like

Origin blog.csdn.net/Qmj2333333/article/details/114013000