Given a real number x ∈ [ 0 ; 1 ) with binary expansion x = ( 0 . a 1 a 2 a 3 … ) 2 , let f ( x ) = ( 0 . a 1 a 2 a 3 … ) 1 0 be the number obtained when interpreting the binary expansion of x as a decimal expansion.
For example, f ( 2 1 ) = f ( ( 0 . 1 0 0 … ) 2 ) = ( 0 . 1 0 0 … ) 1 0 = 1 0 1 .
Given that: I = 0 ∫ 1 f ( x ) d x = n m , where m , n are coprime positive integers.
Find the value of m + n .
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
How would you prove that the integral exists?
Log in to reply
You would just need to show that f is increasing and bounded on (0, 1).
Nice problem!
Let's see what f ( x ) is - Converting to binary involves running it through 2 k s and seeing where it becomes 1 or 0 and then dividing each term by its corresponding 1 0 − k s
So, we can write
f ( x ) = k = 1 ∑ ∞ 1 0 k ⌊ 2 k x ⌋ m o d 2
Now, we need to integrate
∫ 0 1 k = 1 ∑ ∞ 1 0 k ⌊ 2 k x ⌋ m o d 2 d x
k = 1 ∑ ∞ 1 0 k 1 ∫ 0 1 ⌊ 2 k x ⌋ m o d 2 d x
Claim : ∫ 0 1 ⌊ a k x ⌋ m o d a d x = 2 a − 1 where a is a positive integer.
Proof :
a k 1 ∫ 0 a k ⌊ u ⌋ m o d a d u
The integral can be done with the help of graph.
Graph for ⌊ u ⌋ is simple to draw. An increasing graph with critical points at integers. Now, graph for ⌊ u ⌋ m o d a is a periodic one with period a .
Hence, the area under such a graph for the interval ( 0 , a k ) would be -
Area for (0,a) ∗ a k − 1 and as in the given interval, there are a rectangles with width = 1 and length = 0 , 1 , 2 , ⋯ , a − 1 .
= a k − 1 m = 0 ∑ a − 1 m ∗ 1 = 2 a k ( a − 1 )
And there is a factor we've not considered, putting it back -
2 a − 1
Back to the integral, we apply this and get -
k = 1 ∑ ∞ 1 0 k 1 2 1
1 8 1
Problem Loading...
Note Loading...
Set Loading...
We split this integral at x = 2 1 to get: I = 0 ∫ 2 1 f ( x ) d x + 0 ∫ 2 1 f ( 2 1 + x ) d x
Now, if 0 ≤ x < 2 1 then x has the form x = ( 0 . 0 a 2 a 3 … ) 2 , and
f ( x ) = ( 0 . 0 a 2 a 3 … ) 1 0 = 1 0 1 ( 0 . a 2 a 3 … ) 1 0
= 1 0 1 f ( ( 0 . a 2 a 3 … ) 2 ) = 1 0 1 f ( 2 x )
f ( 2 1 + x ) = ( 0 . 1 a 2 a 3 … ) 1 0 = 1 0 1 + 1 0 1 ( 0 . a 2 a 3 … ) 1 0
= 1 0 1 + 1 0 1 f ( ( 0 . a 2 a 3 … ) 2 ) = 1 0 1 + 1 0 1 f ( 2 x )
Hence,
I = 1 0 1 0 ∫ 2 1 ( f ( 2 x ) + 1 + f ( 2 x ) ) d x = 2 0 1 0 ∫ 1 ( 1 + 2 f ( y ) ) d y = 2 0 1 + 1 0 I
Thus, I = 1 8 1 .
So, m + n = 1 9 .