Homepage Solution manuals Ivan Niven An Introduction to the Theory of Numbers Exercise 1.2.44 (Binary decomposition of positive integers)

Exercise 1.2.44 (Binary decomposition of positive integers)

Prove that every positive integer is uniquely expressible in the form

2 j 0 + 2 j 1 + 2 j 2 + + 2 j m

where m 0 and 0 j 0 < j 1 < j 2 < < j m .

Answers

Proof. This is the usual binary representation of a positive integer.

We reason by induction to prove the existence of such a representation.

First 1 = 2 0 = j A 2 j , where A = { 0 } .

Now let n > 1 be any integer, and assume that every positive integer k < n is expressible in the form 2 j 0 + 2 j 1 + 2 j 2 + + 2 j m where m 0 and 0 j 0 < j 1 < j 2 < < j m . In others words, if we denote 𝒫 f ( ) the set of finite subsets of , we assume P ( n ) , where

P ( n ) k [ [ 1 , n ] ] , A k 𝒫 f ( ) { } , k = j A k 2 j .

(If we take A k 𝒫 f ( ) , 0 is represented.)

We prove that n itself is so represented. The division of n by 2 gives

n = 2 m + r , r { 0 , 1 } .

Then 0 < m n 2 < n , so that we may apply the induction hypothesis to m :

m = j A m 2 j ,

for some non empty finite set A m . Then

n = r + 2 j A m 2 j = r + j A m 2 j + 1 = r + k B m 2 k ,

where B m = { k k 1 A m } . Note that B m { 0 } .

Now define A n = B m if r = 0 , and A n = B m { 0 } if r = 1 . Then

n = j A n 2 j .

This proves P ( n + 1 ) , and the induction is done. Therefore every positive integer n is of the form

n = j A 2 j

for some non empty subset A . If A = { j 0 , j 1 , , j m } ,

n = 2 j 0 + 2 j 1 + 2 j 2 + + 2 j m , 0 j 0 < j 1 < < j m , m 0 .

Now we prove the unicity of such a representation. Suppose that

n = j A 2 j = j B 2 j ,

where A , B are non empty finite subsets of . if A = { j 1 , j 2 , , j m } and B = { k 1 , k 2 , , k l } , this gives

n = 2 j 0 + 2 j 1 + 2 j 2 + + 2 j m = 2 k 0 + 2 k 1 + 2 j 2 + + 2 k l ,

where m 0 , 0 j 0 < j 1 < j 2 < < j m and l 0 , 0 k 0 < k 1 < j 2 < < k l .

Without loss of generality, we may suppose that m l . Assume, for the sake of contradiction, that there is some i [ [ 0 , m ] ] such that j i k i , and let I be the first index for which j I k I . Then

2 j I + + 2 j m = 2 k I + + 2 k l .

If j I < k I , then all terms except 2 j I are divisible by 2 j I + 1 , so that 2 j I + 1 2 j I , and 2 1 . This is a contradiction, and we obtain a similar contradiction if j I > k I . Therefore j 0 = k 0 , j 1 = k 1 , , j m = k m . If m < l , we obtain after simplification 0 < 2 k m + 1 + + 2 k l = 0 : this is absurd, so m = l and A = B . The two representations are identical. □

User profile picture
2024-10-02 09:42
Comments
  • Hi Richard, if you want me to delete the comments once the erratas have been discovered, let me know or you can do it yourself to make the final draft without comments. Glad to help. I will work through them all, but I will work slowly. Today I noticed here that the set A=\{j_{1}..etc\} should begin with subscript 0. Same with set B. Also after the phrase "after simplification" the subscript should be k_{m+1} instead of k_{m}+1.
    BretSherfinski2025-02-27
  • Hi Bret. Errors corrected. I don't see any point in destroying the comments: they are part of the history of each exercise. Even if I am a little ashamed of the remaining errors, I know that even a careful rereading on my part would not be enough: I had reread this solution without seeing anything! I regretted until now the lack of reactions to these solutions, and I am happy that an exchange is taking place through these comments. I myself had made some remarks to solutions of other people on this site, unfortunately often remaining unanswered. I will remain attentive to all your suggestions, and I will respond to them regularly. Thank you.
    richardganaye2025-03-03