Define number theoretic function Λ by Λ(n)= log p if n=p^k for prime p and k > 1 or 0 otherwise
Let F(n) = Σ d/n (d divides n) Λ(d)
(a) Compute F(pq) where p and q are prime using def of F and Λ
(b) Compute F(p^3q^5) for p,q prime using def of F and Λ
(c) Show that F(n)=log n. Use prime decomposition for n
(d) Apply Mobius inversion formula to show that Λ(n) = Σd/n μ(n/d)log d = - Σd/n μ(d) log d
Just learning these types of problems and really confused on to how to get through this..
Copyright © 2024 1QUIZZ.COM - All rights reserved.
Answers & Comments
Verified answer
a) F(pq) = Σ(d|pq) Λ(d)
............= Λ(1) + Λ(p) + Λ(q) + Λ(pq)
............= 0 + log p + log q + 0
............= log(pq), via product rule of logs.
b) This is similar to part a, but with more factors (note that Λ = 0 for numbers with more than one prime factor).
(Answer: log(p^3 q^5).)
c) Writing n = (p1)^(a1) * ... * (pk)^(ak) as the prime factorization of n,
F(n) = Λ(1) + Σ(j=1 to k) [Λ(pj) + Λ((pj)^2) + ... + Λ((pj)^(aj))] + 0, since Λ(d) = 0 if
d has more than one prime in its prime factorization
.......= Λ(1) + Σ(j=1 to k) [Λ(pj) + Λ((pj)^2) + ... + Λ((pj)^(aj))]
.......= 0 + Σ(j=1 to k) [log(pj) + log(pj) + ... + log(pj)]
.......= Σ(j=1 to k) aj * log(pj)
.......= Σ(j=1 to k) log ((pj)^(aj)), via power rule for logs
.......= log ((p1)^(a1) * ... * (pk)^(ak)), via product rule for logs
.......= log n.
(d) Λ(n) = Σ(d|n) μ(n/d) log d, via part (c) and Mobius Inversion Formula
............= Σ(d|n) μ(d) log(n/d), by reversing the order of summation
............= Σ(d|n) μ(d) (log n - log d)
............= log n * Σ(d|n) μ(d) - Σ(d|n) μ(d) log d
............= log n * 0 - Σ(d|n) μ(d) log d, fact about μ
............= - Σ(d|n) μ(d) log d.
I hope this helps!
You probably mean Λ(n)= log p if n=p^k for prime p and k > 0 [i.e. if n is a prime or a power of a prime] And I assume you mean that F(n) is the sum of Λ(d) for all divisors d of n.
a) If n=pq, what are its divisors? Clearly, 1, p, q and n itself -- and nothing else.
So F(n) = Λ(1) + Λ(p) + Λ(q) + Λ(n) = 0 + logp + logq + 0 = log(pq) = logn
b) In this case, the divisors of n are 1, p, p², p³, q, qp, qp², qp³, q², q²p, etc. up to q⁵p³.
But of al those numbers, only for p, p², p³, q, q², q³, q⁴ and q⁵ is Λ different from 0.
So Λ(n) = Λ(p) + Λ(ρ²) +... + Λ(q⁴) + Λ(q⁵) = logp + logp + logp + logq + logq + logq + logq + logq =
= 3logp + 5logq = log(p³) + log(q⁵) = log(p³q⁵) = logn
c) In fact, F(n)=logn for all n. For, let n = p^a * q^b * r^c * s^d *... where p, q, r, s... are the different prime divisors of n and a, b, c... their exponents in the prime factorization of n. Then the only divisors of n that are powers of primes are p, p², p³,... p^a; q, q², q³, ... q^b; r, r², r³,... r^c; etc. The Λ of all other divisors of n is 0, and the Λ of the divisors just listed is logp, logp, logp... (a times), logq, logq, logq... (b times), logr, logr, logr... (c times) etc. So the sum of all Λ's is just alogp + blogq + clogr +... = log(p^a)+log(q^b)+log(r^c)+... = log(p^a*q^b*r^c*...) = logn.
d) Möbius' inversion formula states just this, that if F(n) is the sum of Λ(d) over all divisors d of n, then Λ(n) is the sum of μ(n/d)F(d) over all divisors d of n for ANY arithmetic functions Λ and F. Apply this to aour case, where F(n)=logn, to get Λ(n) = Σ μ(n/d)log d. But as d runs over all divisors of n, n/d also runs over all divisors of n; so this sum can also be written Σμ(d)log(n/d) = Σμ(d)(logn - logd) = (Σμ(d))logn - Σμ(d)logd = -Σμ(d)logd, since Σμ(d) taken over all divisors of n is 0 (unless n=1, in which case logn=0 !).
This last assertion is an immediate consequence of the fact that every non-empty finite set has just as many subsets with odd numbers of elements as subsets with even numbers of elements, which is in turn a consequence of the binomial expansion of (1-1)^n = 0
a million. Date one in each of them for an hour, i could somewhat spend an hour with them, then in trouble-free terms a million kiss. 2. e mail tackle, so i will communicate with them. 3. it rather is a confusing one, yet i think of i could somewhat have Nick say i'm somewhat. 4. Meet them and get hugged, that could desire to be remarkable! 5. Haha, i admire this question and the apparent answer is marry a Jonas. . . the two Nick or Joe will do, i'm no longer that choosy lol! stable activity, I enjoyed it! Cool Survey! i could like to make certain yet another! Peace Love Jonas! =]