Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
Drexel University or The Math Forum.


Pavel
Posts:
2
Registered:
3/6/14


Re: last word
Posted:
Mar 6, 2014 1:02 PM


I was looking for the same thing, without much success.
The best paper I found is "ON VANISHING SUMS OF ROOTS OF UNITY" by Lam and Leung. They show that if n has at most two prime divisors then a vanishing sum of nth roots of unity indeed decomposes into the "obvious ones". In the paper, you can also find a counterexample for n having three prime divisors.
Even with many divisors, it is possible that a vanishing sum decomposes into "less obvious ones", but this doesn't seem to be known.



