[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference rusure::math

Title:Mathematics at DEC
Moderator:RUSURE::EDP
Created:Mon Feb 03 1986
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:2083
Total number of notes:14613

1807.0. "Math Magazine 1429" by RUSURE::EDP (Always mount a scratch monkey.) Wed Oct 13 1993 13:24

    Proposed by Wee Liang Gan, student, Singapore.
    
    Let P be a point inside the convex n-gon A[1]A[2]...A[n].  Prove that
    at least one of the angles PA[i]A[i+1], i=1,2,...,n is less than or
    equal to (1/2-1/n)pi.  (All subscripts are taken modulo n.)
T.RTitleUserPersonal
Name
DateLines
1807.1RUSURE::EDPAlways mount a scratch monkey.Tue Oct 25 1994 17:0246
    Solution by The Javelina Problem Solvers, Texas A&M University,
    Kingsville, Texas.

    Let alpha denote the smallest of the angles angle PA[i]A[i+1] and let
    gamma[i]=angle A[i]PA[i+1].  For i=1, 2, ..., n, pick B[i+1] on PA[i+1]
    so that angle PA[i]B[i+1]=alpha (see figure).

    Then, using the law of sines,

		 PA[i]      PA[i]    sin(alpha+gamma[i])
		------- <= ------- = -------------------.
		PA[i+1]    PB[i+1]       sin(alpha)
    
    
               P
               /\   g=gamma[i]
              /g \
             /    \        /
     \      /      \      /
      \    /      ..*B[i+1]
       \  /a .....   \  /     a=alpha
    A[i]\/...______   \/
                   ---*A[i+1]
    
     
    It follows that

		       n     PA[i]             sin(alpha+gamma[i])   
		1 = product ------- <= product ------------------- <=
		            PA[i+1]                sin(alpha)        
    
		 sin(sum i=1..n of (alpha+gamma[i])/n)
		(-------------------------------------)^n =
		              sin(alpha)
    
		 sin(alpha+2pi/n)
		(----------------)^n.
		    sin(alpha)

    where, in the last inequality, we have used a special case of Jensen's
    inequality (see, for example, _Maxima and Minima Without Calculus_,
    Ivan Niven, MAA Dolciani Series, No. 6, 1981, pp. 92-95).

    Thus, sin alpha <= sin(alpha+2pi/n).  We conclude that alpha <= pi/2 -
    pi/n; if alpha > pi/2-pi/n, then sin(alpha+2pi/n) < sin(pi/2+pi/n) <
    sin alpha, a contradiction.