Simple Search:

An Estimation of Exponential Sums Associated with a Cubic Form Polynomial


Heng, Swee Huay (1999) An Estimation of Exponential Sums Associated with a Cubic Form Polynomial. Masters thesis, Universiti Putra Malaysia.

Abstract / Synopsis

The method of exponential sums is one of a few general methods enabling us to solve a wide range of miscellaneous problems from the theory of numbers. The main problem of the theory of exponential sums is to obtain an upper estimate of the modulus of an exponential sum as sharp as possible. Investigation on the sums when f is a two-variable polynomial is studied using the Newton polyhedron technique. One of the methods to obtain the estimate for the above exponential sums is to consider the cardinality of the set of solutions to congruence equations modulo a prime power. A closer look on the actual cardinality on the following polynomial in a cubic form f(x,y) = ax3 + bxi + cx + dy + e has been carried out using the Direct Method with the aid of Mathematica. We reveal that the exact cardinality is much smaller in comparison with the estimation. The necessity to find a more precise estimate arises due to this big gap. By a theorem of Bezout, the number of common zeros of a pair of polynomials does not exceed the product of the degrees of both polynomials. In this research, we attempt to find a better estimate for cardinality by looking at the maximum number of common zeros associated with the partial derivatives fx(x,y) and fy(x,y). Eventually a sharper estimate of cardinality for the various conditions on the coefficients of f(x,y) can be determined and the estimate of S(f; p') obtained.

Download File

[img] PDF

Download (1MB)

Additional Metadata

Item Type: Thesis (Masters)
Subject: Exponential sums
Subject: Polynomials
Call Number: FSAS 1999 45
Chairman Supervisor: Ismail bin Abdullah, PhD
Divisions: Faculty of Science and Environmental Studies
Depositing User: Laila Azwa Ramli
Date Deposited: 11 Feb 2011 12:17
Last Modified: 26 Sep 2013 09:02
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item