Open Access

Retracted Article: On the fourth power mean of the two-term exponential sums

Journal of Inequalities and Applications20142014:116

https://doi.org/10.1186/1029-242X-2014-116

Received: 7 February 2014

Accepted: 6 March 2014

Published: 19 March 2014

The Retraction Note to this article has been published in Journal of Inequalities and Applications 2014 2014:344

Abstract

Abstract

The main purpose of this paper is using the analytic methods and the properties of Gauss sums to study the computational problem of one kind of fourth power mean of two-term exponential sums, and to give an interesting identity and asymptotic formula for it.

MSC: 11L05.

Keywords

the two-term exponential sumsfourth power meanGauss sumsidentityasymptotic formula

1 Introduction

Let q 3 be a positive integer. For any integers m and n, the two-term exponential sum C ( m , n , k ; q ) is defined as follows:
C ( m , n , k ; q ) = a = 1 q e ( m a k + n a q ) ,

where e ( y ) = e 2 π i y .

About the various properties of C ( m , n , k ; q ) , some authors had studied it, and they obtained a series of results, some related works can be found in references [15]. For example, Gauss’ classical work proved the remarkable formula (see [6])
C ( 1 , 0 , 2 ; q ) = 1 2 q ( 1 + i ) ( 1 + e ( q / 4 ) ) = { q , if  q 1 mod 4 , 0 , if  q 2 mod 4 , i q , if  q 3 mod 4 , ( 1 + i ) q , if  q 0 mod 4 ,

where i 2 = 1 .

Generally, for any odd number q and ( 2 m , q ) = 1 , the exact value of | C ( m , n , 2 ; q ) | is (e.g. see Berndt, Evans, Williams and Apostol’s related works). Cochrane and Zheng [2] show for the general sum that
| C ( m , n , k ; q ) | k ω ( q ) q 1 2 ,

where ω ( q ) denotes the number of all distinct prime divisors of q.

In this paper, we study the fourth power mean of the two-term exponential sum C ( m , n , k ; q ) as follows:
m = 1 q | C ( m , n , k ; q ) | 4 ,
(1.1)

where n is any integer with ( n , q ) = 1 .

As regards this problem, it seems that none has yet studied it, at least we have not seen any related result before. The problem is interesting, because it can reflect that the mean value of C ( m , n , k ; q ) is well behaved. The main purpose of this paper is to show this point. That is, we shall prove the following conclusion.

Theorem Let p > 3 be a prime. Then for any integer n with ( n , p ) = 1 , we have the identity
m = 1 p | a = 1 p 1 e ( m a 3 + n a p ) | 4 = { 2 p 3 3 p 2 3 p , if  3 p 1 , 2 p 3 5 p 2 15 p + 4 τ 3 ( χ 1 ) + 4 τ 3 ( χ ¯ 1 ) , if  3 | p 1 ,

where χ 1 is any 3-order character mod p .

Note that for any non-principal character χ mod p , we have | τ ( χ ) | = p , so from our theorem we may immediately deduce the following.

Corollary Let p > 3 be a prime with 3 | p 1 . Then for any integer n with ( n , p ) = 1 , we have the asymptotic formula
m = 1 p | a = 1 p 1 e ( m a 3 + n a p ) | 4 = 2 p 3 5 p 2 + O ( p 3 2 ) .

It seems that our method can also be used to deal with (1.1) for all prime p and integer k 4 . But this time, the computing is very complex.

For any integer h 3 , whether there exists an exact computational formula for
m = 1 p | a = 1 p 1 e ( m a 3 + n a p ) | 2 h ,

where p is an odd prime and ( n , p ) = 1 , is an open problem.

2 Several lemmas

In this section, we will give several lemmas which are necessary in the proof of our theorem. In the proving process of all lemmas, we used many properties of Gauss sums; all these can be found in [6], we will not be repeat them here. First we have the following.

Lemma 1 Let p be an odd prime, χ be any non-principal character mod p . Then for any integer n with ( n , p ) = 1 , we have the identity
| m = 1 p 1 χ ( m ) | a = 1 p 1 e ( m a 3 + n a p ) | 2 | = { p | a = 1 p 2 χ ¯ ( 3 a 2 + 3 a + 1 ) | , if  χ  is not a  3 -order character  mod p , p | 2 + a = 1 p 1 χ ¯ ( a ( 1 a ) ) | , if  χ  is a  3 -order character  mod p .
Proof Note that χ is a non-principal character mod p , so if χ is not a 3-order character mod p (that is, χ 3 χ 0 , the principal character mod p ), then from the properties of Gauss sums we have
m = 1 p 1 χ ( m ) | a = 1 p 1 e ( m a 3 + n a p ) | 2 = a = 1 p 1 b = 1 p 1 m = 1 p 1 χ ( m ) e ( m ( a 3 b 3 ) + n ( a b ) p ) = a = 1 p 1 b = 1 p 1 m = 1 p 1 χ ( m ) e ( m b 3 ( a 3 1 ) + n b ( a 1 ) p ) = τ ( χ ) a = 1 p 1 χ ¯ ( a 3 1 ) b = 1 p 1 χ ¯ 3 ( b ) e ( n b ( a 1 ) p ) = τ ( χ ) τ ( χ ¯ 3 ) a = 1 p 1 χ ¯ ( a 3 1 ) χ 3 ( n ( a 1 ) ) = χ 3 ( n ) τ ( χ ) τ ( χ ¯ 3 ) a = 1 p 2 χ ¯ ( ( a + 1 ) 3 1 ) χ ( a 3 ) = χ 3 ( n ) τ ( χ ) τ ( χ ¯ 3 ) a = 1 p 2 χ ¯ ( ( a ¯ + 1 ) 3 a ¯ 3 ) = χ 3 ( n ) τ ( χ ) τ ( χ ¯ 3 ) a = 1 p 2 χ ¯ ( 3 a 2 + 3 a + 1 ) ,
(2.1)

where τ ( χ ) = a = 1 p 1 χ ( a ) e ( a p ) denotes the classical Gauss sum.

If χ is a 3-order character mod p , then χ 3 = χ 0 ; note that for any integer a with ( a , p ) = 1 , we have
χ 2 ( a ) + χ ( a ) + 1 = { 3 , if  a  is a third residue  mod p , 0 , otherwise .
From the method of proving (2.1) we have the identity
m = 1 p 1 χ ( m ) | a = 1 p 1 e ( m a 3 + n a p ) | 2 = τ ( χ ) a = 1 p 1 χ ¯ ( a 3 1 ) b = 1 p 1 χ ¯ 3 ( b ) e ( n b ( a 1 ) p ) = τ ( χ ) a = 1 p 1 χ ¯ ( a 3 1 ) b = 1 p 1 e ( n b ( a 1 ) p ) = τ ( χ ) a = 1 p 1 χ ¯ ( a 3 1 ) = τ ( χ ) a = 1 p 1 ( χ 2 ( a ) + χ ( a ) + 1 ) χ ¯ ( a 1 ) = τ ( χ ) ( a = 1 p 1 χ ¯ ( a 1 ) + a = 1 p 1 χ ¯ ( a ¯ ( 1 a ¯ ) ) + a = 1 p 1 χ ¯ ( 1 a ¯ ) ) = τ ( χ ) ( 2 + a = 1 p 1 χ ¯ ( a ( 1 a ) ) ) .
(2.2)

Now note that | τ ( χ ) | = p , if χ χ 0 . From (2.1) and (2.2) we may immediately deduce Lemma 1. □

Lemma 2 Let p be an odd prime, χ be any non-real character mod p . Then we have the identity
a = 1 p 1 χ ( a ( a 1 ) ) = τ 2 ( χ ) τ ( χ 2 ) .
Therefore,
| a = 1 p 1 χ ( a ( a 1 ) ) | = p .
Proof From the definition and properties of the classical Gauss sums we have
τ 2 ( χ ) = a = 1 p 1 b = 1 p 1 χ ( a ) χ ( b ) e ( a + b p ) = a = 1 p 1 χ ( a ) b = 1 p 1 χ 2 ( b ) e ( b ( a + 1 ) p ) = τ ( χ 2 ) a = 1 p 1 χ ( a ) χ ¯ ( ( a + 1 ) 2 ) = τ ( χ 2 ) a = 2 p χ ( a 1 ) χ ¯ ( a 2 ) = τ ( χ 2 ) a = 2 p 1 χ ( ( 1 a ¯ ) a ¯ ) = τ ( χ 2 ) a = 1 p 1 χ ( a ( 1 a ) )
or
a = 1 p 1 χ ( a ( a 1 ) ) = τ 2 ( χ ) τ ( χ 2 ) .

This proves Lemma 2. □

3 Proof of the theorem

In this section, we shall complete the proof of our theorem. First from the orthogonality of characters mod p we have
χ mod p | m = 1 p 1 χ ( m ) | a = 1 p 1 e ( m a 3 + n a p ) | 2 | 2 = ( p 1 ) m = 1 p 1 | a = 1 p 1 e ( m a 3 + n a p ) | 4 .
(3.1)
On the other hand, if 3 p 1 , then any non-principal character χ is not a 3-order character mod p . Note that
m = 1 p 1 χ 0 ( m ) | a = 1 p 1 e ( m a 3 + n a p ) | 2 = a = 1 p 1 b = 1 p 1 m = 1 p 1 e ( m b 3 ( a 3 1 ) + n b ( a 1 ) p ) = ( p 1 ) 2 + p 2 = p 2 p 1 .
(3.2)
From (3.2) and Lemma 1 we have
χ mod p | m = 1 p 1 χ ( m ) | a = 1 p 1 e ( m a 3 + n a p ) | 2 | 2 = | m = 1 p 1 χ 0 ( m ) | a = 1 p 1 e ( m a 3 + n a p ) | 2 | 2 + χ mod p χ χ 0 | m = 1 p 1 χ ( m ) | a = 1 p 1 e ( m a 3 + n a p ) | 2 | 2 = ( p 2 p 1 ) 2 + χ mod p χ χ 0 p 2 | a = 1 p 2 χ ¯ ( 3 a 2 + 3 a + 1 ) | 2 = ( p 2 p 1 ) 2 + p 2 χ mod p | a = 1 p 2 χ ¯ ( 3 a 2 + 3 a + 1 ) | 2 p 2 ( a = 1 p 2 χ 0 ( 3 a 2 + 3 a + 1 ) ) 2 = ( p 2 p 1 ) 2 + p 2 ( p 1 ) a = 1 p 2 b = 1 p 2 a 2 + a b 2 + b mod p 1 p 2 ( p 2 ) 2 = ( p 2 p 1 ) 2 + p 2 ( p 1 ) a = 1 p 2 b = 1 p 2 ( a b ) ( a + b + 1 ) 0 mod p 1 p 2 ( p 2 ) 2 = ( p 2 p 1 ) 2 + p 2 ( p 1 ) ( p 2 + p 2 1 ) p 2 ( p 2 ) 2 = ( p 1 ) ( 2 p 3 3 p 2 3 p 1 ) .
(3.3)
If 3 p 1 , then combining (3.1) and (3.3) we may immediately deduce the identity
m = 1 p 1 | a = 1 p 1 e ( m a 3 + n a p ) | 4 = 2 p 3 3 p 2 3 p 1
or
m = 1 p | a = 1 p 1 e ( m a 3 + n a p ) | 4 = 2 p 3 3 p 2 3 p .
(3.4)
If 3 | p 1 , let χ 1 χ 0 be a 3-order character mod p , then χ ¯ 1 = χ 1 2 is also a 3-order character mod p ; this time note that
m = 1 p 1 χ 0 ( m ) | a = 1 p 1 e ( m a 3 + n a p ) | 2 = p ( p 1 ) 2 p 1 = p 2 3 p 1
and
a = 1 p 2 χ 1 ( ( a + 1 ) 3 a 3 ) = a = 1 p 2 χ 1 ( ( a ¯ + 1 ) 3 1 ) = a = 1 p 1 χ 1 ( a 3 1 ) = a = 1 p 1 ( χ ¯ 1 2 ( a ) + χ ¯ 1 ( a ) + 1 ) χ 1 ( a 1 ) = 2 + a = 1 p 1 χ 1 ( a ( 1 a ) ) ,
and from Lemma 2 and the method of proving (3.3) we have
χ mod p | m = 1 p 1 χ ( m ) | a = 1 p 1 e ( m a 3 + n a p ) | 2 | 2 = ( p 2 3 p 1 ) 2 + χ mod p χ χ 0 , χ 1 , χ 1 2 p 2 | a = 1 p 2 χ ¯ ( 3 a 2 + 3 a + 1 ) | 2 + 2 p | 2 + a = 1 p 1 χ 1 ( a ( 1 a ) ) | 2 = ( p 2 3 p 1 ) 2 + p 2 ( p 1 ) ( 2 p 5 ) 2 p 2 | a = 1 p 2 χ 1 ( ( a + 1 ) 3 a 3 ) | 2 + 2 p ( p + 4 2 a = 1 p 1 χ 1 ( a ( 1 a ) ) 2 a = 1 p 1 χ ¯ 1 ( a ( 1 a ) ) ) p 2 ( p 4 ) 2 = ( p 2 3 p 1 ) 2 + p 2 ( p 1 ) ( 2 p 5 ) p 2 ( p 4 ) 2 ( 2 p 2 2 p ) ( p + 4 2 a = 1 p 1 χ 1 ( a ( 1 a ) ) 2 a = 1 p 1 χ ¯ 1 ( a ( 1 a ) ) ) = ( p 1 ) ( 2 p 3 5 p 2 15 p 1 ) + 4 ( p 1 ) ( τ 3 ( χ 1 ) + τ 3 ( χ ¯ 1 ) ) .
(3.5)
So if 3 | p 1 , then combining (3.1) and (3.5) we can deduce the asymptotic formula
m = 1 p | a = 1 p 1 e ( m a 3 + n a p ) | 4 = 1 + m = 1 p 1 | a = 1 p 1 e ( m a 3 + n a p ) | 4 = 2 p 3 5 p 2 15 p + 4 ( τ 3 ( χ 1 ) + τ 3 ( χ ¯ 1 ) ) .
(3.6)

Now our theorem follows from (3.4) and (3.6).

Notes

Declarations

Acknowledgements

The authors would like to thank the referee for his very helpful and detailed comments, which have significantly improved the presentation of this paper. This work is supported by the P.S.F. (2013JZ001) and N.S.F. (11371291) of P.R. China, and the SF of Education of Shaanxi province (12JK0877).

Authors’ Affiliations

(1)
School of Science, Xi’an Polytechnic University
(2)
Department of Mathematics, Northwest University

References

  1. Cochrane T, Zheng Z: Pure and mixed exponential sums. Acta Arith. 1999, 91: 249–278.MathSciNetMATHGoogle Scholar
  2. Cochrane T, Zheng Z: Bounds for certain exponential sums. Asian J. Math. 2000, 4: 757–774.MathSciNetView ArticleMATHGoogle Scholar
  3. Cochrane T, Zheng Z: Upper bounds on a two-term exponential sums. Sci. China Ser. A 2001, 44: 1003–1015. 10.1007/BF02878976MathSciNetView ArticleMATHGoogle Scholar
  4. Cochrane T, Pinner C: A further refinement of Mordell’s bound on exponential sums. Acta Arith. 2005, 116: 35–41. 10.4064/aa116-1-4MathSciNetView ArticleMATHGoogle Scholar
  5. Cochrane T, Pinner C: Using Stepanov’s method for exponential sums involving rational functions. J. Number Theory 2006, 116: 270–292. 10.1016/j.jnt.2005.04.001MathSciNetView ArticleMATHGoogle Scholar
  6. Apostol TM: Introduction to Analytic Number Theory. Springer, New York; 1976.MATHGoogle Scholar

Copyright

© Zhu and Han; licensee Springer 2014

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.