The main purpose of this paper is using the properties of the classical Gauss sums and the analytic methods to study the computational problem of one kind of character sums analogous to high dimensional Kloosterman sums, and give some interesting identities for it.
Citation: Jianghua Li, Xi Zhang. On the character sums analogous to high dimensional Kloosterman sums[J]. AIMS Mathematics, 2022, 7(1): 294-305. doi: 10.3934/math.2022020
[1] | Junfeng Cui, Li Wang . The generalized Kloosterman's sums and its fourth power mean. AIMS Mathematics, 2023, 8(11): 26590-26599. doi: 10.3934/math.20231359 |
[2] | Xiaoxue Li, Wenpeng Zhang . A note on the hybrid power mean involving the cubic Gauss sums and Kloosterman sums. AIMS Mathematics, 2022, 7(9): 16102-16111. doi: 10.3934/math.2022881 |
[3] | He Yanqin, Zhu Chaoxi, Chen Zhuoyu . A sum analogous to Kloosterman sum and its fourth power mean. AIMS Mathematics, 2020, 5(3): 2569-2576. doi: 10.3934/math.2020168 |
[4] | Zhao Xiaoqing, Yi Yuan . Square-free numbers in the intersection of Lehmer set and Piatetski-Shapiro sequence. AIMS Mathematics, 2024, 9(12): 33591-33609. doi: 10.3934/math.20241603 |
[5] | Wenjia Guo, Yuankui Ma, Tianping Zhang . New identities involving Hardy sums S3(h,k) and general Kloosterman sums. AIMS Mathematics, 2021, 6(2): 1596-1606. doi: 10.3934/math.2021095 |
[6] | Wenpeng Zhang, Jiafan Zhang . The hybrid power mean of some special character sums of polynomials and two-term exponential sums modulo p. AIMS Mathematics, 2021, 6(10): 10989-11004. doi: 10.3934/math.2021638 |
[7] | Wenpeng Zhang, Xiaodan Yuan . On the classical Gauss sums and their some new identities. AIMS Mathematics, 2022, 7(4): 5860-5870. doi: 10.3934/math.2022325 |
[8] | Xi Liu . Some identities involving Gauss sums. AIMS Mathematics, 2022, 7(2): 3250-3257. doi: 10.3934/math.2022180 |
[9] | Wenjia Guo, Xiaoge Liu, Tianping Zhang . Dirichlet characters of the rational polynomials. AIMS Mathematics, 2022, 7(3): 3494-3508. doi: 10.3934/math.2022194 |
[10] | Hu Jiayuan, Chen Zhuoyu . On distribution properties of cubic residues. AIMS Mathematics, 2020, 5(6): 6051-6060. doi: 10.3934/math.2020388 |
The main purpose of this paper is using the properties of the classical Gauss sums and the analytic methods to study the computational problem of one kind of character sums analogous to high dimensional Kloosterman sums, and give some interesting identities for it.
Let q>1 be an integer, h, ci and b are integers with h>1 and i=1, 2, ⋯,h. The h-dimensional Kloosterman sums K(c1,c2,⋯,ch,b;q) is defined as:
K(c1,c2,⋯,ch,b;q)=q∑a1=1′q∑a2=1′⋯q∑ah=1′e(c1a1+c2a2+⋯+chah+b¯a1a2⋯ahq), |
where as usual, e(y)=e2πiy, q∑a=1′ denotes the summation over all 1≤a≤q such that (a,q)=1, and c⋅¯c≡1modq.
This sum occupies a very important position in the research of number theory, and many classical problems in analytic number theory are closely related to it. For these reasons, many scholars have studied the properties of K(c1,c2,⋯,ch,b;q), and obtained a series of important results. For example, R. A. Smith [1] studied the properties of the n-dimensional Kloosterman sums, and obtained a sharp upper bound estimate for
S(m,n;q)=q∑a1=1′⋯q∑an=1′e(a1+a2+⋯+an+m¯a1⋅¯a2⋯¯anq). |
W. P. Zhang and D. Han [2] studied the fourth power mean of the 2-dimensional Kloostermann sums, and proved the identity
p−1∑m=1|p−1∑a=1p−1∑b=1e(ma+b+¯abp)|4={7p5−18p4−(bp+6)p3−6p2−3pifp≡1mod6;7p5−22p4−(bp−14)p3−6p2−3pifp≡5mod6, |
where bp is an integer satisfying |bp|<2p32.
W. P. Zhang and X. X. Li [3] obtained the identity
∑χmodpp−1∑m=1|p−1∑a=1p−1∑b=1χ(ab)e(a+b+m¯abp)|4=(p−1)(2p5−7p4+2p3+8p2+4p+1). |
In this paper, we considered the character sums analogous to high dimensional Kloosterman sums as follows:
S(m,h,χ;p)=p−1∑a1=1p−1∑a2=1⋯p−1∑ah=1χ(a1+⋯+ah+m¯a1⋯ah), | (1.1) |
where p is an odd prime, χ is any non-principal Dirichlet character modp, h is any fixed positive integer, and m is any integer.
About summation (1.1), X. X. Lv and W. P. Zhang [4] studied the hybrid power mean involving S(m,2,χ;p) and two-term exponential sums, and proved the following results.
Let p be an odd prime with 3∤(p−1). Then for any non-principal character χ modulo p, one has the identities
p−1∑m=1|p−1∑a=1p−1∑b=1χ(a+b+m¯a¯b)|2⋅|p−1∑b=1e(mb+¯bp)|2=p−1∑m=1|p−1∑a=1p−1∑b=1χ(a+b+m¯a¯b)|2⋅|p−1∑b=1e(mb3+bp)|2=p2⋅(p2−p−1). |
If p is an odd prime with 3∣(p−1). Then for any three-th character χmodp (i.e., there exists a character χ1modp such that χ=χ31), then one has the asymptotic formulae
p−1∑m=1|p−1∑a=1p−1∑b=1χ(a+b+m¯a¯b)|2⋅|p−1∑b=1e(mb+¯bp)|2=3p4+N(p), |
and
p−1∑m=1|p−1∑a=1p−1∑b=1χ(a+b+m¯a¯b)|2⋅|p−1∑b=1e(mb3+bp)|2=3p4+N1(p), |
where N(p) and N1(p) satisfy |N(p)|≤9⋅p72 and |N1(p)|≤15⋅p3 respectively.
X. L. Xu, J. F. Zhang and W. P. Zhang [5] generalized the results in [4] to the high dimensional cases, and obtained the following conclusions:
Theorem A. Let p be an odd prime, h≥1 is an integer with (h+1,p−1)=1. Then for any non-principal character χmodp, one has the identity
p−1∑m=1|p−1∑a1=1⋯p−1∑ah=1χ(a1+⋯+ah+m¯a1⋯ah)|2⋅|p−1∑a=1e(mah+1+ap)|2 |
=ph⋅(p2−p−1). |
Theorem B. Let p be an odd prime, h is an integer with (h+1)∣(p−1), χ is any non-principal character modp. If χ is a (h+1)-th character modp (That is, there exists a character χ1modp such that χ=χh+11), then one has
p−1∑m=1|p−1∑a1=1⋯p−1∑ah=1χ(a1+⋯+ah+m¯a1⋯ah)|2⋅|p−1∑a=1e(mah+1+ap)|2 |
=(h+1)⋅ph+2+O(h2⋅ph+1). |
In addition, some other results related to the Kloosterman sums can also be found in the references [6,7,8,9,10,11,12], we no longer list here.
The main purpose of this paper is using the analytic methods and the properties of Gauss sums to study the computational problem of the fourth power mean of (1.1) with h=2, and give some interesting results. That is, we will prove the following:
Theorem 1.1. Let p be an odd prime with p≡1mod3, χ be any 3-th non-principal character modulo p. Then we have the identity
p−1∑m=1|p−1∑a=1p−1∑b=1χ(a+b+m¯ab)|4=15p4(p−1)+2(p−1)¯τ3(χ)χ3(3)p2⋅E9(χ)+2(p−1)τ3(χ)¯χ3(3)p2⋅¯E9(χ), |
where En(χ)=τn(χ1)+τn(χ1λ)+τn(χ1¯λ), ¯En(χ) denotes the complex conjugate of En(χ), and En(χ) satisfies the third-order linear recursion formula
En+3(χ)=E1(χ)⋅En+2(χ)−τ(χ)⋅¯χ(3)⋅¯E1(χ)⋅En+1(χ)+p⋅τ(χ)⋅¯χ(3)⋅En(χ), |
the first three terms of En(χ) are
E0(χ)=3; E1(χ)=p−1∑a=1χ(a)e(a3p); E2(χ)=E21(χ)−2τ(χ)¯χ(3)⋅¯E1(χ), and
τ(χ)=p−1∑a=1χ(a)e(ap) denotes the classical Gauss sums, |τ(χ)|=√p.
Theorem 1.2. Let p be a prime with p≡1mod3, then we have
1p−1p−1∑m=1|p−1∑a=1p−1∑b=1(a+b+m¯abp)|4=19p4+4p(d2−2p)(d4−4pd2+p2), |
where (∗p) denotes the Legendre symbol modulo p, d is uniquely determined by 4p=d2+27c2 and d≡1mod3.
From our theorems we can also deduce the following:
Corollary 1.1. Let p be an odd prime with p≡1mod3, g be any primitive root modulo p. Then we have the identity
132∑i=0(p−1∑a=1p−1∑b=1(a+b+gi⋅¯abp))2=3p2. |
Corollary 1.2. Let p be a prime, χ be any non-principal character modulo p. Then we have the estimate
p−1∑m=1|p−1∑a=1p−1∑b=1χ(a+b+m¯ab)|4≤27⋅p4⋅(p−1). |
Some notes: Let p be an odd prime, h≥1 is an integer with (h+1)∣(p−1). If χ is not a (h+1)-th character modulo p, then it is easy to prove that
|p−1∑a1=1p−1∑a2=1⋯p−1∑ah=1χ(a1+a2+⋯+ah+m¯a1⋯ah)|=0. |
Let p be an odd prime, k≥1 be an integer with (k+1,p−1)=1. Then for any integer m with (m,p)=1 and any non-principal character χmodp, we have
|p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1χ(a1+a2+⋯+ak+m¯a1⋯ak)|=pk2. |
So in these cases, our conclusions are trivial.
In some places of this paper, we need to use the definition and properties of the classical Gauss sums τ(χ) and character sums, these contents can be found in some analytic number theory books, such as [13], here we will not repeat the related contents. First we have the following:
Lemma 2.1. Let p be an odd prime, k≥1 is an integer with (k+1)∣(p−1), χ is any non-principal character modp, and m is an integer with (m,p)=1. If χ is not a (k+1)-th character modp, then we have the identity
p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1χ(a1+a2+⋯+ak+m¯a1⋯ak)=0. |
If χ is a (k+1)-th character modulo p, then we have the identity
p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1χ(a1+a2+⋯+ak+m¯a1⋯ak)=χ1(m)[τk+1(¯χ1)τ(¯χ)+¯λ(m)τk+1(¯χ1λ)τ(¯χ)+⋯+¯λk(m)τk+1(¯χ1λk)τ(¯χ)], |
where λ denotes any (k+1)-order character modulo p, and χ=χk+11.
Proof. If (k+1)∣(p−1), and χ is not a (k+1)-th character modp, then there exists an integer 1<r<p−1 such that rk+1≡¯rk+1≡1modp and χ(r)≠1. So from the properties of the reduced residue system modulo p we have
p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1χ(a1+a2+⋯+ak+m¯a1⋯ak)=p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1χ(ra1+ra2+⋯+rak+m¯rk¯a1⋯ak)=χ(r)⋅p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1χ(a1+a2+⋯+ak+m¯rk+1¯a1⋯ak)=χ(r)⋅p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1χ(a1+a2+⋯+ak+m¯a1⋯ak). | (2.1) |
Since χ(r)≠1, so from (2.1) we have the identity
p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1χ(a1+a2+⋯+ak+m¯a1⋯ak)=0. | (2.2) |
If (k+1)∣(p−1), and χ is a (k+1)-th character modp, let χ=χk+11, and λ is a k+1-order character modulo p. Then for any integer m with (m,p)=1, note that the identity
1+λ(b)+λ2(b)+⋯+λk(b)={k+1,if(b,p)=1andb≡ck+1modp;0,otherwise. |
From the properties of the classical Gauss sums we have
p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1χ(a1+a2+⋯+ak+m¯a1⋯ak)=1τ(¯χ)⋅p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1p−1∑b=1¯χ(b)e(b(a1+a2+⋯+ak+m¯a1a1⋯ak)p)=1τ(¯χ)⋅p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1e(a1+a2+⋯+akp)p−1∑b=1¯χ(b)e(mbk+1¯a1a2⋯akp)=1τ(¯χ)⋅p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1e(a1+a2+⋯+akp)×p−1∑b=1(1+λ(b)+λ2(b)+⋯+λk(b))¯χ1(b)e(mb¯a1a2⋯akp)=χ1(m)[τk+1(¯χ1)τ(¯χ)+¯λ(m)τk+1(¯χ1λ)τ(¯χ)+⋯+¯λk(m)τk+1(¯χ1λk)τ(¯χ)]. | (2.3) |
Now Lemma 2.1 follows from formulae (2.2) and (2.3).
Lemma 2.2. Let p be an odd prime, k>1 is an integer with (k+1,p−1)=1. Then for any non-principal character χmodp and integer m with (m,p)=1, we have the identity
|p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1χ(a1+a2+⋯+ak+m¯a1⋯ak)|=pk2. |
Proof. Note that (k+1,p−1)=1, so if a passes through a reduced residue system modulo p, then ak+1 also pass through a reduced residue system modulo p. Let h(k+1)≡1mod(p−1), then from the method of proving (2.3) we have
p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1χ(a1+a2+⋯+ak+m¯a1⋯ak)=1τ(¯χ)⋅p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1e(a1+a2+⋯+akp)p−1∑b=1¯χ(b)e(mbk+1¯a1a2⋯akp)=1τ(¯χ)⋅p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1e(a1+a2+⋯+akp)p−1∑b=1¯χ(bh)e(mbh(k+1)¯a1a2⋯akp)=1τ(¯χ)⋅p−1∑a1=1p−1∑a2=1⋯p−1∑ak=1e(a1+a2+⋯+akp)p−1∑b=1¯χh(b)e(mb¯a1a2⋯akp)=χh(m)⋅τ(¯χh)τ(¯χ)⋅(p−1∑a=1¯χh(a)e(ap))k=χh(m)⋅τk+1(¯χh)τ(¯χ). | (2.4) |
Note that |τ(¯χh)|=|τ(¯χ)|=√p, from (2.4) we can deduce Lemma 2.2.
Lemma 2.3. Let p be an odd prime with p≡1mod3. Then for any 3-th character χ=χ31 modulo p, we have the identity
p−1∑m=1|p−1∑a=1p−1∑b=1χ(a+b+m¯ab)|4=15p4(p−1)+2(p−1)¯τ3(χ)χ3(3)p2⋅(τ9(χ1)+τ9(χ1λ)+τ9(χ1¯λ))+2(p−1)τ3(χ)¯χ3(3)p2⋅(¯τ9(χ1)+¯τ9(χ1λ)+¯τ9(χ1¯λ)), |
where λ denotes any cubic-order character modulo p.
Proof. If (3,p−1)=3, then for any non-principal character χmodp with χ=χ31, taking k=2 in Lemma 2.1 we have
p−1∑a=1p−1∑b=1χ(a+b+m¯ab)=χ1(m)τ(¯χ)(τ3(¯χ1)+¯λ(m)τ3(¯χ1λ)+λ(m)τ3(¯χ1¯λ)). | (2.5) |
Note that the identity
p−1∑m=1λ(m)=p−1∑m=1¯λ(m)=0, |
from (2.5) we have
p−1∑m=1|p−1∑a=1p−1∑b=1χ(a+b+m¯ab)|4=1p2p−1∑m=1|τ3(¯χ1)+¯λ(m)τ3(¯χ1λ)+λ(m)τ3(¯χ1¯λ)|4=p4⋅p−1∑m=1(3+λ(m)¯τ3(¯χ1λ)¯τ3(¯χ1)+¯λ(m)¯τ3(¯χ1¯λ)¯τ3(¯χ1)+¯λ(m)τ3(¯χ1λ)τ3(¯χ1)+λ(m)τ3(¯χ1¯λ)τ3(¯χ1)+¯λ(m)p3¯τ3(¯χ1λ)τ3(¯χ1¯λ)+λ(m)p3τ3(¯χ1λ)¯τ3(¯χ1¯λ))2=p4(p−1)(15+2¯τ3(¯χ1λ)¯τ3(¯χ1¯λ)¯τ6(¯χ1)+2¯τ6(¯χ1λ)τ3(¯χ1¯λ)p3⋅¯τ3(¯χ1)+2τ3(¯χ1λ)τ3(¯χ1¯λ)τ6(¯χ1)+2¯τ3(¯χ1λ)τ6(¯χ1¯λ)p3⋅τ3(¯χ1)+2τ6(¯χ1λ)¯τ3(¯χ1¯λ)p3⋅τ3(¯χ1)+2τ3(¯χ1λ)¯τ6(¯χ1¯λ)p3⋅¯τ3(¯χ1)). | (2.6) |
From the triplication formula for Gauss sums (see [14]) we have
τ(χ)=τ(χ31)=1p⋅χ(3)⋅τ(χ1)⋅τ(χ1λ)⋅τ(χ1¯λ). | (2.7) |
Note that τ(χ)⋅¯τ(χ)=p, combining (2.6) and (2.7) we have the identity
p−1∑m=1|p−1∑a=1p−1∑b=1χ(a+b+m¯ab)|4=15p4(p−1)+2(p−1)¯τ3(χ)χ3(3)p2⋅(τ9(χ1)+τ9(χ1λ)+τ9(χ1¯λ))+2(p−1)τ3(χ)¯χ3(3)p2⋅(¯τ9(χ1)+¯τ9(χ1λ)+¯τ9(χ1¯λ)). |
This proves Lemma 2.3.
Lemma 2.4. Let p be a prime with p≡1mod6, χ be any sixth-order character modp. Then about the classical Gauss sums τ(χ), the following holds.
τ3(χ)+τ3(¯χ)={p12(d2−2p)if p=12h+1,−i⋅p12(d2−2p)if p=12h+7, |
where i2=−1, d is uniquely determined by 4p=d2+27c2 and d≡1mod3.
Proof. This result is Lemma 2.3 in S. Chowla, J. Cowles and M. Cowles [15], so the proof details are omitted.
Now we use the several basic lemmas of the previous section to complete the proofs of our theorems. First we prove Theorem 1.1. Let χ=χ31. Then from Lemma 2.3 we have
p−1∑m=1|p−1∑a=1p−1∑b=1χ(a+b+m¯ab)|4=15p4(p−1)+2(p−1)¯τ3(χ)χ3(3)p2⋅E9(χ)+2(p−1)τ3(χ)¯χ3(3)p2⋅¯E9(χ), | (3.1) |
where ¯En(χ) denotes the complex conjugate of En(χ).
Let r1=τ(χ1), r2=τ(χ1λ) and r3=τ(χ1¯λ). It is clear that
E1(χ)=τ(χ1)+τ(χ1λ)+τ(χ1¯λ)=p−1∑a=1χ(a)e(a3p). | (3.2) |
From (2.7) we have
r1r2r3=τ(χ1)⋅τ(χ1λ)⋅τ(χ1¯λ)=p⋅τ(χ)⋅¯χ(3). | (3.3) |
Note that ri⋅¯ri=p, i=1, 2, 3. So we have
E2(χ)=r21+r22+r23=(r1+r2+r3)2−2r1r2−2r1r3−2r2r3=E21(χ)−2r1r2r3p⋅(¯r1+¯r2+¯r3)=E21(χ)−2τ(χ)¯χ(3)⋅¯E1(χ). | (3.4) |
For any integer n≥0, it is easy to get the identity
En+3(χ)=(r1+r2+r3)⋅En+2(χ)−(r1r2+r1r3+r2r3)⋅En+1(χ)+r1r2r3⋅En(χ)=E1(χ)⋅En+2(χ)−τ(χ)⋅¯χ(3)⋅¯E1(χ)⋅En+1(χ)+p⋅τ(χ)⋅¯χ(3)⋅En(χ). |
Now note that E0(χ)=3, combining (3.2) and (3.4) and the above identity we may immediately deduce Theorem 1.1.
Now we prove Theorem 1.2. Taking χ=χ2=(∗p) as the Legendre symbol modulo p in (3.1). It is clear that χ=χ31=χ2. We discuss in two different conditions p=12k+1 and p=12k+7 respectively.
If p=12k+1, then χ2(3)=1, τ(χ2)=√p (see [13], chapter 9.10), from (3.1) we have
p−1∑m=1|p−1∑a=1p−1∑b=1(a+b+m¯abp)|4=15p4(p−1)+4(p−1)√p⋅[τ9(χ2)+τ9(χ2λ)+τ9(χ2¯λ)]. | (3.5) |
Note that χ2λ is a sixth-order character modulo p and τ(χ2λ)⋅τ(χ2¯λ)=p, from Lemma 2.4 we have
τ9(χ2λ)+τ9(χ2¯λ)=(τ3(χ2λ)+τ3(χ2¯λ))⋅((τ3(χ2λ)+τ3(χ2¯λ))2−3p3)=p32⋅(d2−2p)3−3p72⋅(d2−2p). | (3.6) |
Combining (3.5) and (3.6) we have the identity
1p−1p−1∑m=1|p−1∑a=1p−1∑b=1(a+b+m¯abp)|4=19p4+4p(d2−2p)⋅(d4−4pd2+p2). | (3.7) |
If p=12k+7, then χ2(3)=−1, τ(χ2)=i⋅√p and τ(χ2λ)⋅τ(χ2¯λ)=−p, from Lemma 2.4 and the method of proving (3.6) we have
τ9(χ2λ)+τ9(χ2¯λ)=i⋅p32⋅(d2−2p)3−3⋅i⋅p72⋅(d2−2p). | (3.8) |
Combining (3.1) and (3.8) we also have the identity
1p−1p−1∑m=1|p−1∑a=1p−1∑b=1(a+b+m¯abp)|4=19p4+4p(d2−2p)⋅(d4−4pd2+p2). | (3.9) |
It is clear that Theorem 1.2 follows from (3.7) and (3.9).
Now we prove Corollary 1.1. From (2.5) and the method of proving (2.6) we have
p−1∑m=1|p−1∑a=1p−1∑b=1(a+b+m¯abp)|2=1p⋅p−1∑m=1|τ3(χ2)+¯λ(m)τ3(χ2λ)+λ(m)τ3(χ2¯λ)|2=p2⋅p−1∑m=13=3p2(p−1). | (3.10) |
Let g be any primitive root modulo p, note that for any integer r,
|p−1∑a=1p−1∑b=1(a+b+g3r⋅m⋅¯abp)|=|p−1∑a=1p−1∑b=1(gra+grb+gr⋅m⋅¯abp)|=|(grp)⋅p−1∑a=1p−1∑b=1(a+b+m¯abp)|=|p−1∑a=1p−1∑b=1(a+b+m¯abp)|. | (3.11) |
So from (3.10) and (3.11) and the properties of primitive roots and the reduced residue system modulo p we have
3p2(p−1)=p−1∑m=1|p−1∑a=1p−1∑b=1(a+b+m¯abp)|2=2∑i=0p−13−1∑k=0|p−1∑a=1p−1∑b=1(a+b+g3k+i⋅¯abp)|2=2∑i=0p−13−1∑k=0|p−1∑a=1p−1∑b=1(a+b+gi⋅¯abp)|2=p−13⋅2∑i=0|p−1∑a=1p−1∑b=1(a+b+gi⋅¯abp)|2 |
which implies the identity
2∑i=0(13p−1∑a=1p−1∑b=1(a+b+gi⋅¯abp))2=p2. |
This proves Corollary 1.1.
It is easy to deduce Corollary 1.2 from our theorems. In fact, if (3,p−1)=1, then from Lemma 2.1 and Lemma 2.2 we have the estimate
p−1∑m=1|p−1∑a=1p−1∑b=1χ(a+b+m¯ab)|4≤p4(p−1). | (3.12) |
If 3∣(p−1) and χ is not a 3-th character modulo p, then from Lemma 2.1 we have
p−1∑m=1|p−1∑a=1p−1∑b=1χ(a+b+m¯ab)|4=0. | (3.13) |
If 3∣(p−1) and χ is a 3-th character modulo p, then note that the estimate
|E9(χ)|≤3⋅p92, and |τ(χ)|=√p, |
from (3.1) we have the estimate
p−1∑m=1|p−1∑a=1p−1∑b=1χ(a+b+m¯ab)|4≤15p4(p−1)+12p4(p−1)=27p4(p−1). | (3.14) |
Combining (3.12)–(3.14) we may immediately deduce Corollary 1.2.
This completes the proofs of our all results.
The main result of this paper is using the elementary and analytic methods to give an exact calculating formula for the fourth power mean of the character sums analogous to 2-dimensional Kloosterman sums. The result is a new contribution to the relevant fields.
This work is supported by the N. S. F. (11771351, 413610022) of China and the N. S. F. of Shaanxi Province (2021JM-323).
All authors declare no conflicts of interest in this paper.
[1] |
R. A. Smith, On n-dimensional Kloosterman sums, J. Number Theory, 11 (1979), 324–343. doi: 10.1016/0022-314X(79)90006-4. doi: 10.1016/0022-314X(79)90006-4
![]() |
[2] |
W. P. Zhang, D. Han, A new identity involving the classical Kloosterman sums and 2-dimensional Kloostermann sums, Int. J. Number Theory, 12 (2016), 111–119. doi: 10.1142/S179304211650007X. doi: 10.1142/S179304211650007X
![]() |
[3] |
W. P. Zhang, X. X. Li, The fourth power mean of the general 2-dimensional Kloostermann sums modp, Acta Math. Sin. English Ser., 33 (2017), 861–867. doi: 10.1007/s10114-016-6347-9. doi: 10.1007/s10114-016-6347-9
![]() |
[4] |
X. X. Lv, W. P. Zhang, On the character sum of polynomials and the two-term exponential sums, Acta Math. Sin. English Ser., 36 (2020), 196–206. doi: 10.1007/s10114-020-9255-y. doi: 10.1007/s10114-020-9255-y
![]() |
[5] |
X. L. Xu, J. F. Zhang, W. P. Zhang, The character sum of polynomials with k variables and two-term exponential sums, Notes Number Theory, 27 (2021), 112–124. doi: 10.7546/nntdm.2021.27.1.112-124. doi: 10.7546/nntdm.2021.27.1.112-124
![]() |
[6] | S. Chern, On the power mean of a sum analogous to the Kloosterman sum, 2017, arXiv: 1712.01422. |
[7] |
W. P. Zhang, On the fourth and sixth power mean of the classical Kloosterman sums, J. Number Theory, 131 (2011), 228–238. doi: 10.1016/j.jnt.2010.08.008. doi: 10.1016/j.jnt.2010.08.008
![]() |
[8] |
W. P. Zhang, X. X. Lv, The fourth power mean of the general 3-dimensional Kloostermann sums modp, Acta Math. Sin. English Ser., 33 (2017), 369–377. doi: 10.1007/s10114-016-6347-9. doi: 10.1007/s10114-016-6347-9
![]() |
[9] |
W. P. Zhang, On the fourth power mean of the general Kloosterman sums, J. Number Theory, 169 (2016), 315–326. doi: 10.1016/j.jnt.2016.05.018. doi: 10.1016/j.jnt.2016.05.018
![]() |
[10] |
W. P. Zhang, S. M. Shen, A note on the fourth power mean of the generalized Kloosterman sums, J. Number Theory, 174 (2017), 419–426. doi: 10.1016/j.jnt.2016.11.020. doi: 10.1016/j.jnt.2016.11.020
![]() |
[11] | W. Duke, H. Iwaniec, A relation between cubic exponential and Kloosterman sums, Contemp. Math., 145 (1993), 255–258. |
[12] | Y. W. Hou, W. P. Zhang, One kind high dimensional Kloosterman sums and its upper bound estimate, Journal of Shaanxi Normal University (Natural Science Edition), 46 (2018), 28–31. |
[13] | T. M. Apostol, Introduction to analytic number theory, Springer Science & Business Media, 1976. |
[14] |
J. Greene, D. Stanton, The triplication formula for Gauss sums, Aeq. Math., 30 (1986), 134–141. doi: 10.1007/BF02189920. doi: 10.1007/BF02189920
![]() |
[15] |
S. Chowla, J. Cowles, M. Cowles, On the number of zeros of diagonal cubic forms, J. Number Theory, 9 (1977), 502–506. doi: 10.1016/0022-314X(77)90010-5. doi: 10.1016/0022-314X(77)90010-5
![]() |
1. | Jiayuan Hu, Xiaogang Liu, Some New Identities Related to Dedekind Sums Modulo a Prime, 2024, 2024, 2314-4785, 1, 10.1155/2024/8844153 |