Let q≥3 be a positive integer. For any integers m, n, k, h, the two-term exponential sums C(m,n,k,h;q) is defined as C(m,n,k,h;q)=q∑a=1e(mak+nahq), where k>h≥2. The main purpose of this paper is to use analytic methods and the properties of classical Gauss sums to study the mean value involving two-term exponential sums and fourth Gauss sums, and to provide some asymptotic formulas and identities. Previously, only the case of h=1 had been studied.
Citation: Zhefeng Xu, Xiaoying Liu, Luyao Chen. Hybrid mean value involving some two-term exponential sums and fourth Gauss sums[J]. Electronic Research Archive, 2025, 33(3): 1510-1522. doi: 10.3934/era.2025071
[1] | Jianghua Li, Zepeng Zhang, Tianyu Pan . On two-term exponential sums and their mean values. Electronic Research Archive, 2023, 31(9): 5559-5572. doi: 10.3934/era.2023282 |
[2] | Li Rui, Nilanjan Bag . Fourth power mean values of one kind special Kloosterman's sum. Electronic Research Archive, 2023, 31(10): 6445-6453. doi: 10.3934/era.2023326 |
[3] | Li Wang, Yuanyuan Meng . Generalized polynomial exponential sums and their fourth power mean. Electronic Research Archive, 2023, 31(7): 4313-4323. doi: 10.3934/era.2023220 |
[4] | Jin Zhang, Xiaoxue Li . The sixth power mean of one kind generalized two-term exponential sums and their asymptotic properties. Electronic Research Archive, 2023, 31(8): 4579-4591. doi: 10.3934/era.2023234 |
[5] | Jiafan Zhang . On the distribution of primitive roots and Lehmer numbers. Electronic Research Archive, 2023, 31(11): 6913-6927. doi: 10.3934/era.2023350 |
[6] | Shuting Chang, Yaojun Ye . Upper and lower bounds for the blow-up time of a fourth-order parabolic equation with exponential nonlinearity. Electronic Research Archive, 2024, 32(11): 6225-6234. doi: 10.3934/era.2024289 |
[7] | Hongjian Li, Kaili Yang, Pingzhi Yuan . The asymptotic behavior of the reciprocal sum of generalized Fibonacci numbers. Electronic Research Archive, 2025, 33(1): 409-432. doi: 10.3934/era.2025020 |
[8] | Jun Liu, Yue Liu, Xiaoge Yu, Xiao Ye . An efficient numerical method based on QSC for multi-term variable-order time fractional mobile-immobile diffusion equation with Neumann boundary condition. Electronic Research Archive, 2025, 33(2): 642-666. doi: 10.3934/era.2025030 |
[9] | Tingting Du, Zhengang Wu . On the reciprocal sums of products of $ m $th-order linear recurrence sequences. Electronic Research Archive, 2023, 31(9): 5766-5779. doi: 10.3934/era.2023293 |
[10] | Qingjie Chai, Hanyu Wei . The binomial sums for four types of polynomials involving floor and ceiling functions. Electronic Research Archive, 2025, 33(3): 1384-1397. doi: 10.3934/era.2025064 |
Let q≥3 be a positive integer. For any integers m, n, k, h, the two-term exponential sums C(m,n,k,h;q) is defined as C(m,n,k,h;q)=q∑a=1e(mak+nahq), where k>h≥2. The main purpose of this paper is to use analytic methods and the properties of classical Gauss sums to study the mean value involving two-term exponential sums and fourth Gauss sums, and to provide some asymptotic formulas and identities. Previously, only the case of h=1 had been studied.
Let q≥3 be a positive integer. For any integers m, n, k, h, the two-term exponential sums C(m,n,k,h;q) is defined as
C(m,n,k,h;q)=q∑a=1e(mak+nahq), |
where k>h. When n=0, this sum reduces to the k-th Gauss sum, defined by
G(m,k;q)=q∑a=1e(makq). |
These summations are very important in additive number theory because there are close relations between the summation and the Waring's problem. About its arithmetical properties, many authors had studied it. For example, Davenport and Heilbronn [1] proved that
C(m,n,k,1;pr)≪kpθr(b,pr),ifp∤m. | (1.1) |
where θ=23 if k=3, and θ=34 if k≥3. Applying Weil's estimate for exponential sums over finite fields, Hua [2] proved that θ=12 for all k≥2. Loxton and Smith [3], Smith [4], Loxton and Vaughan [5], Dabrowski and Fisher [6], among others, have made improvements to (1.1). Recently, Li [7], utilizing the recursive formula for the fourth Gauss sum and relevant results on two-term exponential sums, further extended the study to derive a fourth-order linear recurrence formula for the fourth Gauss sum and the two-term exponential sum
Mk(p)=p−1∑m=1(p−1∑a=0e(ma4p))k⋅|p−1∑a=0e(ma4+ap)|2 |
in the case where p≡1mod4.
We have been exploring the properties of two-term exponential sums, specifically |p−1∑a=1e(ma4+a2p)|t. The main challenge arises when the lower-order terms of the two-term exponential sum are 2. If a passes through a residue system modulo p while a2 does not pass through the same residue system, direct application of properties of Gauss sums or trigonometric identity becomes complex. Yuan and Wang [8], utilizing the orthogonality and parity properties of characters modulo p, ingeniously transformed the problem and ultimately provided a concise formula. Specifically, for p≡3mod4, the following identity holds:
p−1∑m=0|p−1∑a=1e(ma4+a2p)|4=p⋅(7p2−14p−5), | (1.2) |
however, when p≡1mod4, the research on the problem becomes relatively complex, and they did not provide relevant results.
For the sake of convenience, we have introduced some symbols
α=α(p)=p−12∑a=1(a+¯ap);γ=p−1∑a=1ψ(a2−1)¯ψ(a2+1). |
The main purpose of this paper is to study the asymptotic properties of the t-th power mean
p−1∑m=1|p−1∑a=0e(ma4p)|k⋅|p−1∑a=1e(ma4+a2p)|t, | (1.3) |
and give some exact formulas for (1.3) with t=2; for convenience, we will denote it as Tk(p). That is, we shall prove the following:
Theorem 1. Let p be a prime with p≡5mod8. Then we have the identity
T2(p)=6p3−6p52−15p2+6p32−3p−4pα(1+α−√p). |
Corollary 1. Let p be an odd prime with p≡5mod8. Then we have
T2(p)=6p3−6p52+O(p2). |
Theorem 2. Let p be a prime with p≡1mod8. Then we have the identity
T2(p)=6p3+6p52−23p2+6p32−3p+4pα(3−α−√p)+2p(τ2(ψ)γ+τ2(¯ψ)¯γ), |
where τ(ψ)=p−1∑a=0ψ(a)e(ap), which usually denotes the classical Gauss sums.
Corollary 2. Let p be an odd prime with p≡1mod8. Then we have
T2(p)=6p3+O(p52). |
Theorem 3. Let p be a prime with p≡3mod4. Then for any integer k≥4, we have
p−1∑m=1|p−1∑a=0e(ma4p)|2k⋅|p−1∑a=1e(ma4+a2p)|4=pk+1⋅(7p2−14p−5). |
Some notes:
(1) When p≡3mod4, we have (−1p)=−1. In this case, for any integer m with (m,p)=1, according to Theorem 7.5.4 in reference [9], we have
G(m,2;p)=i(mp)√p, |
where i2=−1. Combining with the properties of Gauss sums, we obtain
G(m,4;p)=1+p−1∑a=1(1+(ap))e(ma2p)=(mp)i√p, |
then
|G(m,4;p)|2=p, | (1.4) |
when p≡3mod4, the hybrid means of Tk(p) are easily obtained. Hence, in this section, we only consider the case when p≡1mod4.
(2) From the recursive formula for G4(m,4;p)=(p−1∑a=0e(ma4p))4 in [10], we can derive a fourth-order linear recurrence formula for Tk(p).
When p=8k+5, the recursive formula for G4(m,4;p) is as follows:
G4(m,4;p)=−2pG2(m,4;p)+4pG(m,4;p)(p−1∑a=1(a+¯ap))−9p2+p(p−1∑a=1(a+¯ap))2=−2pG2(m,4;p)+8pαG(m,4;p)−9p2+4pα2. |
Through simple calculations, we obtain the recursive formula for Tk(p) as follows:
Tk(p)=p−1∑m=1(p−1∑a=0e(ma4p))k⋅|p−1∑a=1e(ma4+a2p)|2=p−1∑m=1Gk(m,4;p)|p−1∑a=1e(ma4+a2p)|2=p−1∑m=1G4(m,4;p)Gk−4(m,4;p)|p−1∑a=1e(ma4+a2p)|2=p−1∑m=1((−2pG2(m,4;p)+8pαG(m,4;p)−9p2+4pα2)Gk−4(m,4;p))⋅|p−1∑a=1e(ma4+a2p)|2=−2pTk−2(p)+8pαTk−3(p)+(4pα2−9p2)Tk−4(p). |
Similarly, when p=8k+1, according to [10], we know
G4(m,4;p)=6pG2(m,4;p)+4pG(m,4;p)(p−1∑a=1(a+¯ap))−p2+p(p−1∑a=1(a+¯ap))2=6pG2(m,4;p)+8pαG(m,4;p)−p2+4pα2, |
then we have
Tk(p)=6pTk−2(p)+8pαTk−3(p)+(4pα2−p2)Tk−4(p). |
(3) When p≡1mod4 and k is any positive integer, for the hybrid mean given by
p−1∑m=1|p−1∑a=0e(ma4p)|2k⋅|p−1∑a=1e(ma4+a2p)|2, |
as k increases, the computation becomes increasingly complex. Whether precise computational formulas or corresponding recurrence relations can be obtained remains a question worth investigating.
In this section, we give some lemmas, which will be applied to prove the main results.
Lemma 1. Let p be an odd prime with p≡1mod4. For any fourth-order character ψ modulo p, we have
p−1∑a=1ψ(a4−1)=1+ψ(−1)√p(τ2(ψ)−√p), | (2.1) |
p−1∑a=1ψ2(a4−1)=−2+ψ(−1)√p(τ2(ψ)+τ2(¯ψ)), | (2.2) |
p−1∑a=1¯ψ(a4−1)=1+¯ψ(−1)√p(τ2(¯ψ)−√p), | (2.3) |
where ψ2=χ2 and χ2=(∗p) denotes the Legendre symbol.
Proof. In fact, this is Lemma 2.1 of [11], so its proof is omitted.
Lemma 2. Let p be an odd prime with p≡1mod4. For any fourth-order character ψ modulo p, we have
τ2(ψ)+τ2(¯ψ)=2√pα. |
Proof. See Lemma 2.2 in [11].
Lemma 3. Let p be an odd prime with p≡1mod4, and a be any integer with (a,p)=1. For any fourth-order character ψ modulo p, we have
p−1∑m=1ψ(m)|p−1∑a=1e(ma4+a2p)|2={2(τ(ψ)−τ(¯ψ)√p)+τ(ψ)√pγ,if p≡1 mod 8,τ(ψ)√pγ,if p≡5 mod 8. |
Proof. First, using the trigonometric identity
q∑m=1e(nmq)={q,if q∣n,0,if q∤n. | (2.4) |
Then, utilizing the properties of Dirichlet characters, ψ4=¯ψ4=χ0 and ¯ψχ2=ψ, we can obtain
p−1∑m=1ψ(m)|p−1∑a=1e(ma4+a2p)|2=p−1∑m=1ψ(m)p−1∑a=1p−1∑b=1e(m(a4−b4)+(a2−b2)p)=p−1∑a=1p−1∑b=1e(b2(a2−1)p)p−1∑m=1ψ(m)e(mb4(a4−1)p)=τ(ψ)p−1∑a=1p−1∑b=1¯ψ(b4)¯ψ(a4−1)e(b2(a2−1)p)=τ(ψ)p−1∑a=1p−1∑b=1¯ψ(a4−1)e(b2(a2−1)p)=τ(ψ)p−1∑a=1¯ψ(a4−1)(p−1∑b=1(1+χ2(b))e(b(a2−1)p))=τ(ψ)p−1∑a=1¯ψ(a4−1)p−1∑b=1e(b(a2−1)p)+τ(ψ)p−1∑a=1¯ψ(a4−1)p−1∑b=1χ2(b)e(b(a2−1)p)=−τ(ψ)p−1∑a=1¯ψ(a4−1)+τ(ψ)τ(χ2)p−1∑a=1¯ψ(a4−1)χ2(a2−1)=−τ(ψ)p−1∑a=1¯ψ(a4−1)+τ(ψ)τ(χ2)p−1∑a=1ψ(a2−1)¯ψ(a2+1). |
When p≡1mod4, τ(χ2)=√p, combined with (2.3) in Lemma 1, we obtain
p−1∑m=1ψ(m)|p−1∑a=1e(ma4+a2p)|2=−τ(ψ)(1+¯ψ(−1)√p(τ2(¯ψ)−√p))+τ(ψ)√pγ=(1+¯ψ(−1))(τ(ψ)−ψ(−1)τ(¯ψ)√p)+τ(ψ)√pγ. |
When p≡1mod8, ψ(−1)=¯ψ(−1)=1; when p≡5mod8, ψ(−1)=¯ψ(−1)=−1. Therefore, we have
p−1∑m=1ψ(m)|p−1∑a=1e(ma4+a2p)|2={2(τ(ψ)−τ(¯ψ)√p)+τ(ψ)√pγ,if p≡1 mod 8,τ(ψ)√pγ,if p≡5 mod 8. |
This proves Lemma 3.
Lemma 4. Let p be an odd prime with p≡1mod4, then
p−1∑m=1χ2(m)|p−1∑a=1e(ma4+a2p)|2={2√p(1−α−√p),if p≡1 mod 8,2√p(1+α−√p),if p≡5 mod 8. |
Proof. From the proof of Lemma 3, we obtain
p−1∑m=1χ2(m)|p−1∑a=1e(ma4+a2p)|2=p−1∑m=1χ2(m)p−1∑a=1p−1∑b=1e(m(a4−b4)+(a2−b2)p)=p−1∑a=1p−1∑b=1e(b2(a2−1)p)p−1∑m=1χ2(m)e(mb4(a4−1)p)=τ(χ2)p−1∑a=1p−1∑b=1χ2(b4)χ2(a4−1)e(b2(a2−1)p)=τ(χ2)p−1∑a=1p−1∑b=1χ2(a4−1)e(b2(a2−1)p)=τ(χ2)p−1∑a=1χ2(a4−1)(p−1∑b=1(1+χ2(b))e(b(a2−1)p))=τ(χ2)p−1∑a=1χ2(a4−1)p−1∑b=1e(b(a2−1)p)+τ(χ2)p−1∑a=1χ2(a4−1)p−1∑b=1χ2(b)e(b(a2−1)p)=−τ(χ2)p−1∑a=1χ2(a4−1)+τ2(χ2)p−1∑a=1χ2(a4−1)χ2(a2−1)=−τ(χ2)p−1∑a=1χ2(a4−1)+τ2(χ2)p−1∑a=1χ2(a2+1). |
Utilizing the properties of the Legendre symbol, then
p−1∑a=0χ2(a2+n)={−1,if (p,n)=1,p−1,if (p,n)=p. | (2.5) |
Combining (2.5) with (2.2) from Lemma 1, we obtain
p−1∑m=1χ2(m)|p−1∑a=1e(ma4+a2p)|2=−τ(χ2)(−2+ψ(−1)√p(τ2(ψ)+τ2(¯ψ)))−2p, |
from Lemma 2, then
p−1∑m=1χ2(m)|p−1∑a=1e(ma4+a2p)|2=−τ(χ2)(−2+ψ(−1)2α)−2p=√p(2−ψ(−1)2α)−2p={2√p(1−α−√p),if p≡1 mod 8,2√p(1+α−√p),if p≡5 mod 8. |
This proves Lemma 4.
Lemma 5. Let p be an odd prime with p≡1mod4, then
p−1∑m=1|p−1∑a=1e(ma4+a2p)|2={2p2+2p32−5p+2p12−1,if p≡1 mod 8,2p2−2p32−5p+2p12−1,if p≡5 mod 8. |
Proof. From the properties of quadratic Gauss sums, we obtain
|p−1∑a=1e(a2p)|2=|p∑a=1e(a2p)−1|2={p−2√p+1,p≡1 mod 4,p+1,p≡3 mod 4. | (2.6) |
Further calculations yield
p−1∑m=1|p−1∑a=1e(ma4+a2p)|2=p−1∑m=0|p−1∑a=1e(ma4+a2p)|2−|p−1∑a=1e(a2p)|2=p−1∑a=1p−1∑b=1p−1∑m=0e(mb4(a4−1)+b2(a2−1)p)−|p−1∑a=1e(a2p)|2=pp−1∑a=1p−1∑b=1a4≡1 mod pe(b2(a2−1)p)−(p−2√p+1)=pp−1∑a=1p−1∑b=1a4≡1 mod p(1+χ2(b))e(b(a2−1)p)−(p−2√p+1)=pp−1∑a=1p−1∑b=1a4≡1 mod pe(b(a2−1)p)+pp−1∑a=1p−1∑b=1a4≡1 mod pχ2(b)e(b(a2−1)p)−(p−2√p+1)=pp−1∑a=1p−1∑b=1a4≡1 mod pa2≡1 mod p1+pp−1∑a=1p−1∑b=1a4≡1 mod pa2≢1 mod pe(b(a2−1)p)+pτ(χ2)p−1∑a=1a4≡1 mod pχ2(a2−1)−(p−2√p+1)=2p(p−1)−2p+2pτ(χ2)(−2p)−(p−2√p+1)=2p2−5p+2p√p(−2p)+2√p−1, |
using the properties of the Legendre symbol, then
(−2p)=(−1p)(2p)=(−1)p−12(−1)p2−18={1,ifp≡1or3 mod 8,−1,ifp≡5or7 mod 8. | (2.7) |
According to (2.7), we can obtain
p−1∑m=1|p−1∑a=1e(ma4+a2p)|2={2p2+2p32−5p+2p12−1,if p≡1 mod 8,2p2−2p32−5p+2p12−1,if p≡5 mod 8. |
This proves Lemma 5.
Lemma 6. Let p be an odd prime with p≡1mod4, then
∑χ mod p|p−1∑a=1ψ(a2−1)¯ψ(a2+1)|2=2(p−1)(p−3). |
Proof. According to the properties of the character modulo p, we have
∑χ mod pχ(a)¯χ(n)={ϕ(p),a≡n mod p,0,a≢n mod p. | (2.8) |
Noting that p∤(a2+1), p∣(a2−1), and using (2.8), we obtain
∑χ mod p|p−1∑a=1ψ(a2−1)¯ψ(a2+1)|2=(p−1)p−2∑a=2p−2∑b=2(a2−1)¯(a2+1)≡(b2−1)¯(b2+1) mod p1=(p−1)p−2∑a=2p−2∑b=2(b2+1)(a2−1)≡(a2+1)(b2−1) mod p1=(p−1)p−2∑a=2p−2∑b=2a2≡b2 mod p1=2(p−1)(p−3). |
This proves Lemma 6.
For any integer m satisfying (m,p)=1, based on the classical properties of Gauss sums and the properties of the fourth-order character ψ, combined with (2.4), we have
G(m,4;p)=p−1∑a=0e(ma4p)=1+p−1∑a=1(1+ψ(a)+χ2(a)+¯ψ(a))e(map)=p−1∑a=0e(map)+p−1∑a=1ψ(a)e(map)+p−1∑a=1χ2(a)e(map)+p−1∑a=1¯ψ(a)e(map)=χ2(m)√p+¯ψ(m)τ(ψ)+ψ(m)τ(¯ψ). | (3.1) |
Let us start by proving Theorem 1.
When p≡5mod8, we have ψ(−1)=−1. Utilizing the property ¯τ(ψ)=ψ(−1)τ(¯ψ)=−τ(¯ψ), it follows that in this scenario, ¯¯ψ(m)τ(ψ)+ψ(m)τ(¯ψ)=−(¯ψ(m)τ(ψ)+ψ(m)τ(¯ψ)). Additionally, since χ2(m)√p is a real number; based on (3.1), we obtain
¯G(m,4;p)=χ2(m)√p−¯ψ(m)τ(ψ)−ψ(m)τ(¯ψ). | (3.2) |
In this particular case, according to (3.1) and (3.2), we derive
|G(m,4;p)|2=G(m,4;p)⋅¯G(m,4;p)=p−(¯ψ(m)τ(ψ)+ψ(m)τ(¯ψ))2=3p−χ2(m)(τ2(ψ)+τ2(¯ψ)). | (3.3) |
Now, considering (3.3) and incorporating Lemmas 2, 4, and 5, we obtain
T2(p)=p−1∑m=1|G(m,4;p)|2|p−1∑a=1e(ma4+a2p)|2=p−1∑m=1(3p−χ2(m)(τ2(ψ)+τ2(¯ψ)))|p−1∑a=1e(ma4+a2p)|2=3pp−1∑m=1|p−1∑a=1e(ma4+a2p)|2−(τ2(ψ)+τ2(¯ψ))p−1∑m=1χ2(m)|p−1∑a=1e(ma4+a2p)|2=3p(2p2−2p32−5p+2p12−1)−2√pα(2√p(1+α−√p))=6p3−6p52−15p2+6p32−3p−4pα(1+α−√p). |
This proves Theorem 1.
When p≡1mod8, it holds that ψ(−1)=1. For any integer m with (m,p)=1, as stated in (3.1), we establish that ¯G(m,4;p)=G(m,4;p), implying that G(m,4;p) is a real number, then
T2(p)=p−1∑m=1|G(m,4;p)|2|p−1∑a=1e(ma4+a2p)|2=p−1∑m=1(χ2(m)√p+¯ψ(m)τ(ψ)+ψ(m)τ(¯ψ))2|p−1∑a=1e(ma4+a2p)|2=p−1∑m=1(p+¯ψ2(m)τ2(ψ)+ψ2(m)τ2(¯ψ)+2τ(ψ)τ(¯ψ)+2χ2(m)√p(¯ψ(m)τ(ψ)+ψ(m)τ(¯ψ)))|p−1∑a=1e(ma4+a2p)|2=3pp−1∑m=1|p−1∑a=1e(ma4+a2p)|2+(τ2(ψ)+τ2(¯ψ))p−1∑m=1χ2(m)|p−1∑a=1e(ma4+a2p)|2+2√pp−1∑m=1(ψ(m)τ(ψ)+¯ψ(m)τ(¯ψ))|p−1∑a=1e(ma4+a2p)|2. |
Since τ(ψ)τ(¯ψ)=ψ(−1)p=p, combining with Lemmas 2, 3, 4, and 5, we obtain
T2(p)=p−1∑m=1|G(m,4;p)|2|p−1∑a=1e(ma4+a2p)|2=3p(2p2+2p32−5p+2p12−1)+2√pα(2√p(1−α−√p))+2√pτ(ψ)p−1∑m=1ψ(m)|p−1∑a=1e(ma4+a2p)|2+2√pτ(¯ψ)p−1∑m=1¯ψ(m)|p−1∑a=1e(ma4+a2p)|2=3p(2p2+2p32−5p+2p12−1)+2√pα(2√p(1−α−√p))+2√pτ(ψ)(2(τ(ψ)−τ(¯ψ)√p)+τ(ψ)√pγ)+2√pτ(¯ψ)(2(τ(¯ψ)−τ(ψ)√p)+τ(¯ψ)√p¯γ)=3p(2p2+2p32−5p+2p12−1)+2√pα(2√p(1−α−√p))+4√pτ2(ψ)−4pτ(ψ)τ(¯ψ)+2pτ2(ψ)γ+4√pτ2(¯ψ)−4pτ(ψ)τ(¯ψ)+2pτ2(¯ψ)¯γ=6p3+6p52−23p2+6p32−3p+4pα(3−α−√p)+2p(τ2(ψ)γ+τ2(¯ψ)¯γ). |
This proves Theorem 2.
When p≡3mod4, as stated in (1.2) and (1.4), we obtain
p−1∑m=1|p−1∑a=0e(ma4p)|2k⋅|p−1∑a=1e(ma4+a2p)|4=pk+1⋅(7p2−14p−5). |
This completes the proof of Theorem 3.
For p≡5mod8, considering |α|≤√p and Theorem 1, we obtain
T2(p)=p−1∑m=1|p−1∑a=0e(ma4p)|2⋅|p−1∑a=1e(ma4+a2p)|2=6p3−6p52+O(p2). |
This completes the proof of Corollary 1.
In the case of p≡1mod8, where |τ2(ψ)|=p, combining with Lemma 6 and Theorem 2, we have
T2(p)=p−1∑m=1|p−1∑a=0e(ma4p)|2⋅|p−1∑a=1e(ma4+a2p)|2=6p3+O(p52). |
This completes the proof of all our theorems and corollaries.
The authors declare they have not used Artificial Intelligence (AI) tools in the creation of this article.
This work is supported by the N.S.F. (12471006, 12371007) of China and Shaanxi Fundamental Science Research Project for Mathematics and Physics (Grant No. 22JSY007).
The authors declare there are no conflicts of interest.
[1] |
H. Davenport, H. Heilbronn, On an exponential sum, Proc. London Math. Soc., 41 (1936), 449–453. https://doi.org10.1112/plms/s2-41.6.449 doi: 10.1112/plms/s2-41.6.449
![]() |
[2] | L. K. Hua, On exponential sums, Sci. Record, 1 (1957), 1–4. |
[3] |
J. H. Loxton, R. A. Smith, On Hua's estimate for exponential sums, J. London Math. Soc., 26 (1982), 15–20. https://doi.org/10.1112/jlms/s2-26.1.15 doi: 10.1112/jlms/s2-26.1.15
![]() |
[4] |
R. A. Smith, Estimate for exponential sums, Proc. Amer. Math. Soc., 79 (1980), 365–368. https://doi.org/10.1090/s0002-9939-1980-0567973-5 doi: 10.1090/s0002-9939-1980-0567973-5
![]() |
[5] |
J. H. Loxton, R. C. Vaughan, The estimate of complete exponential sums, Canad. Math. Bull, 26 (1985), 440–454. https://doi.org/10.4153/cmb-1985-053-7 doi: 10.4153/cmb-1985-053-7
![]() |
[6] |
R. Dabrowski, B. Fisher, A stationary phase formula for exponential sums over Z/pmZ and applications to GL(3)-Kloosterman sums, Acta Arith, 80 (1997), 1–48. https://doi.org/10.4064/aa-80-1-1-48 doi: 10.4064/aa-80-1-1-48
![]() |
[7] |
X. X. Li, The hybrid power mean of the quartic Gauss sums and the two-term exponentials sums, Adv. Differ. Equations, 2018 (2018), 236. https://doi.org/10.1186/s13662-018-1658-z doi: 10.1186/s13662-018-1658-z
![]() |
[8] |
R. J. Yuan, T. T. Wang, On the fourth power mean value of one kind two-term exponential sums, J. Northwest Univ., 53 (2023), 453–458. https://doi.org/10.16152/j.cnki.xdxbzr.2023-03-016 doi: 10.16152/j.cnki.xdxbzr.2023-03-016
![]() |
[9] | L. K. Hua, Introduction to Number Theory, Beijing: Science Press, 1979. |
[10] |
S. M. Shen, W. P. Zhang, On the quartic Gauss sums and their recurrence property, Adv. Differ. Equations, 2017 (2017), 1–9. https://doi.org/10.1186/s13662-017-1097-2 doi: 10.1186/s13662-017-1097-2
![]() |
[11] |
X. C. Du, X. X. Li, On the fourth power mean of the generalized quartic Gauss sums, Acta Math. Sin. Chin. Ser., 61 (2018), 541–548. https://doi.org/10.12386/A2018sxxb0048 doi: 10.12386/A2018sxxb0048
![]() |