In this paper, we defined a new generalized Cochrane sum with Dirichlet characters, and gave the upper bound of the generalized Cochrane sum with Dirichlet characters. Moreover, we studied the asymptotic estimation problem of the mean value of the generalized Cochrane sum with Dirichlet characters and obtained a sharp asymptotic formula for it. By using this asymptotic formula, we also gave the mean value of the generalized Dedekind sum.
Citation: Jiankang Wang, Zhefeng Xu, Minmin Jia. On the generalized Cochrane sum with Dirichlet characters[J]. AIMS Mathematics, 2023, 8(12): 30182-30193. doi: 10.3934/math.20231542
[1] | Xuan Wang, Li Wang, Guohui Chen . The fourth power mean of the generalized quadratic Gauss sums associated with some Dirichlet characters. AIMS Mathematics, 2024, 9(7): 17774-17783. doi: 10.3934/math.2024864 |
[2] | Lei Liu, Zhefeng Xu . Mean value of the Hardy sums over short intervals. AIMS Mathematics, 2020, 5(6): 5551-5563. doi: 10.3934/math.2020356 |
[3] | Jinmin Yu, Renjie Yuan, Tingting Wang . The fourth power mean value of one kind two-term exponential sums. AIMS Mathematics, 2022, 7(9): 17045-17060. doi: 10.3934/math.2022937 |
[4] | Wenjia Guo, Xiaoge Liu, Tianping Zhang . Dirichlet characters of the rational polynomials. AIMS Mathematics, 2022, 7(3): 3494-3508. doi: 10.3934/math.2022194 |
[5] | Zhuoyu Chen, Wenpeng Zhang . A new reciprocity formula of Dedekind sums and its applications. AIMS Mathematics, 2024, 9(5): 12814-12824. doi: 10.3934/math.2024626 |
[6] | Wenjia Guo, Yuankui Ma, Tianping Zhang . New identities involving Hardy sums $S_3(h, k)$ and general Kloosterman sums. AIMS Mathematics, 2021, 6(2): 1596-1606. doi: 10.3934/math.2021095 |
[7] | Guohui Chen, Wenpeng Zhang . One kind two-term exponential sums weighted by third-order character. AIMS Mathematics, 2024, 9(4): 9597-9607. doi: 10.3934/math.2024469 |
[8] | Xue Han, Tingting Wang . The hybrid power mean of the generalized Gauss sums and the generalized two-term exponential sums. AIMS Mathematics, 2024, 9(2): 3722-3739. doi: 10.3934/math.2024183 |
[9] | Yan Ma, Di Han . On the high-th mean of one special character sums modulo a prime. AIMS Mathematics, 2023, 8(11): 25804-25814. doi: 10.3934/math.20231316 |
[10] | 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 |
In this paper, we defined a new generalized Cochrane sum with Dirichlet characters, and gave the upper bound of the generalized Cochrane sum with Dirichlet characters. Moreover, we studied the asymptotic estimation problem of the mean value of the generalized Cochrane sum with Dirichlet characters and obtained a sharp asymptotic formula for it. By using this asymptotic formula, we also gave the mean value of the generalized Dedekind sum.
Let h,q be integers with q>0. The classical Dedekind sum s(h,q) is defined by
s(h,q)=q−1∑a=1((aq))((haq)), |
where
((x))={x−[x]−12,ifxis not an integer;0,ifxis an integer |
and [x] is the largest integer not exceeding x. The Dedekind sum plays an important role in the Dedekind η function and has applications to many parts of mathematics (see [7,11,12]).
For any nonnegative integer n, let Bn and Bn(X) be the n-th Bernoulli number and polynomial, respectively, which is defined by
tet−1=∞∑n=0Bntnn!, tetXet−1=∞∑n=0Bn(X)tnn!, |
where ¯Bn(X)=Bn(X−[X]) is the n-th periodic Bernoulli function in the interval (0,1] for n>1, ¯B1(X)=B1(X−[X]) and if X is an integer, then ¯B1(X)=0. For positive integers m,n, we have the generalized Dedekind sum
S(h,m,n,q)=q−1∑a=1¯Bm(aq)¯Bn(ahq). |
Let p be an odd prime and let χ be any even Dirichlet character mod p. For any integers n,k, Xie and Zhang [8] showed that if n is an odd integer, then
p−1∑h=1χ(h)|S(h,n,n,p)|2k=2(n!)4k42(n−1)k(pζ(2n)2π2n)2k|L(2nk,χ)|2ζ(4nk)+O(p2k−nexp(6lnplnlnp)), | (1.1) |
they also have the following result for any even integer n
p−1∑h=1χ(h)|S(h,n,n,p)|2k=2(n!)4k42(n−1)k(pζ(2n)2π2n)2k|L(2nk,χ)|2ζ(4nk)+O(p2k−1). | (1.2) |
In October 2000, Professor Todd Cochrane first introduced a sum analogous to the Dedekind sum as follows:
C(h,q)=q∑a=1′((ˉaq))((haq)), |
where ˉa satisfies aˉa≡1(modq), ∑′qa=1 denotes the summation over all 1≤a≤q such that (a,q)=1. Many scholars studied the properties of C(h,q). Zhang and Yi [13] gave the following upper bound estimate:
|C(h,q)|≪q12d(q)ln2q, |
where d(q) is the divisor function. Ma et al. [4] gave the upper bound estimate of the incomplete Cochrane sum.
Xu and Zhang [9] defined the high-dimensional Cochrane sum by the following equation:
C(h,k,q)=q∑a1=1′⋯q∑ak=1′((a1q))⋯((akq))((h¯a1⋯akq)). |
For any fixed positive integer k with (q,k(k+1))=1, they gave the following upper bound estimate:
|C(h,k,q)|≪2(k+1)2πk+1qk2d(q)(2k+2k)ω(q)lnk+1q, |
where ω(q) denotes the number of all different prime divisors of q. Liu [5] improved this result.
For positive integers m,n, the main purpose of this paper is to study the generalized Cochrane sum with any Dirichlet character χ mod q as follows:
C(h,m,n,q,χ)=q∑a=1′χ(a)¯Bm(¯aq)¯Bn(ahq), |
which is an interesting generalization of Cochrane sum. Ren and Yi [6] studied the mean square value of C(h,1,1,p,χ), for a prime p≡1mod4 and the Legendre's symbol χmodp, they obtained
p−1∑h=1|C(h,1,1,p,χ)|2=1180p2∏p1∈A(p21+1p21−1)2+O(p1+o(1)), | (1.3) |
where A is the set of quadratic residues of p, p1 is prime which is not equal to p. Liu and Zhang [2,3] studied the mean square value of C(h,m,n,q,χ) and its hybrid mean value formula when χ is the principal Dirichlet character. In this paper, by using an upper bound estimate of the Kloosterman sum with Dirichlet characters, we show that
Theorem 1.1. Let p be an odd prime and let χ be any Dirichlet character mod p. For any integers m,n, we have
C(h,m,n,p,χ)≪m!n!(2π)−(m+n)p12ln2p, |
if χ(−1)≠(−1)n+m, then C(h,m,n,p,χ)=0.
We also obtain the mean square value of C(h,m,n,p,χ) as follows:
Theorem 1.2. Let p be an odd prime and let χ be any Dirichlet character mod p. For any integers m,n with χ(−1)=(−1)n+m, we have
p−1∑h=1|C(h,m,n,p,χ)|2=8(m!n!)2p2(2πi)2(m+n)ζ(2m)ζ(2n)ζ(2m+2n)|L(m+n,χ)|2+O(p2−min(n,m)exp(4lnplnlnp)). |
If χ1 mod a prime p≡1mod4 is the Legendre's symbol, then according to the value of the Dirichlet L-function L(2,χ1), we also can get (1.3):
Corollary 1.3. Let p≡1mod4 be a prime and the Legendre's symbol χ1 mod p. We have
p−1∑h=1|C(h,1,1,p,χ1)|2=1180p2∏p1∈A(p21+1p21−1)2+O(pexp(4lnplnlnp)), |
where A is the set of quadratic residues of p, p1 is prime, which is not equal to p.
Moreover, we give the mean square value of S(h,m,n,p,χ) as follows:
Theorem 1.4. Let p be an odd prime and let χ be any Dirichlet character mod p. For any integers m,n with χ(−1)=(−1)n+m, we have
p−1∑h=1χ(h)S(¯h,m,m,p)S(h,n,n,p)=8(m!n!)2p2(2πi)2(m+n)ζ(2m)ζ(2n)ζ(2m+2n)|L(m+n,χ)|2+O(p2−min(n,m)exp(4lnplnlnp)). |
Obviously, Theorem 1.4 generalizes and improves (1.1) and (1.2) when k=1.
To prove theorems, we need the following several lemmas.
Lemma 2.1. Let p be an odd prime and an integer h with (h,p)=1, and let χ1 be any Dirichlet character mod p. For any positive integers m,n, if χ1(−1)≠(−1)m+n then C(h,m,n,p,χ1)=0, and if χ1(−1)=(−1)n+m, then we have
C(h,m,n,p,χ1)=4m!n!(2πi)m+n1ϕ(p)∑χmodpχχ1(−1)=(−1)m¯χ(h)(+∞∑r=1G(χχ1,r)rm)(+∞∑s=1G(χ,s)sn), |
and
C(h,m,n,p,χ1)=4m!n!(2πi)m+n1ϕ(p)∑χmodpχχ1(−1)=(−1)m¯χ(h)τ(χ)τ(χχ1)L(m,¯χχ1)L(n,¯χ), |
where τ(χ)=G(χ,1), G(χ,r)=∑p−1a=1χ(a)e(rap) denotes Gauss sum, L(n,χ)=∑+∞t=1χ(t)tn is a Dirichlet L-function.
Proof. Applying the orthogonality of multiplicative characters, it follows that
C(h,m,n,p,χ1)=p−1∑a=1χ1(a)¯Bm(¯ap)¯Bn(ahp)=1ϕ(p)∑χmodp{p−1∑a=1χχ1(a)¯Bm(ap)}{p−1∑b=1χ(b)¯Bn(hbp)}. |
Noting that [1]
¯Bn(x)=−n!(2πi)n+∞∑r=−∞r≠0e(xr)rn, |
and G(χ,−hn)=¯χ(−h)G(χ,n). We have
C(h,m,n,p,χ1)=1ϕ(p)∑χmodp{p−1∑a=1(−m!(2πi)m+∞∑r=−∞r≠0χχ1(a)e(rap)rm)}×{p−1∑b=1(−n!(2πi)n+∞∑s=−∞s≠0χ(b)e(sbhp)sn)}=m!n!(2πi)m+n1ϕ(p)∑χmodp{+∞∑r=−∞r≠01rmp−1∑a=1χχ1(a)e(rap)}×{+∞∑s=−∞s≠01snp−1∑b=1χ(b)e(sbhp)}=m!n!(2πi)m+n1ϕ(p)∑χmodp{+∞∑r=−∞r≠0G(χχ1,r)rm}{+∞∑s=−∞s≠0G(χ,sh)sn}=m!n!(2πi)m+n1ϕ(p)∑χmodp¯χ(h)(1+χχ1(−1)(−1)m)(+∞∑r=1G(χχ1,r)rm)×(1+χ(−1)(−1)n)(+∞∑s=1G(χ,s)sn)=4m!n!(2πi)m+n1ϕ(p)∑χmodpχχ1(−1)=(−1)m¯χ(h)(+∞∑r=1G(χχ1,r)rm)(+∞∑s=1G(χ,s)sn), |
where χ1(−1)=(−1)m+n. If χ1(−1)≠(−1)m+n, then C(h,m,n,p,χ1)=0.
Moreover, we also have
C(h,m,n,p,χ1)=4m!n!(2πi)m+n1ϕ(p)∑χmodpχχ1(−1)=(−1)m¯χ(h)τ(χ)τ(χχ1)L(m,¯χχ1)L(n,¯χ), |
where τ(χ)=G(χ,1).
Lemma 2.2. Let p be a prime and let χ be any Dirichlet character mod p. For any integers r,s, we have
|p−1∑a=1χ(a)e(ra+s¯ap)|≤2√p. |
Proof. See Lemma 1 of [10].
Lemma 2.3. Let p be an odd prime and an integer h with (h,p)=1, and let χ be any Dirichlet character mod p, then we have
∑χmodpχχ1(−1)=(−1)m¯χ(h)(∞∑r=1G(χχ1,r)rm)(∞∑s=1G(χ,s)sn)≪p32ln2p. |
Proof. For any fixed parameter N≥p, according to Abel's identity, we have
∞∑r=1G(χχ1,r)rm=∑1≤r≤NG(χχ1,r)rm+m∫+∞N∑N<r≤yG(χχ1,r)rm+1dy. |
Since |G(χ,r)|≤p12, we have
∑1≤r≤NG(χχ1,r)rm≪p12∑1≤r≤N1rm≪p12lnN, |
and from the estimates for trigonometric sums we have
∑N<r≤yG(χχ1,r)=p−1∑a=1χχ1(a)∑N<r≤ye(arp)=p−1∑a=1χχ1(a)e((N+1)ap)−e(([y]+1)ap)1−e(ap)≪p−1∑a=11|sinπap|≪p−1∑a=1pa≪plnp, |
it follows that
m∫+∞N∑N<r≤yG(χχ1,r)rm+1dy≪plnpNm, |
then we have
∑χmodpχχ1(−1)=(−1)m¯χ(h)(+∞∑r=1G(χχ1,r)rm)(+∞∑s=1G(χ,s)sn)=∑χmodpχχ1(−1)=(−1)m¯χ(h)(∑1≤r≤NG(χχ1,r)rm+m∫+∞N∑N<r≤yG(χχ1,r)rm+1dy)×(∑1≤r≤NG(χ,s)sn+n∫+∞N∑N<r≤yG(χ,s)sn+1dy)=∑χmodpχχ1(−1)=(−1)m¯χ(h)(N∑r=1G(χχ1,r)rm)(N∑s=1G(χ,s)sn)+n∑χmodpχχ1(−1)=(−1)m¯χ(h)(N∑r=1G(χχ1,r)rm)(∫+∞N∑N<r≤yG(χ,s)sn+1dy)+m∑χmodpχχ1(−1)=(−1)m¯χ(h)(N∑s=1G(χ,s)sn)(∫+∞N∑N<r≤yG(χχ1,r)rm+1dy)+mn(∫+∞N∑N≤r≤yG(χχ1,r)rm+1dy)(∫+∞N∑N<r≤yG(χ,s)sn+1dy)≪∑χmodpχχ1(−1)=(−1)m¯χ1(h)(N∑r=1G(χχ1,r)rm)(N∑s=1G(χ,s)sn)+O(p32lnNlnpmin(Nm,Nn)). |
Note that
∑χmodpχχ1(−1)=(−1)mχ(a)={12ϕ(p),a≡1modp;(−1)mχ1(−1)12ϕ(p),a≡−1modp;0,otherwise. | (2.1) |
Then, from the orthogonality of Dirichlet characters we find
∑χmodpχχ1(−1)=(−1)m¯χ1(h)(N∑r=1G(χχ1,r)rm)(N∑s=1G(χ,s)sn)=∑χmodpχχ1(−1)=(−1)m¯χ(h)(∑1≤r≤Np−1∑a=1χχ1(a)e(rap)rm)(∑1≤s≤Np−1∑b=1χ(b)e(sbp)sn)=∑1≤r≤N∑1≤s≤N1rmsnp−1∑a=1p−1∑b=1χ1(a)e(ra+sbp)∑χmodpχχ1(−1)=(−1)mχ(a)χ(b)¯χ(h)=ϕ(p)2∑1≤r≤N∑1≤s≤N1rmsnp−1∑a=1p−1∑b=1ab≡hmodpχ1(a)e(ra+sbp)+(−1)mχ1(−1)ϕ(p)2∑1≤r≤N∑1≤s≤N1rmsnp−1∑a=1p−1∑b=1ab≡−hmodpχ1(a)e(ra+sbp). |
According to Lemma 2.2, we obtain
∑χmodpχχ1(−1)=(−1)m¯χ(h)(N∑r=1G(χχ1,r)rm)(N∑s=1G(χ,s)sn)≪p32ln2N. |
Taking N=p2, we can get Lemma 2.3
∑χmodpχχ1(−1)=(−1)m¯χ(h)(+∞∑r=1G(χχ1,r)rm)(+∞∑s=1G(χ,s)sn)≪p32⋅ln2p. |
Lemma 2.4. Let p be a prime and let χ be any Dirichlet character mod p. For any integers m,n, we have
∑χmodpχχ1(−1)=(−1)n|L(n,χχ1)|2|L(m,χ)|2=(p−1)ζ(2m)ζ(2n)2ζ(2m+2n)|L(n+m,χ1)|2+O(p1−min(n,m)exp(4lnplnlnp)). |
Proof. First, we assume that n>m, according to Abel's identity we can write
L(n,χχ1)L(m,χ)=∞∑t=1χ(t)D(t)tm=N∑t=1χ(t)D(t)tm+m∫∞NB(y,χ)ym+1dy, |
where D(t)=∑d|tχ1(d)dn−m, B(y,χ)=∑N<t≤yχ(t)D(t). We know that
B(y,χ)=∑N<t≤yχ(t)D(t)=∑N<t≤yχ(t)∑d∣tχ1(d)dn−m=∑t≤√yχ(t)∑l≤y/tχχ1(l)ln−m+∑l≤√yχχ1(l)ln−m∑t≤y/lχ(t)−∑t≤√Nχ(t)∑l≤N/tχχ1(l)ln−m−∑l≤√Nχχ1(l)ln−m∑t≤N/lχ(t)−∑t≤√yχ(t)∑l≤√yχχ1(l)ln−m+∑l≤√Nχχ1(l)ln−m∑t≤√Nχ(t). |
It follows that
|B(y,χ)|≪√ylny, |
∑χmodpχχ1(−1)=(−1)n|B(y,χ)|2≪yϕ(p)ln2y. |
From the Cauchy inequality we have
∑χmodpχχ1(−1)=(−1)n|m∫∞NB(y,χ)ym+1dy|2≪{m∫∞N1ym+1(∑χmodpχχ1(−1)=(−1)n|B(y,χ)|2)1/2dt}2≪{m∫∞Ny−m−12ϕ12(p)lnydt}2≪ϕ(p)ln2NN2m−1. |
From (2.1), we have
∑χmodpχχ1(−1)=(−1)n|N∑n1=1χ(n1)D(n1)nm1|2=ϕ(p)2∑1≤n1,n2≤Nn1≡n2(modp)(n1n2,p)=1D(n1)¯D(n2)nm1nm2+χ1(−1)(−1)nϕ(p)2∑1≤n1,n2≤Nn1≡−n2(modp)(n1n2,p)=1D(n1)¯D(n2)nm1nm2=ϕ(p)2∑1≤n1≤N(n1,p)=1|D(n1)|2n2m1+O(ϕ(p)N∑n2=1[N/p]∑l=1d(n2)d(lp+n2)nm2(lp+n2)m)+O(ϕ(p)p−1∑n1=1d(n1)d(q−n1)nm1(p−n1)m)+O(ϕ(p)N∑n1=1[N/p]∑l=1+n1/pd(n1)d(lp−n1)nm1(lp−n1)m)=ϕ(p)2∑1≤n1≤∞(n1,p)=1|D(n1)|2n2m1+O(ϕ(p)pmexp(2lnNlnlnN)), |
noting that |D(n1)|≤d(n1)=∑t∣n11≤exp((1+ϵ)ln2lnNlnlnN), it follows that
∑χmodpχχ1(−1)=(−1)n(N∑t=1χ(t)D(t)tm)(m∫∞NB(y,χ)ym+1dy)≪(lnN)2∫∞N1ym+1(∑χmodpχ(−1)=(−1)n|B(y,χ)|)dy≪ϕ(p)N12−m(lnN)3. |
Taking N=p2 and ϕ(p)=p−1, then
∑χmodpχχ1(−1)=(−1)n|L(n,χχ1)|2|L(m,χ)|2=p−12∑1≤n1≤∞(n1,p)=1|D(n1)|2n2m1+O(p1−mexp(4lnplnlnp)). |
From the Euler product formula, we can get
∞∑n1=1(n1,p)=1|D(n1)|2n2m1=∏p1∤p(1+|D(p1)|2p2m1+|D(p21)|2p4m1+⋯+|D(pk1)|2p2mk1+⋯) |
and
D(pk1)=1+χ1(p1)pn−m1+(χ1(p1)pn−m1)2+⋯+(χ1(p1)pn−m1)k=1−(χ1(p1)pn−m1)k+11−χ1(p1)pn−m1, |
and it is straightforward to show that
∞∑n1=1(n1,p)=1|D(n1)|2n2m1=∏p1∤p11−1p2m1+1p2n−2m11−1p2n1−χ1(p1)pn−m11−χ1(p1)pn+m1−¯χ1(p1)pn−m11−¯χ1(p1)pn+m1|1−χ1(p1)pn−m1|2=∏p1∤p1−1p2m+2n1(1−1p2m1)(1−1p2n1)1|1−χ1(p1)pn+m1|2=ζ(2m)ζ(2n)ζ(2m+2n)|L(n+m,χ1)|2(1−1p2m)(1−1p2n)1−1p2m+2n=ζ(2m)ζ(2n)ζ(2m+2n)|L(n+m,χ1)|2+O(p−2m). |
Similarly, we also have Lemma 2.4 for n≤m.
Now, we prove our theorems by using the above lemmas.
Proof of Theorem 1.1. Combining Lemma 2.1 and Lemma 2.3, we have
C(h,m,n,p,χ)≪m!n!(2π)−(m+n)p12ln2p. |
Proof of Theorem 1.2. From Lemma 2.1 and Lemma 2.4, we have
p−1∑h=1|C(h,m,n,p,χ)|2=16(m!n!)2(2πi)2(m+n)1ϕ2(p)p−1∑h=1|∑χ1modpχχ1(−1)=(−1)m¯χ1(h)τ(χ1)τ(χχ1)L(m,¯χχ1)L(n,¯χ1)|2=16(m!n!)2(2πi)2(m+n)p2ϕ(p)∑χ1modpχχ1(−1)=(−1)m|L(m,¯χχ1)L(n,¯χ1)|2=8(m!n!)2p2(2πi)2(m+n)ζ(2m)ζ(2n)ζ(2m+2n)|L(m+n,χ)|2+O(p2−min(n,m)exp(4lnplnlnp)). |
This proves Theorem 1.2.
Proof of Theorem 1.4. From the definition of C(h,m,n,p,χ) and the orthogonality of Dirichlet characters, we have
|C(h,m,n,p,χ)|2=|p−1∑a=1χ(a)¯Bm(¯ap)¯Bn(ahp)|2={p−1∑a=1χ(a)¯Bm(¯ap)¯Bn(ahp)}{p−1∑b=1¯χ(b)¯Bm(¯bp)¯Bn(bhp)}=p−1∑a=1p−1∑b=1χ(a)¯Bm(¯abp)¯Bn(abhp)¯Bm(¯bp)¯Bn(bhp)=1ϕ(p)∑χ1modpp−1∑a=1χ(a){p−1∑b=1χ1(b)¯Bm(¯abp)¯Bm(bp)}×{p−1∑c=1χ1(c)¯Bn(hcp)¯Bn(ahcp)}=1ϕ(p)∑χ1modp¯χ1(h)p−1∑a=1χ(a){p−1∑b=1χ1(b)¯Bm(¯abp)¯Bm(bp)}×{p−1∑c=1χ1(c)¯Bn(cp)¯Bn(acp)}, |
it follows that
p−1∑h=1|C(h,m,n,p,χ)|2=1ϕ(p)p−1∑h=1∑χ1modp¯χ1(h)p−1∑a=1χ(a){p−1∑b=1χ1(b)¯Bm(¯abp)¯Bm(bp)}×{p−1∑c=1χ1(c)¯Bn(cp)¯Bn(acp)}=p−1∑a=1χ(a){p−1∑b=1¯Bm(¯abp)¯Bm(bp)}{p−1∑c=1¯Bn(cp)¯Bn(acp)}. |
Thus, from Theorem 1.2, we have
p−1∑h=1χ(h)S(¯h,m,m,p)S(h,n,n,p)=8(m!n!)2p2(2πi)2(m+n)ζ(2m)ζ(2n)ζ(2m+2n)|L(m+n,χ)|2+O(p2−min(n,m)exp(4lnplnlnp)). |
The authors declare they have not used Artificial Intelligence (AI) tools in the creation of this article.
This work is supported by the National Natural Science Foundation of China (11971381).
The authors declare that there are no conflicts of interest regarding the publication of this paper.
[1] | T. M. Apostol, Introduction to Analytic Number Theory, Springer-Verlag, New York, 1976. |
[2] | H. Y. Liu, W. P. Zhang, On the mean square value of a generalized Cochrane sum, Soochow J. Math., 30 (2004), 165–175. |
[3] |
H. Y. Liu, W. P. Zhang, On a generalized Cochrane sum and its hybrid mean value formula, Ramanujan J., 9 (2005), 373–380. https://doi.org/10.1007/s11139-005-1874-5 doi: 10.1007/s11139-005-1874-5
![]() |
[4] |
Y. K. Ma, W. P. Zhang, T. P. Zhang, Upper bound estimate of incomplete Cochrane sum, Open Math., 15 (2017), 852–858. https://doi.org/10.1515/math-2017-0068 doi: 10.1515/math-2017-0068
![]() |
[5] |
H. Y. Liu, A note on the upper bound estimate of high-dimensional Cochrane sums, J. Number Theory, 125 (2007), 7–13. https://doi.org/10.1016/j.jnt.2006.10.008 doi: 10.1016/j.jnt.2006.10.008
![]() |
[6] |
D. M. Ren, Y. Yi, On the mean value of general Cochrane sum, Proc. Japan Acad. Ser. A Math. Sci., 86 (2010), 1–5. https://doi.org/10.3792/pjaa.86.1 doi: 10.3792/pjaa.86.1
![]() |
[7] | H. Rademacher, E. Grosswald, Dedekind sums, Carus Math. Monographs, 1972. |
[8] | M. Xie, W. P. Zhang, On the 2kth mean value formula of general Dedekind sums, Acta Math. Sin. (Chinese Series), 44 (2001), 85–94. |
[9] |
Z. F. Xu, W. P. Zhang, On the order of the high-dimensional Cochrane sum and its mean value, J. Number Theory, 117 (2006), 131–145. https://doi.org/10.1016/j.jnt.2005.05.005 doi: 10.1016/j.jnt.2005.05.005
![]() |
[10] |
H. Zhang, W. P. Zhang, Some new sums related to D. H. Lehmer problem, Czech. Math. J., 65 (2015), 915–922. https://doi.org/10.1007/s10587-015-0217-y doi: 10.1007/s10587-015-0217-y
![]() |
[11] |
W. P. Zhang, On the mean values of Dedekind sums, J. Théor. Nombres Bordeaux, 8 (1996), 429–422. https://doi.org/10.5802/JTNB.179 doi: 10.5802/JTNB.179
![]() |
[12] |
W. P. Zhang, A note on the mean square value of the Dedekind sums, Acta Math. Hungar., 86 (2000), 275–289. https://doi.org/10.1023/A:1006724724840 doi: 10.1023/A:1006724724840
![]() |
[13] | W. P. Zhang, Y. Yi, On the upper bound estimate of Cochrane sums, Soochow J. Math., 28 (2002), 297–304. |