Abstract
Let \(\mathbb F_{q}\) be a finite field with \(q=p^{m}\) elements, where p is an odd prime and m is a positive integer. In this paper, let \(D=\{(x_{1},x_{2},\ldots ,x_{n})\in \mathbb F_{q}^{n}\backslash \{(0,0,\ldots )\}: Tr(x_{1}^{p^{k_{1}}+1}+x_{2}^{p^{k_{2}}+1}+\cdots +x_{n}^{p^{k_{n}}+1})=c\}\), where \(c\in \mathbb F_p\), Tr is the trace function from \(\mathbb F_{q}\) to \(\mathbb F_{p}\) and each \(m/(m,k_{i})\) ( \(1\le i\le n\) ) is odd. we define a p-ary linear code \(C_{D}=\{c(a_{1},a_{2},\ldots ,a_{n}):(a_{1},a_{2},\ldots ,a_{n})\in \mathbb F_{q}^{n}\}\), where \(c(a_{1},a_{2},\ldots ,a_{n})=(Tr(a_{1}x_{1}+a_{2}x_{2}+\cdots +a_{n}x_{n}))_{(x_{1},x_{2},\ldots ,x_{n})\in D}\). We present the weight distributions of the classes of linear codes which have at most three weights.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
Throughout this paper, let \(\mathbb F_q\) be a finite field with \(q=p^m\) elements, where p is an odd prime and m is a positive integer, and let Tr be the trace function from \(\mathbb F_q\) to \(\mathbb F_p\). An [n, k, d] p-ary linear code \(\mathscr {C}\) is a k-dimensional subspace of \(\mathbb F_{p}^{n}\) and has minimum Hamming distance d. Let \(A_{i}\) denote the number of codewords with Hamming weight i in a code \(\mathscr {C}\) of length n. The weight enumerator is defined by
The sequence \((1,A_{1},\ldots ,A_{n})\) is called the weight distribution of the code \(\mathscr {C}\). A code \(\mathscr {C}\) is said to be a t-weight code if the number of nonzero \(A_{i}\) is equal to t. Weight distribution is an interesting topic and was investigated in [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]. The weight distribution of a code can not only give the error correcting ability of the code, but also allow the computation of the error probability of error detection and correction.
For a set \(D=\{d_{1},d_{2},\ldots ,d_{n}\}\subseteq \mathbb F_{q}\), define a linear code of length n over \(\mathbb F_{p}\) by
We call D the defining set of \(\mathscr {C}_{D}\). Many known linear codes could be produced by selecting the defining set. For details of these known codes, the reader is referred to [3, 13, 14].
In this paper, we always assume that \(n, m, k_1, \ldots , k_n\) are positive integers with each \(m/\gcd (m, k_i)\) odd. Then each \(f_i(x)=x^{p^{k_i}+1}\) is a planar function over \(\mathbb F_q\) (see [16]). Fixing \(c\in \mathbb F_p\), we define
where
In fact, we have some well-known results as follows. If \(n=1\) and either \(k_1=0\) or \(m/\gcd (m,k_1)\) is odd, then it is just the result in [17, 18].
In the paper, we will determine the weight distribution of the linear codes \(\mathscr {C}_D\) in three cases: (1) \(c=0\), (2) \(c\in {\mathbb F_p^*}^2\), (3) \(c\in \mathbb F_p^* \setminus {\mathbb F_p^*}^2\). In the cases (2) and (3), we use the cyclotomic numbers of order 2 to get their distributions.
2 Preliminaries
Let \(\mathbb F_{q}\) be a finite fields with q elements, where q is a power of a prime p. We define the additive character of \(\mathbb F_{q}\) as follows:
where \(\zeta _{p}\) is a complex p-th primitive root of unity and Tr denotes the trace function from \(\mathbb F_{q}\) to \(\mathbb F_{p}\). The orthogonal property of additive characters [19] is given by
Let \(\lambda : \mathbb F_{q}^{*}\longrightarrow \mathbb {C}^{*}\) be a multiplicative character of \(\mathbb F_{q}^{*}\). The trivial character \(\lambda _{0}\) defined by \(\lambda _{0}(x)=1\) for all \(x\in \mathbb F_{q}^{*}\). The orthogonal property of multiplicative characters is given by
Let \(\bar{\lambda }\) be the conjugate character of \(\lambda \) defined by \(\bar{\lambda }(x)=\overline{\lambda (x)}\). It is easy to obtain that \(\lambda ^{-1}=\bar{\lambda }\). The multiplicative group \(\widehat{\mathbb F}_{q}^{*}\) is isomorphic to \(\mathbb F_{q}^{*}\). For \(\mathbb F_{q}^{*}=\langle \alpha \rangle \), define a multiplicative character by \(\psi (\alpha )=\zeta _{q-1}\), where \(\zeta _{q-1}\) denotes the primitive \(q-1\)-th root of complex unity. Then we have \(\widehat{\mathbb F}_{q}^{*}=\langle \psi \rangle \). Set \(\eta =\psi ^{\frac{q-1}{2}}\) be the quadratic character of \(\mathbb F_q\).
Define the Gauss sum over \(\mathbb F_{q}\) by
Let \((\frac{\cdot }{p})\) denote the Legendre symbol. The quadratic Gauss sums are known and given in the following.
Lemma 1
[19] Suppose that \(q=p^{m}\) and \(\eta \) is the quadratic multiplicative character of \(\mathbb F_{q}\), where p is odd prime. Then
where \(p^{*}=(-1)^{\frac{p-1}{2}}p\) is the discriminant of a prime p.
Let \(\chi '\) be the canonical additive character of \(\mathbb F_p\) such that \(\chi (x)=\chi '({\text {Tr}}(x))\) for \(x\in \mathbb F_q\). Let \(\eta '\) be a quadratic character of \(\mathbb F_p\), then \(\eta (x)=\eta '(N_{q/p}(x))\) for \(x\in \mathbb F_q^*\).
Lemma 2
[20] Let \(x\in \mathbb F_p^*\) and \(q=p^m\), where p is odd prime.
If m is even, then \(\eta (x)=1\).
If m is odd, then \(\eta (x)=\eta '(x)\).
Moreover, \(G(\eta )=(-1)^{m-1}G(\eta ')^m\), where \(G(\eta )\) and \(G(\eta ')\) are the Gauss sums over \(\mathbb F_q\) and \(\mathbb F_p\), respectively.
We now give a brief introduction to the theory of quadratic forms over finite fields. Quadratic forms have been well studied and have applications in sequence design [11, 21] and coding theory [7, 22].
Lemma 3
Let \(d=gcd(k,m)\). Then
In this paper, we assume that k an integer and \(m/\gcd (k,m)\) odd. Then it is well-known that \(f(x)=x^{p^k+1}\) is a planar function from \(\mathbb F_q\) to \(\mathbb F_q\). In [23, 24], Coulter gave the valuations of the following Weil sums:
Lemma 4
[23, Theorem 1] Let m / d be odd. Then
Lemma 5
[24, Theorem 1] Let q be odd and suppose \(f(X)=a^{p^{k}}X^{p^{2k}}+aX\) is a permutation polynomial over \(F_{q}\). Let \(x_{0}\) be the unique solution of the equation \( f(X)=-b^{p^{k}}\). The evaluation of \(S_{k}(a,b)\) partitions into the following two cases:
-
(1)
If m / d is odd, then
$$\begin{aligned} S_{k}(a,b)= & {} \eta (a)G(\eta )\bar{\chi }\left( ax_{0}^{p^{k}+1}\right) \\= & {} \left\{ \begin{array}{ll} (-1)^{m-1}\sqrt{q}\eta (a)\bar{\chi }\left( ax_{0}^{p^{k}+1}\right) ,&{} \quad \text{ if } p\equiv 1 \pmod 4,\\ (-1)^{m-1}i^{m}\sqrt{q}\eta (a)\bar{\chi }\left( ax_{0}^{p^{k}+1}\right) ,&{} \quad \text{ if } p\equiv 3 \pmod 4. \end{array} \right. \end{aligned}$$ -
(2)
If m / d is even, then \(m=2e\), \(a^{\frac{q-1}{p^{d}+1}}\ne (-1)^{m/d}\) and
$$\begin{aligned} S_{k}(a,b)=(-1)^{m/d}p^{e}\bar{\chi }\left( ax_{0}^{p^{k}+1}\right) . \end{aligned}$$
In fact, Lemma 4 is made of some revision in [24, Theorem 1].
3 Linear codes
Let \(\mathbb F_{q}\) be the finite field with \(q=p^m\) elements, where p is an odd prime and m is an positive integer. Let Tr denote the trace function from \(\mathbb F_{q}\) to \(\mathbb F_{p}\). In this section, we always assume that \(n, k_1, \ldots , k_n\) are positive integers with each \(m/\gcd (m, k_i)\) odd. Let \(f_i(x)=x^{p^{k_i}+1},x\in \mathbb F_{q}\), \(i=1,\dots , n\). It is known from [16] that \(f_i\), \(1\le i\le n\), are planar functions from \(\mathbb F_{q}\) to \(\mathbb F_{q}\).
3.1 The first case
Define
where
Lemma 6
Let \(n_0=|D_0|\). Suppose that mn is even, then
Suppose that mn is odd, then
Proof
By Lemma 3, we have that
If m is even or m is odd and n is even, then \(\eta (y)^n=1\) for each \(y\in \mathbb F_p^*\). Hence \(n_0=\frac{q^n-p}{p}+\frac{p-1}{p}G(\eta )^n\).
If mn is odd, then m is odd. Let \(\eta '\) be a quadratic character of \(\mathbb F_p^*\), then \(\eta (y)=\eta '(y)\) for each \(y\in \mathbb F_p^*\). Hence \(\sum _{y\in \mathbb F_p^*}\eta (y)^n=\sum _{y\in \mathbb F_p^*}\eta '(y)=0\), so \(n_0=\frac{q^n-p}{p}\).
By Lemma 1, we can get the exact value of \(n_0\).
Theorem 1
Let \(\mathscr {C}_{D_{0}}\) be the linear code defined as (3.1).
If mn is even, then \(\mathscr {C}_{D_{0}}\) is a two-weight code with the Hamming weight distribution in Table 1.
If mn is odd, then \(\mathscr {C}_{D_0}\) is a three-weight code with the Hamming weight distribution in Table 2.
Proof
Firstly, we determine the weight distribution of the code \(\mathscr {C}_{D_{0}}\). Define the following parameter
where \(a=(a_1,\ldots ,a_n)\in \mathbb F_{q}^{n}\). By definition and the basic facts of additive characters, for each \(a=(a_1,\ldots ,a_n)\in \mathbb F_{q}^{n}\setminus \{(0,\ldots ,0)\}\), we have
By Lemmas 2 and 4, we have that
By \(a=(a_1,\ldots , a_n)\in \mathbb F_q^n\setminus \{(0,\ldots ,0)\}\), we have that
To compute \(N_{a}\), it is sufficient to determine the value of the exponential sum
For each \(k_i\) and \(d_i=\gcd (k_i,m)\), \(m/d_i\) is odd. Hence for \(y\in \mathbb F_p^*\), the polynomial \(f_i(x)=y^{p^{k_i}}x^{p^{2k_i}}+yx=y(x^{p^{2k_i}}+x)\) must be a permutation polynomial over \(\mathbb F_q\). In fact, suppose that there is \(0\ne b\in \mathbb F_q\) such that \(f_i(b)=0\). Then \(b^{p^{2k_i}-1}=-1\). Let \(\alpha \) be a primitive element of \(\mathbb F_q^*\) and \(b=\alpha ^t\), then
Let \(d_i=\gcd (m,k_i)\), then \(\gcd (2k_i, m)=d_i\) by \(m/d_i\) odd. Hence \(\gcd (p^{2k_i}-1, p^m-1)=(p^{d_i}-1)\) and \((p^{d_i}-1)\not \mid \frac{p^m-1}{2}\), so (3.2) is contradictory.
Since \(f_i(x)=y(x^{p^{2k_i}}+x)\) is a permutation polynomial over \(\mathbb F_q\), for each \(a_i\in \mathbb F_q\) there is the unique solution \(b_i\in \mathbb F_q\) of the equation \(x_i^{p^{2k_i}}+x_i+a_i^{p^{k_i}}=0\). In fact, there is a one-to-one correspondence between \(a_i\in \mathbb F_q\) and \(b_i\in \mathbb F_q\), and \(a_i=0\) is correspond to \(b_i=0\). Hence there is the unique solution \(wb_i\in \mathbb F_q\) of the equation \(y(x_i^{p^{2k_i}}+x_i+wa_i^{p^{k_i}})=0\), where \(w=\frac{z}{y}\in \mathbb F_p^*\).
By Lemma 5, we have that
Set
To compute the value of \(\varOmega _3\), we divide into two cases.
The first case: mn is even, i.e. either m is even or n is even. Then we have that \(\eta (y)^n=1\) for \(y\in \mathbb F_p^*\).
If \((b_1,\ldots , b_n)\in \varGamma _0\), then \(\varOmega _3=\frac{(p-1)^2}{p^2}G(\eta )^n,\)
Hence by Lemma 6, the weight of \(\mathscr {C}_{D_{0}}\) is
If \((b_1,\ldots , b_n)\in \varGamma _1\cup \varGamma _2\), then
Hence by Lemma 6, the weight of \(\mathscr {C}_{D_{0}}\) is
The second case: mn is odd. Then we have that that \(\eta (y)^n=\eta '(y)\) for \(y\in \mathbb F_p^*\).
If \((b_1,\ldots , b_n)\in \varGamma _0\), then
If \((b_1,\ldots , b_n)\in \varGamma _2\), so \({\text {Tr}}(\sum _{i=1}^nb_i^{p^{k_i}+1})=c\in {\mathbb F_p^*}^2\), then
If \((b_1,\ldots , b_n)\in \varGamma _1\), so \({\text {Tr}}(\sum _{i=1}^nb_i^{p^{k_i}+1})=c\in \mathbb F_p^*\setminus {\mathbb F_p^*}^2\), then
Secondly, we determine the frequency of each nonzero weight of \(\mathscr {C}_{D_0}\). It is sufficient to consider the values of \(|\varGamma _i|,i=0,1,2\).
By Lemma 6, it is clear that
If mn is even, then by Lemma 1,
Since \(|\varGamma _0|<q^n\). Without loss of generality, suppose that \(\varGamma _2\ne \emptyset \). For some \(c\in {\mathbb F_p^*}^2\), there are \((x_1,\ldots , x_n)\in \mathbb F_q^n\) such that \({\text {Tr}}(x_1^{p^{k_1}+1}+\cdots +x_n^{p^{k_n}+1})=c\in {\mathbb F_p^*}^2\). By the property of the trace function, the values are presented averagely from \({\mathbb F_p^*}^2\). Hence
If mn is even, then by Lemma 1,
If mn is odd, then by Lemma 1,
Since \(|\varGamma _0|+|\varGamma _2|<q^n\), \(\varGamma _1\ne \emptyset \). Similarly, the values of the trace function are presented averagely from \(\mathbb F_p^*\setminus {\mathbb F_p^*}^2\) and
If mn is even, then
If mn is odd, then by Lemma 1,
Hence, we get the Tables 1 and 2. \(\square \)
Example 1
Let \(m=3\), \(n=1\), \(p=3\), the code \(\mathscr {C}_{D_0}\) has parameters [8,3,4] and weight enumerator \(1+6z^{4}+8z^{6}+12z^{8}\). This code is almost optimal linear code,as the optimal one has parameters [8,3,5] by the Griesmer bound.
Example 2
Let \(m=2\), \(n=2\), \(p=3\), the code \(\mathscr {C}_{D_0}\) has parameters [44,4,18] and weight enumerator \(1+32z^{18}+36z^{24}\).
3.2 The second case
Fix \(c\in {\mathbb F_p^*}^2\) and define
where
Since \(\mathscr {C}_{D_2}\) is a linear code over \(\mathbb F_p\), it is independent of the choice of c. For convenience, we take \(c=1\).
By Lemma 1 and the computation of \(\varGamma _2\) as above, we can get the result.
Lemma 7
Let \(n_2=|D_2|\). Suppose that mn is even, then
Suppose that mn is odd, then
Theorem 2
Let \(\mathscr {C}_{D_{2}}\) be the linear code defined as (3.3).
If mn is even, then \(\mathscr {C}_{D_{2}}\) is a two-weight code with the Hamming weight distribution in Table 3.
If mn is odd, then \(\mathscr {C}_{D_1}\) is a two-weight code with the Hamming weight distribution in Table 4.
Proof
Firstly, we determine the weight distribution of the code \(\mathscr {C}_{D_{2}}\). Fix \(c=1\in {\mathbb F_p^*}^2\). Define the following parameter
where \(a=(a_1,\ldots ,a_n)\in \mathbb F_{q}^{n}\setminus \{(0,\ldots ,0)\}\). We have
We have that
By \((a_1,\ldots , a_n)\in \mathbb F_p^n\setminus \{(0,\ldots ,0)\}\), we have that
Similarly, we have
where \((b_1,\ldots , b_n)\) is one-to-one correspondent to \((a_1,\ldots , a_n)\), and \((0,\ldots ,0)\) is correspondence to \((0,\ldots ,0)\).
To compute the value of \(\varOmega _3'\), we divide into two cases.
The first case: mn is even. Then \(\eta (y)^n=1\) for \(y\in \mathbb F_p^*\).
If \((b_1,\ldots , b_n)\in \varGamma _0\), then
If \((b_1,\ldots , b_n)\in \varGamma _1\), then \(w^2{\text {Tr}}((\sum _{i=1}^n(b_i)^{p^{k_i}+1})\ne 1\) for any \(w\in \mathbb F_p^*\). Hence
If \((b_1,\ldots , b_n)\in \varGamma _2\), then there are only two \(\pm w_0\in \mathbb F_p^*\) such that \(w_0^2{\text {Tr}}((\sum _{i=1}^n(b_i)^{p^{k_i}+1})=1\), so
The second case: mn is odd. Then \(\eta (y)^n=\eta '(y)\) for any \(y\in \mathbb F_p^*\).
If \((b_1,\ldots , b_n)\in \varGamma _0\), then
If \((b_1,\ldots , b_n)\in \varGamma _1\), then \(w^2{\text {Tr}}(\sum _{i=1}^n(b_i)^{p^{k_i}+1})\ne 1\) for any \(w\in \mathbb F_p^*\). let \(\beta \) be a primitive root of \(\mathbb F_p^*\), then \(C_0=<\beta ^2>\) be a subgroup of \(\mathbb F_p^*\) and \(C_1=\beta C_0\), so \(\mathbb F_p^*=C_0\cup C_1\). Define
all cyclotomic numbers of order 2. We have the following results [25]: If \(p\equiv 1\pmod 4\), then \((1,0)_2=(0,1)_2=(1,1)_2=\frac{p-1}{4}\), \((0,0)_2=\frac{p-5}{4}\). If \(p\equiv 3\pmod 4\), then \((1,0)_2=(0,0)_2=(1,1)_2=\frac{p-3}{4}\), \((0,1)_2=\frac{p+1}{4}\).
Hence
If \((b_1,\ldots , b_n)\in \varGamma _2\), then there are only two \(\pm w_0\in \mathbb F_p^*\) such that \(w_0^2{\text {Tr}}((\sum _{i=1}^n(b_i)^{p^{k_i}+1})=1\). Hence
By the computations of \(\varGamma _0,\varGamma _1,\varGamma _2\), we get the Tables 3 and 4. \(\square \)
Example 3
Let \(m=1\), \(n=2\), \(p=5\), the code \(\mathscr {C}_{D_2}\) is a [4,2,2] almost optimal linear code, as the optimal one has parameters [4,2,3] by the Griesmer bound.
Example 4
Let \(m=2\), \(n=2\), \(p=3\), the code \(\mathscr {C}_{D_2}\) has parameters [24,4,12] and weight enumerator \(1+24z^{12}+50z^{18}\).
Example 5
Let \(m=3\), \(n=3\), \(p=3\), the code \(\mathscr {C}_{D_2}\) has parameters [6642,9,4374] and weight enumerator \(1+13202z^{4374}+6480z^{4401}\).
3.3 The third case
Fix \(c\in \mathbb F_p^* \setminus {\mathbb F_p^*}^2\) and define
where
Since \(\mathscr {C}_{D_1}\) is a linear code over \(\mathbb F_p\), it is independent of the choice of c.
By Lemma 1 and the computation of \(\varGamma _{1}\) as above, we can get the result similarly with Lemma 7.
Lemma 8
Let \(n_1=|D_1|\). Suppose that mn is even, then
Suppose that mn is odd, then
Theorem 3
Let \(\mathscr {C}_{D_{1}}\) be the linear code defined as (3.4).
If mn is even, then \(\mathscr {C}_{D_{1}}\) is a two-weight code with the Hamming weight distribution in Table 5.
If mn is odd, then \(\mathscr {C}_{D_1}\) is a two-weight code with the Hamming weight distribution in Table 6.
Proof
By the process of proving Theorem 2, we can get the result. \(\square \)
4 Concluding remarks
There is a recent survey on three-weight codes [3, 9, 17, 18, 22, 26, 27]. We did not find the the parameters of the binary three-weight codes of this paper in these literatures.
Linear codes can be used to construct secret sharing schemes [28]. Let \(w_{\min }\) and \(w_{\max }\) denote the minimum and maximum nonzero Hamming weights of a linear code \(\mathscr {C}\). To obtain secret sharing schemes with interesting access structures, we would like to construct linear codes which have the property that
We remark that the linear codes in this paper can be employed in secret sharing schemes using the framework in [28].
For the code of \(\mathscr {C}_{D}\) the Theorem 3.2, we have
For the code of \(\mathscr {C}_{D}\) the Theorem 3.4, we have
References
Carlet, C., Ding, C.: Highly nonlinear mappings. J. Complex. 20(2), 205–244 (2004)
Calderbank, R., Kantor, W.M.: The geometry of two-weight codes. Bull. Lond. Math. Soc. 18, 97–122 (1986)
Ding, C.: Linear codes from some 2-designs. IEEE Trans. Inf. Theory 61(6), 2065–2075 (2015)
Ding, C., Luo, J., Niederreiter, H.: Two-weight codes punctured from irreducible cyclic codes. In: Li, Y., Ling, S., Niederreiter, H., Xing, C., Zhang, S. (eds.) Proceedings of the First International Workshop on Coding Theory and Cryptography, pp. 119–124. World Scientific, Singapore (2008)
Ding, C., Niederreiter, H.: Cyclotomic linear codes of order 3. IEEE Trans. Inf. Theory 53(6), 2229–2235 (2007)
Ding, C., Yang, J.: Hamming weights in irreducible cyclic codes. Discret. Math. 313(4), 434–446 (2013)
Luo, J., Feng, K.: On the weight distributions of two classes of cyclic codes. IEEE Trans. Inf. Theory 54(12), 5332–5344 (2008)
Wang, X., Zheng, D., Hu, L., Zeng, X.: The weight distributions of two classes of binary cyclic codes. Finite Fields Appl. 34, 192–207 (2015)
Zhou, Z., Ding, C.: Seven classes of three-weight cyclic codes. IEEE Trans. Inf. Theory 61(10), 4120–4126 (2013)
Zeng, X., Hu, L., Jiang, W., Yue, Q., Cao, X.: The weight distribution of a class of \(p\)-ary cyclic codes. Finite Fields Appl. 16, 56–73 (2010)
Zeng, X., Liu, J.Q., Hu, L.: Generalized Kasami sequences: the large set. IEEE Trans. Inf. Theory 53(7), 2587–2598 (2007)
Chen, Y., Li, N., Zeng, X.: A class of binary cyclic codes with generalized Niho exponents. Finite Fields Appl. 43, 123–140 (2017)
Heng, Z., Yue, Q.: A class of linear codes with at most three weights. IEEE Commun. Lett. 14(9), 1488–1491 (2015)
Heng, Z., Yue, Q.: Evaluation of Hamming weights of a class of linear codes based on Gauss sums. Des. Codes Cryptogr. doi:10.1007/S10623-016-0222-7
Grassl, M.: Bounds on the minumum distance of linear codes. http://www.codetables.de
Coulter, R.S., Matthews, R.W.: Planar functions and planes of Lenz-Barlotti class II. Des. Codes Cryptogr. 10, 167–184 (1997)
Ding, K., Ding, C.: A class of two-weight and three weight codes and their applicationsin secret sharing. IEEE Trans. Inf. Theory 61(11), 5835–5842 (2015)
Ding, K., Ding, C.: Binary linear codes with three weights. IEEE Commun. Lett. 18(11), 1879–1882 (2014)
Lidl, R., Niederreiter, H.: Finite Fields. Addison-Wesley Publishing Inc., Reading (1983)
Berndt, B., Evans, R., Williams, K.: Gauss and Jacobi Sums. Wiley, New York (1997)
Luo, J., Tang, Y., Wang, H.: Cyclic codes and sequences: the generalized Kasami case. IEEE Trans. Inf. Theory 56(5), 2130–2142 (2010)
Zhou, Z., Ding, C.: A class of three-weight cyclic codes. Finite Fields Appl. 25, 79–93 (2014)
Coulter, R.S.: Explicit evaluations of some Weil sums. Acta Arith. 83(3), 241–251 (1998)
Coulter, R.S.: Further evaluations of Weil sums. Acta Arith. 86(3), 217–226 (1998)
Ding, C., Helleseth, T.: New generalized cycltomy and its applications. Finite Fields Appl. 4, 140–166 (1998)
Li, C., Yue, Q., Li, F.: Weight distributions of cyclic codes with respect to pairwise coprime order elements. Finite Fields Appl. 28, 94–114 (2014)
Li, C., Yue, Q., Li, F.: Hamming weights of the duals of cyclic codes with two zeros. IEEE Trans. Inf. Theory 60(7), 3895–3902 (2014)
Yuan, J., Ding, C.: Secret sharing schmes from three classes of linear codes. IEEE Trans. Inf. Theory 52(1), 206–212 (2006)
Acknowledgements
The authors are very grateful to the editor and anonymous reviewers for their valuable comments and suggestions that improved the quality of this paper. This paper is supported by Guangxi Science Research and Technology Development Project (1599005-2-13).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Zhu, X., Yang, F. A class of linear codes with two weights or three weights from some planar functions. J. Appl. Math. Comput. 56, 235–252 (2018). https://doi.org/10.1007/s12190-016-1071-2
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12190-016-1071-2