On a classification of planar functions in characteristic three

Abstract

Planar functions are functions over a finite field that have optimal combinatorial properties and they have applications in several branches of mathematics, including algebra, projective geometry and cryptography. There are two relevant equivalence relations for planar functions, that are isotopic equivalence and CCZ-equivalence. Classification of planar functions is performed via CCZ-equivalence which arises from cryptographic applications. In the case of quadratic planar functions, isotopic equivalence, coming from connections to commutative semifields, is more general than CCZ-equivalence and isotopic transformations can be considered as a construction method providing up to two CCZ-inequivalent map**s. In this paper, we first survey known infinite classes and sporadic cases of planar functions up to CCZ-equivalence, aiming to exclude equivalent cases and to identify those with the potential to provide additional functions via isotopic equivalence. In particular, for fields of order $3^{n}$ with $n\leq 11$, we completely resolve if and when isotopic equivalence provides different CCZ-classes for all currently known planar functions. Further, we perform an extensive computational investigation on some of these fields and find seven new sporadic planar functions over $\mathbb{F}_{3^{6}}$ and two over $\mathbb{F}_{3^{9}}$. Finally, we give new simple quadrinomial representatives for the Dickson family of planar functions.

Keywords: planar functions, semifields, classification, equivalence

1 Introduction

Functions over finite fields have numerous applications in many areas of research within mathematics and computer science. Notably, they have been heavily studied for the design of cryptographic ciphers. For many years, the design of cryptographic ciphers has been mostly based on binary fields, due to the efficiency of implementing binary operations in hardware and software. More recently there has been a renewed interest in the case of odd characteristic, especially for building MPC-friendly cryptography primitives and in the field of side-channel countermeasures, see Grassi, Rechberger, Rotaru, Scholl and Smart [GRRSS-2016-mfskp] or Masure, Méaux, Moos and Standaert [MMMS-2023-eaemi] for example. Thus, it seems desirable to investigate functions over finite fields of odd characteristic with useful cryptographic properties. In 1989, Meier and Staffelbach [MS-1989-nlccs] showed the importance of using highly nonlinear functions in designing cryptographic primitives. Moreover, after the discovery of differential cryptanalysis by Biham and Shamir [BS-1991-dcodl] in 1991, Nyberg [N-1993-dumic] introduced the mathematical concept of differential uniformity of a function between finite fields with the idea that functions with optimal differential uniformity would provide the strongest resistance to differential cryptanalysis. In binary fields, the optimal differential uniformity is $2$, and functions obtaining this optimal value are called Almost Perfect Nonlinear (APN) functions. Over fields of odd characteristics, the optimal differential uniformity is 1, and functions that obtain this value are called Perfect Nonlinear (PN) or planar functions. The applications of planar functions, however, are not limited to their possible uses in cryptography. Indeed, when Dembowski and Ostrom [DO-1968-poowc] first introduced planar functions in 1968 as tools for constructing projective planes admitting collineation groups with specific properties. Additionally, Coulter and Henderson [CH-2008-cpas] established a 1-to-1 correspondence between commutative presemifields of odd order and quadratic planar functions. This connection was subsequently used by Budaghyan and Helleseth [BH-2011-ncsdb] to construct the first new infinite family of commutative semifields of arbitrary odd characteristic since the 1950s.

Information about the known families and constructions of planar functions (and commutative semifields) is scattered. There has not been a serious attempt to provide an exhaustive survey of the known classes over small fields since 2010, when Pott and Zhou [PZ-2010-scopf] gave an account of what was known up to CCZ-equivalence for fields of order $3^{n}$ with $n\leq 6$. There is also some confusion over the historical record, in part because some papers introducing “new” examples of planar functions have failed to fully address the question of equivalence with known classes. A central motivation of this paper is to rectify the above two problems by presenting, as best we can, a complete picture of the current state of knowledge with regards to planar functions over fields of order $3^{n}$ with $n\leq 11$. As part of this account, we give a nomenclature of the known families of planar functions, which is as accurate as it can be. With this, we hope to provide a complete account of the current situation, including whatever is known about the intersections between the known families. Following a section on background material and notation, this nomenclature is given in Section 3 along with explanations and remarks.

In Section 4 we introduce new planar DO polynomials representing the Dickson semifields. The best previous representatives for the entire class were all hexanomials. We use a linear transformation on that previous class to reduce them to quadrinomials. Our motivation stems from the fact that most computations involving equivalence are made easier by two things, namely reducing the number of terms of the polynomials involved and restricting the coefficients of the polynomials to as small a subfield as possible.

Finally, in Section 5 we turn to our central goal of giving a complete account of the current knowledge of planar functions, up to CCZ-equivalence, over fields of order $3^{n}$ with $n\leq 11$. The latest effort to produce such a classification, to the best of our knowledge, was made by Pott and Zhou in 2010 [PZ-2010-scopf], who gave a classification up to CCZ-equivalence of all planar functions known at the time in characteristic 3, for dimensions up to 6. We update this classification, adding the representatives of families and sporadic instances discovered in the intervening years. Testing for equivalence in dimensions larger than 6 has been exceedingly difficult until the recent discovery of a more efficient equivalence test by Ivkovic and Kaleyski [IK-2022-darle]. Using this improved equivalence test, we have expanded the classification up to dimension 11. Notably, we find that the Coulter-Henderson-Cosick(CHK) semifield [CHK-2007-ppfcs], considered a sporadic instance until now, belongs to the Budaghyan-Helleseth family [BH-2008-npnmo]. We classify planar functions up to CCZ-equivalence because it is the natural equivalence relation to consider when investigating differential properties. However, for quadratic planar functions, there is a more general equivalence relation based on the isotopy of the corresponding semifields, namely isotopic equivalence. The families presented in Section 3 are equivalent with respect to isotopic equivalence. This means that a complete classification, consistent with the list of families provided is only possible if we investigate the number of CCZ-classes (or, equivalently, strong isotopic equivalence classes), inside the isotopic equivalence classes. The Coulter-Henderson Theorem [CH-2008-cpas] provides necessary conditions for an isotopic equivalence class to split into two strong isotopic equivalence classes, as well as conditions on the form of isotopisms that are not also strong isotopisms. Using the latter, we can computationally prove whether an isotopy class splits into two strong isotopy classes for $n\leq 11$. We do so either by finding a split or by exhausting the search space for possible isotopisms. We find that some of the instances of the Budaghyan-Helleseth (BH) family and the instances of the Cohen-Ganley (CG) family over $\mathbb{F}_{3^{8}}$ define isotopic equivalence classes that split into two strong isotopic equivalence classes. To the best of our knowledge, this was not previously known. The complete account, along with tables of the invariants used, is reported in the tables of Appendix A. With this in place, we proceed to search for new planar functions up to CCZ-equivalence. Our approach is to perform an expansion search, a procedure analogous to the one employed by Aleksandersen, Budaghyan and Kaleyski [ABK-2022-sfafb] for APN functions. This procedure, and the complete results of the searches, are outlined in Section 5.1. As a result of this search, we find seven new CCZ-inequivalent examples of sporadic planar functions over $\mathbb{F}_{3^{6}}$ and two over $\mathbb{F}_{3^{9}}$. Finding so many new examples of planar functions over such a low dimension is a little surprsing, and we believe a more involved computational search might reveal even more instances. Additionally, we find that some of the new instances belong to an isotopy class that splits into two CCZ-equivalence classes.

2 Preliminaries

This section introduces the necessary background on planar function and semifields, as well as establishing notation. We refer the reader to Budaghyan [B-2014-caaoc, Chapter 2] for further details.

2.1 Planar functions

Let $p$ be an odd prime and $n$ be a positive integer. We denote by $\mathbb{F}_{p^{n}}$ the finite field with ${p^{n}}$ elements, and by $\mathbb{F}_{p}^{n}$ the $n$-dimensional vector space over $\mathbb{F}_{p}$. It is well known that one can identify the vector space $\mathbb{F}_{p}^{n}$ with the field $\mathbb{F}_{p^{n}}$ through the use of a basis over $\mathbb{F}_{p}$. Let $k$ be a divisor of $n,$ we denote the trace function of $\mathbb{F}_{p^{n}}$ over $\mathbb{F}_{p^{k}}$ as $\mathrm{Tr}^{n}_{k}(x)=x+x^{p^{k}}+\cdots+x^{p^{n-k}}$. Moreover, we denote by $\mathrm{Tr}=\mathrm{Tr}_{n}=\mathrm{Tr}^{n}_{1}$ the absolute trace function.

Any function $F\colon\mathbb{F}_{p}^{n}\to\mathbb{F}_{p}^{n}$ can be represented uniquely as a univariate polynomial $F(x)\in\mathbb{F}_{p^{n}}\left[X\right]$ of the form

$$F(x)=\sum_{i=0}^{p^{n}-1}\alpha_{i}x^{i},$$ (1)

where $\alpha_{i}\in\mathbb{F}_{p^{n}}$. For any positive integer $i$ strictly less than $p^{n}$, we set $\mathrm{d}_{p}(i)$ to be equal to $\sum_{j=0}^{n-1}c_{j}$ for an appropriate choice of $0\leq c_{j}<p$ such that $i=\sum_{j=0}^{n-1}c_{j}p^{j}$. We denote by $\mathrm{d}^{\circ}(F)$ the algebraic degree of $F$, that is defined as

$$\mathrm{d}^{\circ}(F)=\max\{\mathrm{d}_{p}(i)\colon 0\leq i<p^{n},\,\alpha_{i}\neq 0\}$$

where the $\alpha_{i}$ are as in (1). If $F$ has algebraic degree at most $1$, then $F$ is called affine and if additionally $F(0)=0$ we also say that $F$ is linear. If $F$ has algebraic degree $2$ (resp. $3$), then $F$ is said to be quadratic (resp. cubic). The function $F$ is called Dembrowski-Ostrom (DO) polynomial if all the non-zero terms in its polynomial representation have algebraic degree $2$, that is

$$F(x)=\sum_{0\leq i\leq j<p^{n}}\alpha_{i,j}x^{p^{j}+p^{i}}$$

where $\alpha_{i,j}\in\mathbb{F}_{p^{n}}$. We denote by $D_{a}F(x)$ the derivative of $F$ in direction $a\in\mathbb{F}_{p^{n}}\setminus\{0\}$,

$$D_{a}F(x)=F(x+a)-F(x).$$

Let $\delta$ be a positive integer. A function $F$ is differentially $\delta$-uniform if

$$\delta\geq\max_{a,b\in\mathbb{F}_{p^{n}},\,a\neq 0}|\{x\in\mathbb{F}_{p^{n}}\colon D_{a}F(x)=b\}|.$$

A $1$-uniform function is also called a Perfect Nonlinear (PN) function or planar function. In the case of $F$ being a DO polynomial, then $F$ is planar if and only if for all $a\in\mathbb{F}_{p^{n}}\setminus\{0\}$ the only solution to the equation $D_{a}F(x)=D_{a}F(0)$ is $x=0.$

2.2 Semifields

Let $S$ be a nonempty set and $+,\star$ be two binary operations over $S$. Then $(S,+,\star)$ is called a presemifield if the following holds

• $(S,+)$ is an Abelian group;

• For any $a,b\in S$, we have that $(a+b)\star c=(a\star c)+(b\star c)$ and that $a\star(b+c)=(a\star b)+(a\star c)$;

• If $a,b,c\in S$ are such that $a\star b=0$, then $a=0$ or $b=0$.

If there exists $1_{S}\in S$ such that $1_{S}\star a=a\star 1_{S}=a$ for all $a\in S$, then $(S,+,\star)$ is called a semifield. In 1965, Knuth [K-1965-fsapp] showed that the additive group of a presemifield is isomorphic to the additive group of a finite field $\mathbb{F}_{q}$. So we can always represent a presemifield as $(\mathbb{F}_{q},+,\star)$, where $+$ is the usual addition over the finite field $\mathbb{F}_{q}$. Moreover, any finite field is also a semifield with the usual multiplication.

Two presemifields $\mathbb{PS}_{1}=(\mathbb{F}_{q},+,\star)$ and $\mathbb{PS}_{2}=(\mathbb{F}_{q},+,*)$ are called isotopic if there are three linear permutations $L$, $M$, and $N$ of $\mathbb{F}_{q}$ such that $L(x\star y)=M(x)*N(y)$ for any $x,y\in\mathbb{F}_{q}$. If $M=N$, then $\mathbb{PS}_{1}$ and $\mathbb{PS}_{2}$ are called strongly isotopic. Every finite commutative presemifield $\mathbb{PS}=(\mathbb{F}_{q},+,\star)$ is isotopic to a finite commutative semifield $\mathbb{S}=(\mathbb{F}_{q},+,*)$ where we choose any $a\in\mathbb{F}_{q}\setminus\{0\}$ and we set

$$(x\star a)*(y\star a)=x\star y.$$

Then $1_{\mathbb{S}}=a\star a$. We observe that $\mathbb{PS}$ and $\mathbb{S}$ are strongly isotopic by using the transformation $(L,M,N)=(\mathrm{id},\mathrm{id}\star a,\mathrm{id}\star a)$ where $\mathrm{id}$ is the identity function over $\mathbb{F}_{q}$.

The left, middle and right nucleus of a finite semifield $\mathbb{S}=(\mathbb{F}_{q},+,\star)$ are denoted by, respectively,

$$N_{\ell}(\mathbb{S}) =\left\{\alpha\in\mathbb{F}_{q}\colon(\alpha\star x)\star y=\alpha\star(x\star y)\text{ for all }x,y\in\mathbb{F}_{q}\right\},$$

$$N_{m}(\mathbb{S}) =\left\{\alpha\in\mathbb{F}_{q}\colon(x\star\alpha)\star y=x\star(\alpha\star y)\text{ for all }x,y\in\mathbb{F}_{q}\right\},$$

$$N_{r}(\mathbb{S}) =\left\{\alpha\in\mathbb{F}_{q}\colon(x\star y)\star\alpha=x\star(y\star\alpha)\text{ for all }x,y\in\mathbb{F}_{q}\right\}.$$

We denote by $N(\mathbb{S})=N_{\ell}(\mathbb{S})\cap N_{m}(\mathbb{S})\cap N_{r}(\mathbb{S})$ the nucleus of $\mathbb{S}$. If $\mathbb{S}_{1}$ and $\mathbb{S}_{2}$ are isotopic, then their nuclei (left, middle, right, and the nucleus) have the same order. We have that if $\mathbb{S}$ is commutative, then $N(\mathbb{S})=N_{\ell}(\mathbb{S})=N_{r}(\mathbb{S})$ because $N_{\ell}(\mathbb{S})=N_{r}(\mathbb{S})$ and $N_{\ell}(\mathbb{S})\subseteq N_{m}(\mathbb{S})$.

When $q$ is odd, Coulter and Henderson [CH-2008-cpas] showed there is a 1-to-1 correspondence between finite commutative semifields and planar DO polynomials which can be realised in the following way. From any finite commutative presemifield $(\mathbb{F}_{q},+\star)$ we can obtain a planar DO polynomial $F\in\mathbb{F}_{q}\left[X\right]$ by $F(x)=\frac{1}{2}(x\star x)$. Conversely, any planar DO polynomial $F\in\mathbb{F}_{q}\left[X\right]$ defines a commutative presemifield $(\mathbb{F}_{q},+,\star)$ via field addition and multiplication given by $x\star_{F}y=F(x+y)-F(x)-F(y)$.

2.3 Equivalence relations of functions

Let $F$ and $G$ be two functions over $\mathbb{F}_{p^{n}}$. The functions $F$ and $G$ are said to be:

• affine equivalent (resp. linear equivalent) if there are two affine (resp. linear) permutations $A_{1}$ and $A_{2}$ over $\mathbb{F}_{p^{n}}$ such that $G=A_{1}\circ F\circ A_{2}$;

• Extended Affine equivalent, or EA-equivalent, if there is an affine function $A$ over $\mathbb{F}_{p^{n}}$ such that $G$ is affine equivalent to $F+A$.

• Carlet, Charpin and Zinoviev equivalent, or CCZ-equivalent, if there is an affine permutation $\mathcal{A}$ over $\mathbb{F}_{p^{n}}^{2}$ map** $\Gamma_{F}$ to $\Gamma_{G}$. Where by $\Gamma_{F}$ we denote the graph of the function $F$, $\Gamma_{F}=\left\{(x,F(x))\colon x\in\mathbb{F}_{p^{n}}\right\}$.

We specify that CCZ-equivalence is strictly more general than EA-equivalence (any pair of functions that are EA-equivalent are also CCZ-equivalent, but not necessarily vice-versa); EA-equivalence is strictly more general than affine equivalence; and affine equivalence is strictly more general than linear equivalence.

Dempwolff [U-2018-ceopf] proved that two monomial functions $F(x)=x^{d}$ and $G(x)=x^{e}$ are CCZ-equivalent if and only if they are cyclotomic equivalent, that is there exists a positive integer $i$ such that $d=p^{i}e\pmod{p^{n}-1}$ or $d=p^{i}e^{-1}\pmod{p^{n}-1}$ if $e$ is invertible modulo $p^{n}-1$. Budaghyan and Helleseth [BH-2008-npnmo] proved that the equivalence relations linear, affine, EA, and CCZ all coincide for planar DO polynomials over $\mathbb{F}_{p^{n}}$. Two planar DO polynomials $F$ and $G$ are strongly isotopic if and only if they are CCZ-equivalent, and any isotopic class contains at most two CCZ-equivalence classes. The following theorem gives more insights on this topic.

Theorem 1 (Coulter Henderson [CH-2008-cpas]).

Let $\mathbb{PS}$ and $\mathbb{PS}^{\prime}$ be two commutative presemifields where $p$ is a prime and $n$ is a positive integer. Let $\mathbb{S}=(\mathbb{F}_{p^{n}},+,\star)$ and $\mathbb{S}^{\prime}=(\mathbb{F}_{p^{n}},+,*)$ be two commutative semifields strongly isotopic respectively to $\mathbb{PS}$ and $\mathbb{PS}^{\prime}$. Suppose that $\mathbb{PS}$ and $\mathbb{PS}^{\prime}$ are isotopic. Let $|N_{m}(\mathbb{S})|=p^{m}$ and $|N(\mathbb{S})|=p^{k}$. Then one of the following holds

• If $m/k$ is odd, then $\mathbb{PS}$ and $\mathbb{PS}^{\prime}$ are strongly isotopic.

• If $m/k$ is even, then either $\mathbb{PS}$ and $\mathbb{PS}^{\prime}$ are strongly isotopic or all isotopisms $(L,M,N)$ from $\mathbb{S}$ to $\mathbb{S}^{\prime}$ are such that $M=\alpha\star N$ where $\alpha\in N_{m}(\mathbb{S})$ is a non-square (w.r.t. to the multiplication $\star$) or, equivalently, $\alpha\in N_{m}(\mathbb{S})\setminus N(\mathbb{S})$.

Moreover, if $n$ is odd, the notion of strong isotopism coincides with the one of isotopism for commutative presemifields.

One important implication of this theorem is that if the dimension of the middle nucleus over the left nucleus of a commutative semifield is even, then the isotopy class of the semifield can split into two strong isotopy classes. Consequently, we have checked all situations where this might be possible as part of our accounting of the known classes of planar polynomials that we give below. While this may sound computationally difficult, in practice it is not, as the specific format of the splitting scenario outlined in the theorem means we need only check the isotopisms $(L,M,N)=(x,\alpha\star x,x)$ with $\alpha\in N_{m}(\mathbb{S})\setminus N(\mathbb{S})$ a non-square. This observation is not new, having already been used by Zhou [Z-2012-anoti] for example. Moreover, one can improve the computational investigation by taking only one representative $\alpha$ in each coset $\beta\star N(\mathbb{S})$ where $\beta\in N_{m}(\mathbb{S})\setminus N(\mathbb{S})$ non-square. Indeed, for any $\gamma\in N(\mathbb{S})\setminus\{0\}$ the two isotopism $(L,M,N)=(x,\beta\star x,x)$ and $(L,M,N)=(x,(\gamma\star\beta)\star x,x)$ give strongly isotopic semifields because

$$\left((\gamma\star\beta)\star x\right)\star y=\left(\gamma\star(\beta\star x)\right)\star y=\gamma\star\left((\beta\star x)\star y\right).$$

A property that is preserved by an equivalence relation is called an invariant. In particular, if a property is preserved by linear equivalence (respectively, affine, EA, CCZ, isotopic), then it is a linear invariant (respectively, affine, EA, CCZ, isotopic). Invariants can be useful to speed up equivalence tests because functions with a different value for an invariant must be inequivalent. Not many invariants are known for planar DO polynomials $F$ over $\mathbb{F}_{q}$, but some have been useful in past research. Firstly, the orders of the various nuclei of any commutative semifield $\mathbb{S}$ constructed from $F$ are isotopic invariants. Now define a linear code $\mathcal{C}_{F}$ with generator matrix

$$M_{F}=\left[\begin{matrix}1\\ x\\ F(x)\end{matrix}\right]_{x\in\mathbb{F}_{q}}.$$

The monomial automorphism group, or automorphism group for short, is the group of monomial matrices that map the code $\mathcal{C}_{F}$ to itself. A monomial matrix is an invertible matrix where each column has exactly one non-zero entry. Pott and Zhou [PZ-2010-scopf] showed that the order of the monomial automorphism group of $\mathcal{C}_{F}$ is a CCZ-invariant. Finally, we consider the set of linear self-equivalences of $F$, $EQ(F,F)$, that is the set of pairs of linear permutations $(L_{1},L_{2})$ such that $L_{1}\circ F\circ L_{2}=F$. The right orbit of $x\in\mathbb{F}_{q}$ with respect to the set of linear self-equivalences is defined as the set

$$RO_{F,x}=\left\{y\in\mathbb{F}_{q}\colon(L_{1},L_{2})\in EQ(F,F),L_{2}(x)=y\right\}.$$

The multiset of the cardinalities of each orbit is a linear invariant [IK-2022-darle].

3 On the known families of Planar Functions

In this section we address the nomenclature of planar functions. Our aim is to clarify any confusion there may be in the literature regarding precedent and discovery while at the same time explaining our naming convention and hopefully standardizing the names of known families. We do this in part to address the recent proliferation of claims of new planar functions, especially over the last decade, which have since been verified as being examples of known families. The result of Coulter and Henderson [CH-2008-cpas] linking planar DO polynomials with commutative semifields does complicate the history a little, but we are somewhat fortunate in the sense that almost no planar DO polynomials known at the time of the discovery corresponded with known families of commutative semifields. That said, given the equivalence, any families are named after whatever came first, whether that be a semifield or a planar function.

Finally, we note that in her Ph.D. thesis [K-2009-csooo], Kosick determines planar DO polynomial representatives for all of the commutative semifield families discovered before 2009. Specifically, planar DO polynomial representatives are given for the Dickson, Cohen-Ganley, Ganley families, as well as the sporadic example of Pentilla and Williams. See also the unpublished paper [CHHKXZ-2007-ppacs] which, among other things, gives a family of planar DO binomials that represent a subset of the Dickson semifields.

1. FF – Finite fields

   • $\mathbb{S}=(\mathbb{F}_{p}^{n},+,\star)$ given by

     $$x\star y=xy\text{ for all }x,y\in\mathbb{F}_{p}^{n}$$

   • Planar DO representative:

     $$F(x)=x^{2}.$$

   • $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(p^{n},p^{n})$.

   The monomial $x^{2}$ is planar over any field, finite or infinite, which is not of characteristic 2. It is easily seen to be equivalent to the field on which it is defined. We give 1893 as the year as it was then that Moore first established the uniqueness of finite fields of arbitrary order, thereby completing their classification, see [M-1893-adiso, M-1896-adiso].

2. D – The commutative semifields of Dickson

   • $\mathbb{S}=(\mathbb{F}_{p^{m}}^{2},+,\star)$ given by

     $$(a,b)\star(c,d)=(ac+\alpha(bd)^{p^{i}},ad+bc)\text{ for all }a,b,c,d\in\mathbb{F}_{p^{m}}.$$

     Here, $\alpha\in\mathbb{F}_{p^{m}}$ is a nonsquare and $0<i\leq\lfloor\frac{m}{2}\rfloor$.

   • Different choices of $i$ lead to non-isotopic classes, while the two integers $i$ and $i^{\prime}=m-i$ yield isotopic semifields, see Burmester [B-1962-otcna].

   • For fixed i, all non-square $\alpha$ produce isotopic semifields, see [D-1906-oclai].

   • There are multiple choices for the planar DO representatives for this class. Perhaps the best general forms known prior to this article are those given by Kosick [K-2009-csooo] (see Lemma 2 below). Also, a notable one is in binomial form [CHHKXZ-2007-ppacs] but has many restrictions (see Lemma 1 below). In this article, we shall establish a class of quadrinomials that can be used as the representatives, see Theorem 2.

   • $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(p^{\gcd(i,m)},p^{m})$.

   Dickson produced his family of commutative semifields in 1906 in [D-1906-oclai]. (Note that a typo in Dembowski’s book [bD-1968-fg], page 241, inadvertently cites his nearfield paper [D-1905-ofa] instead.) He was motivated by his work [D-1905-ofa] from the previous year in which he was the first to construct nearfields (a division ring where only one of the two distributive laws holds), describing an infinite family, along with a further 7 sporadic examples. That he had described all possible nearfields was only confirmed in 1935 by Zassenhaus [Z-1935-ubef]. It is, perhaps, worth mentioning that in 1907 Veblen and Wedderburn [VW-1907-ndanp] used Dickson’s nearfields to construct the first known examples of non-Desarguesian projective planes, and that their construction could also have used Dickson’s commutative semifields.

3. A – Albert’s twisted fields

   • $\mathbb{S}=(\mathbb{F}_{p}^{n},+,\star)$ given by

     $$x\star y=x^{p^{i}}y+xy^{p^{i}}\text{ for all }x,y\in\mathbb{F}_{p}^{n}$$

   • Planar DO representative:

     $$F(x)=x^{p^{i}+1}$$

     where $1\leq i\leq\lfloor\frac{n}{2}\rfloor$ with $\frac{n}{\gcd(n,i)}$ odd.

   • Different choices of $i$ lead to non-isotopic classes, see [A-1952-onda].

   • $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(p^{\gcd(n,i)},p^{\gcd(n,i)})$.

   Albert’s twisted fields were first published in [A-1952-onda], and subsequently generalised and extensively studied by him in the papers [A-1960-fdaaf, A-1961-gtf, A-1961-ifgtf, A-1963-otcga]. The form of the equivalent planar monomials, the DO monomials, were first described by Dembowski and Ostrom [DO-1968-poowc], but with an erroneous condition, subsequently corrected by Coulter and Matthews in [CM-1997-pfapo]. Note that both Albert’s original construction and the DO monomial examples include the finite field case as an example.

4. CG – The commutative semifields of Cohen and Ganley

   • $\mathbb{S}=(\mathbb{F}_{3^{m}}^{2},+,\star)$ given by

     $$(a,b)\star(c,d)=(ac+\alpha(bd)+\alpha^{3}(bd)^{9},ad+bc+\alpha(bd)^{3})\text{ for all }a,b,c,d\in\mathbb{F}_{3^{m}}.$$

     Here, $m\geq 3$, and $\alpha\in\mathbb{F}_{3^{m}}$ is a nonsquare.

   • All non-square $\alpha$ produce isotopic semifields, see [CG-1982-cstdo].

   • Planar DO representative over $\mathbb{F}_{3^{2m}}$:

     $$F(x)=L(t^{2}(x))+\frac{1}{2}x^{2}.$$

     Here $m\geq 3$ odd, $t=x^{3^{m}}-x$, $\beta\in\mathbb{F}_{3^{2m}}\setminus\mathbb{F}_{3^{m}}$, $\alpha=t(\beta)$, and $L(x)=-x^{9}-\alpha x^{3}+(1-\alpha^{4})x$.

   • Each member of this family is isotopic inequivalent to the Dickson (D) family.

   • There are multiple choices for the planar DO representatives for this class. The planar DO representatives we use here were given by Kosick [K-2009-csooo]. We remark that there is a typo in the DO representatives given by Kosick, but the proofs are correct for the forms we give.

   • $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(3,3^{m})$.

   They were first constructed by Cohen and Ganley [CG-1982-cstdo], in their study of a general construction method for commutative semifields two dimensional over their middle nuclei.

5. G – The commutative semifields of Ganley

   • $\mathbb{S}=(\mathbb{F}_{3^{m}}^{2},+,\star)$ given by

     $$(a,b)\star(c,d)=(ac-b^{9}d-bd^{9},ad+bc+b^{3}d^{3})\text{ for all }a,b,c,d\in\mathbb{F}_{3^{m}}.$$

     Here, $m\geq 3$ odd.

   • Planar DO representative over $\mathbb{F}_{3^{2m}}$:

     $$F(x)=L(t^{2}(x))+D(t(x))+\frac{1}{2}x^{2}.$$

     Here $m\geq 3$ odd, $t=x^{3^{m}}-x$, $\beta\in\mathbb{F}_{3^{2m}}\setminus\mathbb{F}_{3^{m}}$, $\alpha=t(\beta)$, $L(x)=-\alpha^{-5}x^{3}+x$, and $D(x)=-\alpha^{-10}x^{10}$.

   • There are multiple choices for the planar DO representatives for this class. The planar DO representatives we use here were given by Kosick [K-2009-csooo]. As for the CG family, we remark that there is a typo in the DO representatives given by Kosick, but the proofs are correct for the forms we give.

   • $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(3,3)$.

   Building on the work in [CG-1982-cstdo], Ganley used the concept of weak nuclei to produce the commutative semifields in [G-1981-cwns].

6. CM – The planar monomials of Coulter and Matthews

   • Planar DO representative over $\mathbb{F}_{3^{n}}$:

     $$F(x)=x^{(3^{i}+1)/2}.$$

     Here, $i\geq 3$ is odd and $\gcd(n,i)=1$.

   These are the only non-DO examples of planar functions known, and as they are not DO polynomials, they do not produce commutative semifields. The Coulter-Matthews monomials were announced on an online forum in 1994 and presented in full at Fq3, the 3rd International Conference on Finite Fields and Applications, in Glasgow in July, 1995. Delays in publication meant that Coulter and Matthews’ results only appeared in print in 1997 in [CM-1997-pfapo]. In the same year, Helleseth and Sandberg [HS-1997-spmwl] published a paper which included the Coulter-Matthews examples. However, precedent was confirmed publicly by Tor Helleseth at BFA 2018, the third International Workshop on Boolean Functions and Their Application, in Loen in June 2018. We are most grateful to Tor for the clarification. These are the only known non-quadratic planar. It is known that if a quadratic function is CCZ-equivalent to a monomial, then such monomial must be quadratic. To check CCZ-equivalence among monomials, it is enough to check that their exponents belong to the same cyclotomic set. Since EA coincides with CCZ for planar functions, we can classify the instances of CM separately from the quadratic planar functions.

7. TST – The Ten-Six-Two family of planar DO polynomials

   • Planar DO representative over $\mathbb{F}_{3^{n}}$:

     $$F(x)=x^{10}\pm x^{6}-x^{2}.$$

     Here, $n\geq 5$ odd.

   • The two functions are inequivalent up to strong isotopism, see [CH-2008-cpas].

   • $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(3,3)$.

   The Ten-Six-Two family of functions are so named for their form: $g_{5}(x^{2},a)=x^{10}+ax^{6}-a^{2}x^{2}$, where $g_{k}(x,a)$ denotes the $k$th Dickson polynomial of the first kind. (For information regarding the Dickson polynomials of the first and second kind, see the monograph [bLMT-1993-dp] of Lidl, Mullen and Turnwald.) Coulter and Matthews established the planarity of $g_{5}(x^{2},1)$ in [CM-1997-pfapo]. Ding and Yuan [DY-2006-afosh] extended the proof to all $a\neq 0$ in 2007 as part of their disproof of a 70 year old conjecture on skew Hadamard difference sets. Coulter and Henderson [CH-2008-cpas] settled the question of equivalence by showing there are two equivalence families, one for all square $a$ and one for all non-square $a$. Thus, one can use $a=\pm 1$ to describe the two families.

8. PW – A sporadic commutative semifield of order $3^{10}$

   • $\mathbb{S}=(\mathbb{F}_{3^{5}}^{2},+,\star)$ given by

     $$(a,b)\star(c,d)=((ac+bd)^{9},ad+bc+(bd)^{27})\text{ for all }a,b,c,d\in\mathbb{F}_{3^{5}}.$$

   • Planar DO representatives over

   • It is not isotopic equivalent to a Dickson semifield or to a Cohen-Ganley semifield, see [PW-2000-oops].

   • $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(3,3^{5})$.

   In their study of ovoids in the orthogonal space $O(5,3^{5})$, Penttila and Williams [PW-2000-oops] discovered this commutative semifield as part of a computational search. At the time of writing, this example is not part of any known infinite family.

9. ACW – A sporadic planar DO binomial over $\mathbb{F}_{3^{5}}$

   • Planar DO representative over $\mathbb{F}_{3^{5}}$:

     $$F(x)=x^{90}+x^{2}.$$

   • $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(3,3).$

   At and Cohen [AC-2008-antfa] found a new planar binomial over $\mathbb{F}_{3^{5}}$ in the course of testing a proposed method for establishing planarity. In July, 2009, in Dublin, at Fq9, the 9th International Conference on Finite Fields and Applications, Coulter and Kosick [CK-2010-csooa] attributed the very same planar example to Guibiao Weng, who had informed them of the example in personal correspondence dating from March 2008 at the latest. It appears both At and Cohen, and Weng discovered this binomial independently at around the same time.

10. BH – The planar DO polynomials of Budaghyan and Helleseth

    • Planar DO representative over $\mathbb{F}_{p^{2m}}$:

      $$F(x)=x^{p^{m}+1}+\omega\mathrm{Tr}_{m}^{2m}(\beta x^{p^{s}+1}).$$

      Here, $\omega\in\mathbb{F}_{p^{2m}}\setminus\mathbb{F}_{p^{m}}$, $\beta\in\mathbb{F}_{p^{2m}}$ is a non-square and $0<s<m$ with $\nu_{2}(s)\neq\nu_{2}(m)$ where, for any positive integer $k$, $\nu_{2}(k)$ is the non-negative integer such that $\frac{k}{2^{\nu_{2}(k)}}$ is odd.

    • For fixed $s$, any choice of $\omega$ and $\beta$ produces functions in the same strong isotopy class, see Bierbrauer [B-2011-csfpm] or Feng and Li [FL-2018-otico].

    • For any $m>1$ odd, each member splits into two isotopic classes (see Remark 2 below).

    • $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(p^{\gcd(m,s)},p^{2\gcd(m,s)})$ [MP-2012-otnoa].

    Budaghyan and Helleseth [BH-2008-npnmo] established a family of planar DO polynomials with many terms. These produced the first new infinite family of commutative semifields without a restriction on the characteristic to be discovered in over 50 years. For convenience, here we present the form defined by Bierbrauer in [B-2011-csfpm, B-2010-nspaa] (see [ZP-2013-anfos, Theorem 7]) because it is shorter. Many families of planar DO polynomials discovered since have been shown to be equivalent to this family such as the Lunardon-Marino-Polverino-Trombetti-Bierbrauer (LMPTB) family [B-2011-csfpm], the Zha-Wang (ZW) family [ZW-2009-nfopn], and a family by Bierbrauer[BBFMP-2018-afosi]. Even some parts of the ZP family, see below, turns out to be equivalent to this family. Over $\mathbb{F}_{3^{4}}$, all the instances of the BH family do not split. In this paper, we show that all the instances of the BH family over $\mathbb{F}_{3^{8}}$ that have nuclei $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(3,3^{2})$ split and the ones that have nuclei $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(3^{2},3^{4})$ do not split.

11. ZKW – The planar DO binomials of Zha, Kyureghyan and Wang

    • Planar DO representative over $\mathbb{F}_{p^{n}}$:

      $$F(x)=x^{p^{s}+1}-\alpha^{p^{k}-1}x^{p^{k}+p^{2k+s}}.$$

      Here $n=3k$, $\gcd(3,k)=1$, $s$ positive integer such that $k\equiv s\pmod{3}$, $\frac{n}{\gcd(s,n)}$ is odd and $\alpha\in\mathbb{F}_{p^{3k}}$ is primitive.

    • $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(p^{\gcd(s,k)},p^{\gcd(s,k)})$ [MP-2012-otnoa].

    A family of planar DO binomials was given in 2009 by Zha, Kyureghyan and Wang [ZKW-2009-pnlba]. This family was obtained as a generalization of a family of APN binomials presented by Budaghyan, Carlet, and Leander [BCL-2008-tcoqa]. They also showed that the family contains planar functions that are not monomials.

12. B – The planar DO binomials of Bierbrauer

    • Planar DO representative over $\mathbb{F}_{p^{n}}$:

      $$F(x)=x^{p^{s}+1}-\alpha^{p^{k}-1}x^{p^{3k}+p^{k+s}}.$$

      Here $n=4k$, such that $\frac{2k}{\gcd(2k,s)}$ is odd, $p^{s}=p^{k}=1\pmod{4}$, and $\alpha\in\mathbb{F}_{p^{3k}}$ is primitive.

    • $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(p^{\gcd(s,k)},p^{2\gcd(s,k)})$ [MP-2012-otnoa].

    This family of planar DO binomials was established by Bierbrauer [B-2010-nspaa]. These binomials have a remarkably similar structure to those discovered in [ZKW-2009-pnlba], but have distinct dimensions. They were also obtained as a generalization of the same family of APN binomials [BCL-2008-tcoqa].

13. CK – Two sporadic planar DO polynomials over fields of order $3^{5}$ and $5^{5}$

    • Planar DO representatives over $\mathbb{F}_{p^{5}}$ with $p=3,5$:

      $$F(x)=L(t^{2}(x))+D(t(x))+\frac{1}{2}x^{2}.$$

    • The first with $p=3$, $t(x)=x^{3}-x$, $L(x)=-x^{3}$ and $D(x)=-x^{36}+x^{30}+x^{28}+x^{4}$ (or $L(x)=-x$ and $D(x)=-x^{36}+x^{28}+x^{12}+x^{4}$).

    • $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(3,3)$.

    • The second with $p=5$, $t(x)=x^{5}-x$, $L(x)=x^{5^{3}}+x^{5^{2}}+2x^{5}+3x$ and $D(x)=0$ (or $L(x)=2x^{5^{2}}+x^{5}$ and $D(x)=2x^{5^{3}+5}+2x^{5^{2}+1}$).

    • $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(5,5)$.

    Using the form described in [CHK-2007-ppfcs], Coulter and Kosick [CK-2010-csooa] conducted an exhaustive search for families of planar DO polynomials that had representatives with coefficients in the prime subfield. This search was carried out for fields of order $3^{5}$ and $5^{5}$, and two new examples were discovered, one for each order. Both remain outside of any known infinite family.

14. ZP – The commutative semifields of Zhou and Pott

    • $\mathbb{S}=(\mathbb{F}_{p^{m}}^{2},+,\star)$ given by

      $$(a,b)\star(c,d)=(a\circ_{k}c+\alpha(b\circ_{k}d)^{p^{i}},ad+bc)\text{ for all }a,b,c,d\in\mathbb{F}_{p^{m}}^{2}.$$

      Here $0\leq k,i\leq\lfloor\frac{m}{2}\rfloor$ with $(i,k)\neq(0,0)$, $\frac{m}{\gcd(m,k)}$ odd, $x\circ_{k}y=x^{p^{k}}y+xy^{p^{k}}$ and $\alpha\in\mathbb{F}_{p^{m}}$ is a nonsquare.

    • By fixing $i$ and $k$ as above, any choice of $\alpha\in\mathbb{F}_{p^{m}}$ non-square leads to a semifield that fall into the same strong isotopic class [ZP-2013-anfos].

    • The D family coincides with ZP with parameters $k=0$ and $0<i\leq\lfloor\frac{m}{2}\rfloor$. The BH family over the finite field $\mathbb{F}_{p^{m}}$ such that $m$ is odd and $-1$ is a square in $\mathbb{F}_{p^{m}}$ coincides with ZP with parameters $i=0$ and $0<k\leq\lfloor\frac{m}{2}\rfloor$ [ZP-2013-anfos, Theorem 5].

    • Let $0\leq i_{2},i_{2}\leq\lfloor\frac{m}{2}\rfloor$ and $0<k_{1},k_{2}\leq\lfloor\frac{m}{2}\rfloor$ with $(i_{1},k_{1})\neq(i_{2},k_{2})$ then the two semifileds defined by $(i_{1},k_{1})$ and $(i_{2},k_{2})$ are not isotopic. Moreover if $i=0$ and $0<k\leq\lfloor\frac{m}{2}\rfloor$, then the isotopic class of each semifiled defined by $(0,k)$ contains exactly two strong isotopic classes [ZP-2013-anfos, Theorem 6].

    • $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=\begin{cases}(p^{\gcd(m,k)},p^{2\gcd(m,k)})&i=0\\ (p^{\gcd(m,k,i)},p^{\gcd(m,k)})&i>0.\end{cases}$

    By cleverly replacing the field multiplication with twisted field multiplication in the general form of commutative semifields studied by Cohen and Ganley [CG-1982-cstdo], itself a form based on Dickson’s original construction [D-1906-oclai], Zhou and Pott [ZP-2013-anfos] produced a new general family of commutative semifields. This construction method generated many new inequivalent families.

15. GK – The commutative semifields of Göloğlu and Kölsch

    • Planar DO representative over $\mathbb{F}_{p^{m}}^{2}$:

      $$F(x,y)=(x^{p^{k}+1}+\alpha y^{p^{k}+1},x^{p^{k+m/2}}y+\beta\alpha^{-1}xy^{p^{k+m/2}}).$$

      Here $m$ is even and not a power of $2$, $\alpha\in\mathbb{F}_{p^{m}}$ is a nonsquare, $\beta\in\mathbb{F}_{p^{m}}$ is not a power of $p^{m/2}+1$ and $\frac{m}{\gcd(m,k)}$ is odd.

    • $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(p^{\gcd(k,m)/2},p^{\gcd(k,m)})$.

    Göloğlu and Kölsch dramatically changed the number of known inequivalent commutative semifields in a landmark paper [GK-2023-aebot]. They produced a new construction method which allowed them to combine some of the previously known commutative semifields into new semifields. Through their construction method they were able to show that the number of non-isotopic commutative semifields of order $p^{n}$ grows exponentially with $n$. This confirmed a conjecture of Kantor, who had proved a corresponding result for commutative semifields of even order in 2003 [K-2003-csass], but the result did not transfer to odd characteristic.

Remark 1.

In 2007, Coulter, Henderson and Kosick [CHK-2007-ppfcs] developed a general form for planar DO polynomials that can be used to describe all planar DO families. Using this form they published a purported example over $\mathbb{F}_{3^{8}}$ with middle nucleus of order 9 and nucleus of order 3. Unfortunately, the paper has numerous typographical errors and the original example, often called the CHK semifield, was wrong (it is not even planar). A replacement example was given on Coulter’s personal website, but it was sourced from a file of Dickson isotopes by mistake. At the time [CHK-2007-ppfcs] was published, Coulter, Henderson and Kosick knew of 4 potentially inequivalent commutative semifields of order $3^{8}$ with $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(3,9)$, but had been unable to prove the inequivalence of these among themselves. We have now confirmed that the four potential classes are all now covered by the BH class (indeed, Classes 8.5 and 8.6 in Table A.1). For these reasons, even though the 4 examples known to the authors of [CHK-2007-ppfcs] do pre-date the BH class, we do not list the CHK example as part of the nomenclature.

Remark 2.

Let $m>1$. In [ZP-2013-anfos], it is shown that ZP with $i=0$ and $k\neq 0$ coincides with BH if $m$ is odd and $-1$ is a square in $\mathbb{F}_{p^{m}}$ and that each of those instances split. Since $m$ is odd, $-1$ is a square in $\mathbb{F}_{p^{m}}$ if and only if $p=1\pmod{4}$. Marino and Polverino [MP-2012-oiasi] prove that the BH family splits if $m$ is not a power of $2$, $\frac{m}{\gcd(m,s)}$ is odd and $p^{\gcd(m,s)}=3\pmod{4}$. Observe that if m is odd, then $p^{\gcd(m,s)}=p\pmod{4}$. If $m$ is even and $\frac{m}{\gcd(m,s)}$ is odd, then $\gcd(m,s)$ is even and $p^{\gcd(m,s)}=1\pmod{4}$. Combining the two results, we get that the BH family splits if $m$ is odd.

3.1 Some remarks on the classification of planar DO polynomials

We record some known facts on the classification of planar functions, planar DO polynomials and commutative semifields.

• Planar functions over prime fields were classified in 1989 and 1990. All are quadratic, and all are equivalent to $x^{2}$, which produces the finite field. This was established independently by three sets of authors: Gluck [G-1990-apapp], Hiramine [H-1989-acoap], and Rónyai and Szönyi [RS-1989-pfaff].

• Planar monomials have been classified over fields of order $p^{n}$ with $n\in\{1,2,3,4\}$. Johnson [J-1987-ppoot] proved the prime field case in 1987. Order $p^{2}$ was completed by Coulter [C-2006-tcopm], and order $p^{4}$ by Coulter and Lazebnik [CL-2012-otcop]. The order $p^{3}$ case was completed in 2022 by Bergman, Coulter and Villa [BCV-2022-cpmof].

• Knuth [K-1965-fsapp] showed that commutative semifields of order $p^{2}$ are necessarily isotopic to finite fields in 1965. In 1977, Menichetti [M-1977-oakcc] proved that any commutative semifield of dimension 3 over its nucleus is necessarily isotopic to an Albert twisted field. Together these two results complete the classification of commutative semifields of orders $p^{2}$ and $p^{3}$.

• Menichetti later proved in [M-1996-daoaf] that if $n$ is prime and $q$ is sufficiently large, any commutative semifield of order $q^{n}$ with nucleus of order $q$ is equivalent to an Albert’s twisted field.

• Let $q$ be an odd prime power and $l$ a positive integer. Blokhuis, Lavrauw and Ball [BLB-2003-otcos] proved that if $q\geq 4l^{2}-8l+2$, then any commutative semifield of order $q^{2l}$ with $|N(\mathbb{S})|\geq q$, $|N_{m}(\mathbb{S})|\geq q^{l}$ is either equivalent to the finite field or a Dickson semifield. In particular, this shows that any commutative semifield of order $p^{4}$ with $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(p,p^{2})$ is necessarily a Dickson semifield or a finite field. Note that the only remaining case to consider for order $p^{4}$ is with $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(p,p)$. At the time of writing, we know of no examples.

• Commutative semifields of orders $3^{n}$ have been classified for $n\leq 5$. The results of Knuth and Menichetti mentioned above deal with $n\leq 3$. For order 81, Dickson [35] showed in 1906 “by a tedious computation” that the only commutative semifields were those given by his construction. To put this in context, it wasn’t until 2008 that Dempwolff [D-2008-spoo] managed to enumerate all semifields of order 81. Weng and Zeng [WZ-2012-fropd] computed all commutative semifields of order 243 in 2012.

• For $m\in\{2,3,4,5\}$, every commutative semifield $\mathbb{S}$ over $\mathbb{F}_{3^{2m}}$ with nuclei $(|N(\mathbb{S})|,|N_{m}(\mathbb{S})|)=(3,3^{m})$ is either in Dickson, in Cohen-Ganley, or in Pentilla-Williams (if $m=5$) [BLB-2003-otcos, MPT-2007-oflso, LR-2019-codrt].

• For $n$ up to $7$, all planar DO polynomials in $\mathbb{F}_{3^{n}}$ with coefficients in the prime field $\mathbb{F}_{3}$ were classified by Davidova and Kaleyski [DK-2022-coadp].

4 Quadrinomials representing the Dickson family

There are a number of planar DO representatives of the Dickson family known. Some of these are particularly simple. For example, [CHHKXZ-2007-ppacs, Theorem 4.2] gives binomial representatives for the specific Dickson semifields of dimension 2 over the middle nucleus and dimension 4 over the nucleus. Kosick [K-2009-csooo] gives also planar DO representatives for any Dickson semifield. Here we present both for completeness.

Lemma 1 ([CHHKXZ-2007-ppacs, Theorem 4.2]).

Let $p$ be any odd prime and let $m\geq 2$ be an even positive integer such that $p^{m/2}=1\pmod{4}$. Let $q=p^{m}$, let $\alpha$ be a primitive element of $\mathbb{F}_{q^{2}}$ and $e$ a positive integer. Then $F(x)=x^{q+1}+\alpha^{e(p^{m/2}-1)}x^{2p^{m/2}}$ is a planar DO polynomial over $\mathbb{F}_{q^{2}}$ representing the Dickson semifield as described in class D.

Lemma 2 (Kosick [K-2009-csooo, Theorem 3.1.2]).

Let $p$ be any odd prime, $q=p^{m}$ for some integer $m\geq 2$ and let $0<i<m$. Let $t(x)=x^{q}-x$ and $L(x)=8^{-1}(x^{p^{i}}-x)$ be functions over $\mathbb{F}_{q^{2}}$. Then $F(x)=L(t^{2}(x))+2^{-1}x^{2}$ is a planar DO polynomial over $\mathbb{F}_{q^{2}}$ representing the Dickson semifield as described in class D.

We note that the proof given by Kosick in [K-2009-csooo] actually proves a slightly more general statement than the one we gave above.

The representatives given in Lemma 2 have 6 terms. We now show that there are always quadrinomial representatives of Dickson semifields.

Theorem 2.

Let $p$ be any odd prime, $q=p^{m}$ for some integer $m\geq 2$ and let $0<i<m$. If $\gcd(i,m)=\gcd(i,2m)$, then the polynomial

$$x^{2}+x^{q+1}-x^{(q+1)p^{i}}+x^{2qp^{i}}$$ (2)

is a planar DO polynomial over $\mathbb{F}_{q^{2}}$ representing the Dickson semifield as described in class D above.

Proof.

For simplicity of exposition, set $r=p^{i}$. We shall first prove that the linearised polynomial $M(x)=3x-x^{r}+x^{q}+x^{qr}$ is a permutation polynomial over $\mathbb{F}_{q^{2}}$ and then we show that $M(F(x))$, where $F(x)$ is as given in Lemma 2, is the quadrinomial (2) of our statement. That it is planar follows at once from the fact $M$ is a linear permutation.

Let us show that $M$ is a permutation. To do so, we need only show the only root of $M$ in $\mathbb{F}_{q^{2}}$ is $z=0$. To that end, suppose $z\in\mathbb{F}_{q^{2}}$ satisfies $M(z)=0$. We have

$$0 =3z+z^{q}-z^{r}+z^{qr}$$

$$=(z^{q}+z)+2z+(z^{q}-z)^{r},$$

We conclude $2z+(z^{q}-z)^{r}\in\mathbb{F}_{q}$. We therefore have

$$2z^{q}+(z-z^{q})^{r}=2z+(z^{q}-z)^{r},$$

and rearranging and simplifying yields

$$0=(z^{q}-z)^{r}-(z^{q}-z).$$

Now, all roots of $x^{r}-x$ in $\mathbb{F}_{q^{2}}$ lie in the field $\mathbb{F}_{r}\cap\mathbb{F}_{q^{2}}$, which is equal to the field $\mathbb{F}_{r}\cap\mathbb{F}_{q}$ by hypothesis since $\mathbb{F}_{r}\cap\mathbb{F}_{q^{2}}=\mathbb{F}_{p^{\gcd(i,2m)}}=\mathbb{F}_{p^{\gcd(i,m)}}=\mathbb{F}_{r}\cap\mathbb{F}_{q}$. In particular, all roots of $x^{r}-x$ lie in $\mathbb{F}_{q}$, and so $z^{q}-z\in\mathbb{F}_{q}$. This implies $z^{q}-z=z-z^{q}$ from which we conclude $z\in\mathbb{F}_{q}$. Thus,

$$0=M(z)=3z+z-z^{r}+z^{r}=4z,$$

proving $z=0$ is our only solution. We have proved $M$ is a permutation.

Let us show that $M(F(x))$, where $F(x)$ is as given in Lemma 2, is the quadrinomial (2) of our statement. Observe that

$$8\cdot M(L(x)) =3x^{r}-3x+x^{qr}-x^{q}-x^{r^{2}}+x^{r}+x^{qr^{2}}-x^{qr}$$

$$=x^{qr^{2}}-x^{q}-x^{r^{2}}+4x^{r}-3x.$$

and that $8\cdot M(L(t^{2}(x)))$ is equal to

$$(x^{2q}+x^{2}-2x^{q+1})^{qr^{2}}-(x^{2q}+x^{2}-2x^{q+1})^{q}-(x^{2q}+x^{2}-2x^{q+1})^{r^{2}}$$

$$+4(x^{2q}+x^{2}-2x^{q+1})^{r}-3(x^{2q}+x^{2}-2x^{q+1})$$

$$= 4(x^{2q}+x^{2}-2x^{q+1})^{r}-4(x^{2q}+x^{2}-2x^{q+1}).$$

Therefore, we have that $2\cdot M(F(x))=2\cdot M(L(t^{2}(x)))+M(x^{2})$ is equal to

$$(x^{2q}+x^{2}-2x^{q+1})^{r}-(x^{2q}+x^{2}-2x^{q+1})+\left(3x^{2}-x^{2r}+x^{2q}+x^{2qr}\right)$$

$$= 2x^{2}+2x^{q+1}-2x^{r(q+1)}+2x^{2qr}.$$

This concludes the proof. ∎

The condition $\gcd(i,m)=\gcd(i,2m)$, while a restriction, does not stop us from obtaining all of the non-isotopic versions in Dickson’s class. This is because $i$ and $i^{\prime}=m-i$ produce the same Dickson isotopes. Let $i=2^{h}g$ and $m=2^{l}k$ with $g,k$ odd integers. Note that $i^{\prime}=2^{a}b$ with $a=\min(l,h)$ and for some odd integer $b$. If $\gcd(i,m)\neq\gcd(i,2m)$, then $h>l$ and so $\gcd(i^{\prime},m)=\gcd(i^{\prime},2m)$. Consequently, we can cover all of the non-isotopic semifields in the Dickson class by choosing $i$ or $i^{\prime}$ as appropriate.

5 Computational results

5.1 The expansion search for new planar functions

We describe the procedure we used to search for new instances of planar DO polynomials over $\mathbb{F}_{3^{6}}$, $\mathbb{F}_{3^{7}}$, $\mathbb{F}_{3^{8}}$, and $\mathbb{F}_{3^{9}}$. It consists of taking a quadratic power function $F(x)=x^{d}$ in univariate representation, adding quadratic terms of the form $cx^{d^{\prime}}$ progressively, and testing for planarity each time. This method is based on a similar method as the one introduced by Aleksandersen, Budaghyan, and Kaleyski [ABK-2022-sfafb] for the search of APN functions over binary fields. Then, we test if these functions are new up to CCZ-equivalence using the linear equivalence algorithm [IK-2022-darle, Algorithm 1]. We performed this computational investigation for $d\in\{2,10\}$ over $\mathbb{F}_{3^{6}}$, $d\in\{2,4,10,28\}$ over $\mathbb{F}_{3^{7}}$, $d\in\{2,4,10,28,82\}$ over $\mathbb{F}_{3^{8}}$, and $d\in\{2,4,10,28,82,244\}$ over $\mathbb{F}_{3^{9}}$.

We carried out our searches on a server with a Dell Inc. Poweredge C4130 motherboard, Intel Xeon CPU E5-2690 v4 @ 2.60GHz, NVIDIA Tesla K80, and 512 GB DDR4 RAM @2300MHz. We give a complete report of the searches conducted and the necessary time in Table A.4.

Let $\alpha$ be the primitive element of $\mathbb{F}_{3^{n}}$ over $\mathbb{F}_{3}$ chosen by MAGMA [BCP-1997-tmasi]. In dimension 6, we find seven new CCZ-classes of planar functions. All of these can be obtained by expanding $x^{2}$ and $x^{10}$ by two or three terms with coefficients in $\mathbb{F}_{3^{2}}$. Two of the classes can be represented by trinomials, namely

$$f_{1}(x)=\alpha^{91}x^{30}+x^{10}+x^{2}$$

and

$$f_{2}(x)=\alpha^{91}x^{486}+x^{10}+x^{2}.$$

Note also that $\alpha^{91}$ is primitive in $\mathbb{F}_{3^{2}}$, and so all the coefficients of these representations lie in the subfield $\mathbb{F}_{3^{2}}$. The remaining five classes do not appear to have a trinomial representation but can be expressed using quadrinomials with coefficients in $\mathbb{F}_{3^{2}}$. These are:

$$\begin{split}f_{3}(x)&=\alpha^{182}x^{82}+{2}x^{10}+\alpha^{91}x^{6}+x^{2},\\ f_{4}(x)&=\alpha^{182}x^{82}+{2}x^{10}+\alpha^{273}x^{6}+x^{2},\\ f_{5}(x)&=\alpha^{91}x^{486}+\alpha^{182}x^{90}+{2}x^{10}+x^{2},\\ f_{6}(x)&=\alpha^{273}x^{486}+\alpha^{182}x^{90}+{2}x^{10}+x^{2},\\ f_{7}(x)&=\alpha^{273}x^{246}+\alpha^{182}x^{82}+\alpha^{91}x^{6}+x^{2}.\end{split}$$

In dimension $9$, we find two new CCZ-classes of planar functions with coefficients in $\mathbb{F}_{3}$. Both of these can be obtained by expanding $x^{2}$ by four elements. They are:

$$\begin{split}f_{1}(x)&=x^{756}+x^{486}+x^{162}+x^{6}+x^{2},\\ f_{2}(x)&=x^{486}+x^{162}+2x^{84}+2x^{18}+x^{2}.\end{split}$$

Throughout all of our computational investigations in $\mathbb{F}_{3^{7}}$, $\mathbb{F}_{3^{8}}$, and $F_{3^{10}}$ we were not able to find any previously unknown new planar functions (up to CCZ-equivalence). However, we found some nice representations with coefficients in $\mathbb{F}_{3}$ of some known planar functions. We recall that this is an important result because it allows us to compute the orbits more efficiently, for instance using [IK-2022-darle, Algorithm 5].

5.2 On the known planar functions in characteristic 3

We now present a full account of the known planar functions over $\mathbb{F}_{3^{n}}$ for $3\leq n\leq 11$, up to CCZ-equivalence. To start, we focus on the procedures used to complete this categorisation and the accompanying data related to invariants. Our results can be found in Appendix A. We use as reference the known families of functions listed in Section 3, the newly discovered sporadic functions over $\mathbb{F}_{3^{6}}$ and $\mathbb{F}_{3^{9}}$ described in Subsection 5.1, and the family of quadrinomials described in Section 4. By doing this, we update the previous classification done in 2010 [PZ-2010-scopf]. We list the representatives of the known CCZ-classes in these dimensions and their CCZ-invariants. Those are the order of the monomial automorphism group, the sizes of the nuclei of the associated semifield in the case of quadratic functions, and the multiset of the cardinalities of each orbit as described in Subsection 2.3. Moreover, we report a list of representatives of the right orbits of each CCZ-class representative we chose. Although this is not an invariant, knowledge of the orbit representatives can be used to speed up the equivalence test defined in [IK-2022-darle] significantly, making it easier to run future tests for equivalence with the proposed representatives. We recall that the orbit representatives are not a linear invariant, but if two functions $F$ and $G$ are linearly equivalent such that $G=L_{1}\circ F\circ L_{2}$, then $RO_{G,x}=RO_{F,L_{2}(x)}$.

Whenever we have some quadratic planar functions that are either sporadic or come from a family, we must verify if each leads to $1$ or $2$ strong isotopic classes by using the Coulter-Henderson Theorem 1. We recall the order of the nuclei are invariants up to isotopism, while the order of the automorphism group and the multiset of the cardinalities of each orbit are only invariants up to strong isotopism.

The classification of known planar functions over $\mathbb{F}_{3^{n}}$ with $n\geq 7$ odd does not require an equivalence test because the known functions are either power functions for which we can use cyclotomic equivalence instead of CCZ-or belong to the two TST instances which are known to be CCZ-inequivalent to any power function. However, computing the invariants in dimension $n=7$ and $n=9$ was important for the expansion search. The cases $n=8$ and $n=10$ are still feasible because we can use all the theoretical results available to reduce the number of tests. The case $n=12$ is left for future work both because of limitations of computational power and because the number of representatives is too high.

In order to compute the nuclei efficiently, we use the following procedure. Let $\mathbb{S}=(\mathbb{F}_{q},+,*)$ be a commutative semifield, then the function that maps $(x,y,z)\in\mathbb{F}_{q}$ to $(x*y)*z$ has a unique polynomial representation $f(x,y,z)$ where all the nonzero monomial in the representation are of the form $x^{a}y^{b}z^{c}$ where $0\leq a,b,c\leq q-1$. Then $N(\mathbb{S})$ is equal to the set of all $\alpha\in\mathbb{F}_{q}$ such that $f(\alpha,x,y)=f(x,y,\alpha)$ since $N(\mathbb{S})=N_{\ell}(\mathbb{S})$. Moreover, the set $N_{m}(\mathbb{S})$ is equal to the set of $\alpha\in\mathbb{F}_{q}$ such that $f(x,\alpha,y)=f(y,\alpha,x)$. This procedure is efficient if constructing $f(x,y,z)$ is not computationally expensive and this can be the case if $\mathbb{S}$ is constructed starting from a planar function $F$ with a sufficiently sparse polynomial representation.

The automorphism group of the associated linear code from [PZ-2010-scopf] is computed in a straightforward way using the Magma algebra system. Unfortunately, this is only possible for $\mathbb{F}_{3^{n}}$ with $n\leq 6$; for higher dimensions, the memory needed to perform the computation becomes prohibitive. We also note that computing the automorphism group of a DO planar function over $\mathbb{F}_{3^{6}}$ equivalent to a power function can be more computationally expensive than checking for equivalence with [IK-2022-darle]. So we took this into account throughout our computations by checking first if a function is equivalent to a power function and only if it is not, we compute the order of its automorphism group.

Linear equivalence tests were performed using [IK-2022-darle, Algorithm 1]. Once a classification was compiled, it was possible to compute the linear invariant using the right orbits, as well as the orbit representatives, using [IK-2022-darle, Algorithm 5]. We made an effort in searching for the class representatives that have polynomial representation with coefficients in the smallest possible subfield, since this significantly speeds up the computation of the orbits.

Appendix A Appendix

A.1 Functions

—c—c—X—c—c— CCZ-inequivalent planar functions over $\mathbb{F}_{3^{n}}$, $n=2,\ldots,8$
n $\mathbf{N^{O}}$ Representative Family Splits

2 2.1 $x^{2}$ FF No
3 3.1 $x^{2}$ FF No
3.2 $x^{4}$ A No
4 4.1 $x^{2}$ FF No
4.2 $x^{14}$ CM NA
4.3 $x^{36}+2x^{10}+2x^{4}$ BH/D/ZP No
5 5.1 $x^{2}$ FF No
5.2 $x^{4}$ A No
5.3 $x^{10}$ A No
5.4 $x^{14}$ CM NA
5.5 $x^{10}+x^{6}+2x^{2}$ TST No
5.6 $x^{10}+2x^{6}+2x^{2}$ TST No
5.7 $x^{90}+x^{2}$ ACW No
5.8 $x^{162}+x^{108}+2x^{84}+x^{2}$ CK No
6 6.1 $x^{2}$ FF No
6.2 $x^{10}$ A No
6.3 $x^{14}$ CM NA
6.4 $x^{122}$ CM NA
6.5 $x^{162}+2x^{84}+x^{28}+x^{2}$ D/ZP No
6.6 $\alpha^{455}x^{270}+x^{28}+\alpha^{273}x^{10}$ BH/ZP 6.7
6.7 $2x^{270}+x^{246}+2x^{90}+x^{82}+x^{54}+2x^{30}+x^{10}+x^{2}$ BH/ZP 6.6
6.8 $x^{270}+2x^{244}+\alpha^{449}x^{162}+\alpha^{449}x^{84}+\alpha^{534}x^{54}+2x^{36}+\alpha^{534}x^{28}+x^{10}+\alpha^{449}x^{6}+\alpha^{279}x^{2}$ G No
6.9 $x^{486}+x^{252}+\alpha^{561}x^{162}+\alpha^{561}x^{84}+\alpha^{183}x^{54}+\alpha^{183}x^{28}+x^{18}+\alpha^{561}x^{6}+\alpha^{209}x^{2}$ CG No
6.10 $x^{162}+2x^{108}+2x^{90}+x^{82}+2x^{10}+x^{4}+x^{2}$ ZP No
6.11 $\alpha^{91}x^{30}+x^{10}+x^{2}$ This work No
6.12 $\alpha^{91}x^{486}+x^{10}+x^{2}$ This work No
6.13 $\alpha^{182}x^{82}+2x^{10}+\alpha^{91}x^{6}+x^{2}$ This work 6.14
6.14 $\alpha^{182}x^{82}+2x^{10}+\alpha^{273}x^{6}+x^{2}$ This work 6.13
6.15 $\alpha^{91}x^{486}+\alpha^{182}x^{90}+2x^{10}+x^{2}$ This work 6.16
6.16 $\alpha^{273}x^{486}+\alpha^{182}x^{90}+2x^{10}+x^{2}$ This work 6.15

6.17 $\alpha^{273}x^{246}+\alpha^{182}x^{82}+\alpha^{91}x^{6}+x^{2}$ This work No
7 7.1 $x^{2}$ FF No
7.2 $x^{4}$ A No
7.3 $x^{10}$ A No
7.4 $x^{28}$ A No
7.5 $x^{14}$ CM NA
7.6 $x^{122}$ CM NA
7.7 $x^{10}+x^{6}+2x^{2}$ TST No
7.8 $x^{10}+2x^{6}+2x^{2}$ TST No

8 8.1 $x^{2}$ FF No
8.2 $x^{14}$ CM NA
8.3 $x^{122}$ CM NA
8.4 $x^{1094}$ CM NA
8.5 $x^{244}+2x^{84}+2x^{82}$ BH 8.6
8.6 $x^{324}+x^{82}+2x^{4}$ BH 8.5
8.7 $x^{1458}+2x^{738}+x^{82}+x^{2}$ B/BH/D/ZP No
8.8 $x^{486}+2x^{246}+x^{82}+x^{2}$ D/ZP No
8.9 $\alpha^{3608}x^{1458}+\alpha^{3608}x^{738}+\alpha^{3810}x^{486}+\alpha^{3810}x^{246}+\alpha^{3413}x^{162}+\alpha^{3413}x^{82}+\alpha^{3608}x^{18}+\alpha^{3810}x^{6}+\alpha^{2565}x^{2}$ CG 8.10
8.10 $\alpha^{164}x^{1458}+\alpha^{164}x^{738}+\alpha^{950}x^{486}+\alpha^{950}x^{246}+\alpha^{616}x^{162}+\alpha^{616}x^{82}+\alpha^{164}x^{18}+\alpha^{950}x^{6}+\alpha^{6297}x^{2}$ CG 8.9
9 9.1 $x^{2}$ FF No
9.2 $x^{4}$ A No
9.3 $x^{10}$ A No
9.4 $x^{28}$ A No
9.5 $x^{82}$ A No
9.6 $x^{122}$ CM NA
9.7 $x^{1094}$ CM NA
9.8 $x^{10}+x^{6}+2x^{2}$ TST No
9.9 $x^{10}+2x^{6}+2x^{2}$ TST No
9.10 $x^{486}+x^{162}+2x^{84}+2x^{18}+x^{2}$ This work No
9.11 $x^{756}+x^{486}+x^{162}+x^{6}+x^{2}$ This work No
10 10.1 $x^{2}$ FF No
10.2 $x^{10}$ A No
10.3 $x^{82}$ A No
10.4 $x^{14}$ CM NA
10.5 $x^{1094}$ CM NA
10.6 $x^{9842}$ CM NA
10.7 $2x^{4374}+2x^{2196}+\alpha^{7686}x^{1458}+\alpha^{7686}x^{732}+\alpha^{244}x^{486}+\alpha^{244}x^{244}+2x^{18}+\alpha^{7686}x^{6}+\alpha^{1220}x^{2}$ CG No
10.8 $x^{1458}+2x^{732}+x^{244}+x^{2}$ D/ZP No
10.9 $x^{13122}+2x^{6588}+x^{244}+x^{2}$ D/ZP No
10.10 $\alpha^{44286}x^{13122}+\alpha^{44286}x^{6588}+x^{4374}+x^{2196}+x^{486}+x^{244}+\alpha^{44286}x^{54}+x^{18}$ PW No
10.11 $x^{2430}+2x^{2188}+\alpha^{14762}x^{1458}+\alpha^{14762}x^{732}+x^{486}+2x^{252}+x^{244}+x^{10}+\alpha^{14762}x^{6}$ G No
10.12 $2x^{2916}+x^{738}+x^{12}$ ZP No
10.13 $x^{21870}+x^{19692}+x^{2430}+2x^{252}$ ZP No
10.14 $x^{21870}+x^{19692}+2x^{2268}$ ZP No
10.15 $x^{21870}+2x^{19692}+x^{2430}+2x^{2268}+x^{2188}+2x^{486}+x^{252}+x^{90}+x^{10}+x^{2}$ ZP No
10.16 $x^{2430}+x^{244}+\alpha^{44286}x^{10}$ BH 10.17
10.17 $x^{6562}+x^{486}+2x^{270}+x^{2}$ BH 10.16
10.18 $x^{19926}+x^{244}+\alpha^{44286}x^{82}$ BH 10.19
10.19 $2x^{19926}+x^{486}+x^{82}+x^{2}$ BH 10.18
10.20 $2x^{19926}+2x^{486}+2x^{82}+x^{2}$ ZP 10.21
10.21 $x^{19764}+x^{2190}+a^{242}x^{738}$ ZP 10.20
10.22 $2x^{2430}+2x^{486}+2x^{10}+x^{2}$ ZP 10.23
10.23 $x^{19764}+a^{242}x^{6562}+x^{732}+x^{270}$ ZP 10.22
11 11.1 $x^{2}$ FF No
11.2 $x^{4}$ A No
11.3 $x^{10}$ A No
11.4 $x^{28}$ A No
11.5 $x^{82}$ A No
11.6 $x^{244}$ A No
11.7 $x^{14}$ CM NA
11.8 $x^{122}$ CM NA
11.9 $x^{1094}$ CM NA
11.10 $x^{9842}$ CM NA
11.11 $x^{10}+x^{6}+2x^{2}$ TST No
11.12 $x^{10}+2x^{6}+2x^{2}$ TST No

A.2 Invariants

—c—c—c—c—c—X— Invariants for the classes in Table A.1
n $\mathbf{N^{O}}$ $|N|$ $|N_{m}|$ Orbits Aut. group order
3 3.1 $3^{3}$ $3^{3}$ $\left\{*26*\right\}$ $4212$
3.2 $3$ $3$ $\left\{*26*\right\}$ $4212$
4 4.1 $3^{4}$ $3^{4}$ $\left\{*80*\right\}$ $51840$
4.2 NA NA $\left\{*80*\right\}$ $640$
4.3 $3$ $3^{2}$ $\left\{*16,64*\right\}$ $10368$
5 5.1 $3^{5}$ $3^{5}$ $\left\{*242*\right\}$ $588060$
5.2 $3$ $3$ $\left\{*242*\right\}$ $588060$
5.3 $3$ $3$ $\left\{*242*\right\}$ $588060$
5.4 NA NA $\left\{*242*\right\}$ $2420$
5.5 $3$ $3$ $\left\{*2,10^{24}*\right\}$ $4860$
5.6 $3$ $3$ $\left\{*2,10^{24}*\right\}$ $4860$
5.7 $3$ $3$ $\left\{*22,110^{2}*\right\}$ $53460$
5.8 $3$ $3$ $\left\{*2,10^{24}*\right\}$ $4860$
6 6.1 $3^{6}$ $3^{6}$ $\left\{*728*\right\}$ $6368544$
6.2 $3^{2}$ $3^{2}$ $\left\{*728*\right\}$ $6368544$
6.3 NA NA $\left\{*728*\right\}$ $8736$
6.4 NA NA $\left\{*728*\right\}$ $8736$
6.5 $3$ $3^{3}$ $\left\{*26^{2},52,156^{4}*\right\}$ $227448$
6.6 $3$ $3^{2}$ $\left\{*104,312^{2}*\right\}$ $454896$
6.7 $3$ $3^{2}$ $\left\{*52^{2},312^{2}*\right\}$ $454896$
6.8 $3$ $3$ $\left\{*26^{4},78^{8}*\right\}$ $113724$
6.9 $3$ $3^{3}$ $\left\{*8^{4},24^{29}*\right\}$ $34992$
6.10 $3$ $3$ $\left\{*26^{2},52,156^{4}*\right\}$ $227448$
6.11 $3$ $3^{2}$ $\left\{*4^{2},12^{60}*\right\}$ $17496$
6.12 $3$ $3^{2}$ $\left\{*4^{2},12^{60}*\right\}$ $17496$
6.13 $3$ $3^{2}$ $\left\{*4^{2},12^{60}*\right\}$ $17496$
6.14 $3$ $3^{2}$ $\left\{*4^{2},12^{60}*\right\}$ $17496$
6.15 $3$ $3^{2}$ $\left\{*4^{2},12^{60}*\right\}$ $17496$
6.16 $3$ $3^{2}$ $\left\{*4^{2},12^{60}*\right\}$ $17496$
6.17 $3$ $3^{2}$ $\left\{*4^{2},12^{60}*\right\}$ $17496$
7 7.1 $3^{7}$ $3^{7}$ $\left\{*2186*\right\}$ $-$
7.2 $3$ $3$ $\left\{*2186*\right\}$ $-$
7.3 $3$ $3$ $\left\{*2186*\right\}$ $-$
7.4 $3$ $3$ $\left\{*2186*\right\}$ $-$
7.5 NA NA $\left\{*2186*\right\}$ $-$
7.6 NA NA $\left\{*2186*\right\}$ $-$
7.7 $3$ $3$ $\left\{*2,14^{156}*\right\}$ $-$
7.8 $3$ $3$ $\left\{*2,14^{156}*\right\}$ $-$
8 8.1 $3^{8}$ $3^{8}$ $\left\{*6560*\right\}$ $-$
8.2 NA NA $\left\{*6560*\right\}$ $-$
8.3 NA NA $\left\{*6560*\right\}$ $-$
8.4 NA NA $\left\{*6560*\right\}$ $-$
8.5 $3$ $3^{2}$ $\left\{*160,1280^{5}*\right\}$ $-$
8.6 $3$ $3^{2}$ $\left\{*160,1280^{5}*\right\}$ $-$
8.7 $3^{2}$ $3^{4}$ $\left\{*160,1280^{5}*\right\}$ $-$
8.8 $3$ $3^{4}$ $\left\{*80^{2},640^{10}*\right\}$ $-$
8.9 $3$ $3^{4}$ $\left\{*16^{410}*\right\}$ $-$
8.10 $3$ $3^{4}$ $\left\{*16^{410}*\right\}$ $-$
9 9.1 $3^{9}$ $3^{9}$ $\left\{*19682*\right\}$ $-$
9.2 $3$ $3$ $\left\{*19682*\right\}$ $-$
9.3 $3$ $3$ $\left\{*19682*\right\}$ $-$
9.4 $3^{3}$ $3^{3}$ $\left\{*19682*\right\}$ $-$
9.5 $3$ $3$ $\left\{*19682*\right\}$ $-$
9.6 NA NA $\left\{*19682*\right\}$ $-$
9.7 NA NA $\left\{*19682*\right\}$ $-$
9.8 $3$ $3$ $\left\{*2,6^{4},18^{1092}*\right\}$ $-$
9.9 $3$ $3$ $\left\{*2,6^{4},18^{1092}*\right\}$ $-$
9.10 $3$ $3^{3}$ $-$ $-$
9.11 $3$ $3^{3}$ $-$ $-$
10 10.1 $3^{10}$ $3^{10}$ $\left\{*59048*\right\}$ $-$
10.2 $3^{2}$ $3^{2}$ $\left\{*59048*\right\}$ $-$
10.3 $3^{2}$ $3^{2}$ $\left\{*59048*\right\}$ $-$
10.4 NA NA $\left\{*59048*\right\}$ $-$
10.5 NA NA $\left\{*59048*\right\}$ $-$
10.6 NA NA $\left\{*59048*\right\}$ $-$
10.7 $3$ $3^{5}$ $-$ $-$
10.8 $3$ $3^{5}$ $\left\{*242^{2},484,2420^{24}*\right\}$ $-$
10.9 $3$ $3^{5}$ $\left\{*242^{2},484,2420^{24}*\right\}$ $-$
10.10 $3$ $3^{5}$ $-$ $-$
10.11 $3$ $3$ $-$ $-$
10.12 $3$ $3$ $\left\{*242^{2},484,2420^{24}*\right\}$ $-$
10.13 $3$ $3$ $\left\{*242^{2},484,2420^{24}*\right\}$ $-$
10.14 $3$ $3$ $\left\{*242^{2},484,2420^{24}*\right\}$ $-$
10.15 $3$ $3$ $\left\{*242^{2},484,2420^{24}*\right\}$ $-$
10.16 $3$ $3^{2}$ $\left\{*968,4840^{12}*\right\}$ $-$
10.17 $3$ $3^{2}$ $\left\{*484^{2},4840^{12}*\right\}$ $-$
10.18 $3$ $3^{2}$ $\left\{*968,4840^{12}*\right\}$ $-$
10.19 $3$ $3^{2}$ $\left\{*484^{2},4840^{12}*\right\}$ $-$
10.20 $3$ $3^{2}$ $\left\{*484^{2},4840^{12}*\right\}$ $-$
10.21 $3$ $3^{2}$ $\left\{*968,4840^{12}*\right\}$ $-$
10.22 $3$ $3^{2}$ $\left\{*484^{2},4840^{12}*\right\}$ $-$
10.23 $3$ $3^{2}$ $\left\{*968,4840^{12}*\right\}$ $-$
11 11.1 $3^{11}$ $3^{11}$ $\left\{*177146*\right\}$ $-$
11.2 $3$ $3$ $\left\{*177146*\right\}$ $-$
11.3 $3$ $3$ $\left\{*177146*\right\}$ $-$
11.4 $3$ $3$ $\left\{*177146*\right\}$ $-$
11.5 $3$ $3$ $\left\{*177146*\right\}$ $-$
11.6 $3$ $3$ $\left\{*177146*\right\}$ $-$
11.7 NA NA $\left\{*177146*\right\}$ $-$
11.8 NA NA $\left\{*177146*\right\}$ $-$
11.9 NA NA $\left\{*177146*\right\}$ $-$
11.10 NA NA $\left\{*177146*\right\}$ $-$
11.11 $3$ $3$ $-$ $-$
11.12 $3$ $3$ $-$ $-$

A.3 Orbits

—c—c—¿X— Right orbit representatives for the classes in Table A.1.
n $\mathbf{N^{O}}$ Orbit representatives ($\alpha^{i}$)
4 4.3 $0,1$
5 5.5 0, 1, 2, 4, 5, 7, 8, 10, 11, 13, 16, 17, 19, 20, 22, 25, 26, 31, 34, 35, 38, 40, 61, 67, 76
5.6 same as 5.5
5.7 $0,1,2$
5.8 same as 5.5
6 6.5 0, 1, 4, 7, 8, 11, 14
6.6 0, 1, 2
6.7 0, 1, 2, 7
6.8 0, 1, 2, 3, 4, 5, 6, 8, 10, 15, 17, 20
6.9 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 13, 16, 17, 19, 22, 23, 31, 34, 35, 36, 38, 39, 44, 45, 47, 48, 50, 54, 66, 72, 90
6.10 0, 1, 2, 6, 8, 13, 15
6.11 Table A.3
6.12 Table A.3
6.13 Table A.3
6.14 Table A.3
6.15 Table A.3
6.16 Table A.3
6.17 Table A.3
7 7.7 Table A.3
7.8 Table A.3
8 8.5 0, 1, 2, 4, 7, 8
8.6 0, 1, 2, 4, 7, 8
8.7 0, 1, 2, 4, 5, 7
8.8 0, 1, 2, 4, 10, 11, 13, 16, 17, 28, 35, 41
8.9 Table A.3
8.10 Table A.3
10 10.8 0, 1, 4, 5, 7, 10, 11, 14, 16, 19, 20, 22, 26, 31, 34, 38, 40, 41, 49, 55, 65, 76, 82, 91, 104, 122, 133
10.9 0, 1, 2, 4, 7, 8, 10, 13, 14, 16, 19, 20, 23, 25, 26, 32, 34, 38, 40, 41, 44, 55, 61, 86, 122, 125, 188
10.12 0, 1, 2, 4, 5, 7, 8, 10, 11, 13, 14, 16, 17, 19, 20, 22, 23, 31, 32, 34, 38, 40, 41, 43, 47, 61, 122
10.13 Same as 10.9
10.14 0, 1, 2, 3, 4, 5, 6, 7, 8, 11, 16, 20, 21, 23, 28, 29, 31, 34, 38, 42, 43, 48, 53, 64, 79, 183, 192
10.15 Same as 10.12
10.16 0, 1, 2, 3, 4, 5, 6, 8, 10, 11, 12, 13, 17
10.17 0, 1, 2, 4, 5, 7, 8, 10, 11, 16, 17, 19, 20, 61
10.18 same as 10.16
10.19 same as 10.17
10.20 same as 10.17
10.21 0, 1, 2, 3, 4, 6, 7, 8, 10, 13, 42, 44, 51
10.22 same as 10.17
10.23 0, 1, 2, 3, 4, 6, 7, 10, 12, 13, 14, 15, 19

—c—X— Right orbits representatives of the classes missing from Table A.3
$\mathbf{N^{O}}$ Orbit representatives ($\alpha^{i}$)
6.11
to
6.17 0, 1, 2, 3, 4, 5, 6, 7, 8, 10, 11, 12, 13, 14, 15, 16, 17, 19, 20, 23, 24, 26, 28, 29, 30, 31, 32, 33, 35, 37, 38, 39, 40, 46, 47, 48, 49, 51, 53, 55, 56, 57, 58, 60, 69, 71, 73, 74, 76, 78, 80, 91, 92, 94, 96, 98, 101, 114, 119, 121, 137, 139
7.7
7.8 0, 1, 2, 4, 5, 7, 8, 10, 11, 13, 14, 16, 17, 19, 20, 22, 23, 25, 26, 28, 29, 31, 32, 34, 35, 37, 38, 40, 43, 44, 46, 47, 49, 50, 52, 53, 55, 56, 58, 59, 61, 62, 64, 65, 67, 70, 71, 73, 74, 76, 77, 79, 80, 85, 86, 88, 89, 91, 92, 94, 97, 98, 100, 101, 103, 104, 106, 107, 110, 112, 113, 115, 116, 118, 119, 121, 137, 139, 142, 143, 145, 146, 148, 151, 152, 154, 155, 157, 160, 161, 169, 170, 172, 173, 175, 178, 179, 181, 182, 184, 187, 188, 193, 196, 197, 199, 200, 202, 211, 214, 215, 220, 223, 224, 226, 227, 229, 233, 235, 236, 238, 241, 242, 274, 277, 278, 281, 283, 295, 296, 301, 304, 305, 308, 310, 317, 319, 322, 323, 337, 344, 346, 349, 350, 358, 359, 362, 364, 547, 553, 562, 565, 589, 592, 607, 688, 715
8.9 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 62, 63, 64, 65, 66, 67, 68, 69, 70, 72, 73, 75, 76, 78, 79, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 99, 100, 101, 102, 103, 104, 105, 107, 109, 110, 111, 112, 113, 114, 115, 117, 118, 119, 120, 121, 122, 123, 125, 126, 127, 129, 130, 131, 132, 133, 134, 135, 136, 137, 139, 142, 143, 146, 147, 148, 149, 150, 151, 153, 154, 155, 157, 158, 159, 161, 162, 163, 164, 165, 166, 167, 171, 172, 173, 174, 175, 177, 178, 179, 181, 182, 183, 185, 186, 188, 189, 190, 191, 192, 195, 196, 197, 199, 200, 201, 202, 203, 204, 205, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 219, 220, 222, 224, 225, 226, 227, 228, 231, 232, 233, 234, 235, 236, 237, 238, 239, 241, 242, 243, 244, 246, 247, 248, 249, 253, 254, 256, 257, 258, 260, 261, 263, 264, 266, 267, 269, 270, 271, 274, 275, 278, 281, 282, 284, 287, 289, 291, 292, 293, 294, 299, 304, 305, 306, 307, 308, 309, 311, 312, 314, 315, 317, 320, 321, 323, 324, 326, 329, 333, 334, 336, 340, 342, 346, 347, 349, 351, 359, 360, 361, 363, 365, 369, 372, 373, 374, 377, 378, 380, 381, 382, 383, 386, 390, 394, 396, 397, 400, 401, 402, 405, 406, 412, 414, 415, 419, 420, 427, 429, 431, 434, 435, 439, 442, 445, 447, 449, 455, 456, 457, 461, 462, 469, 470, 472, 473, 475, 477, 493, 495, 496, 497, 498, 500, 504, 506, 507, 513, 515, 519, 520, 521, 523, 525, 529, 533, 539, 540, 544, 547, 552, 557, 558, 560, 564, 565, 567, 568, 576, 583, 584, 591, 593, 598, 601, 611, 613, 620, 621, 623, 637, 641, 643, 646, 649, 652, 653, 656, 658, 659, 667, 668, 670, 679, 681, 685, 694, 697, 699, 703, 704, 716, 721, 724, 736, 775, 784, 787, 791, 796, 807, 810, 816, 822, 830, 837, 845, 855, 865, 867, 880, 933, 935, 939, 954, 975, 986, 994, 1080, 1107, 1134, 1194
8.10 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 46, 47, 48, 49, 50, 51, 53, 54, 56, 57, 58, 59, 60, 61, 62, 63, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 84, 85, 86, 87, 88, 89, 90, 92, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 111, 112, 113, 114, 115, 116, 117, 118, 120, 121, 122, 124, 125, 126, 128, 129, 132, 133, 134, 136, 138, 139, 140, 141, 143, 144, 145, 146, 147, 149, 150, 151, 152, 153, 154, 155, 156, 157, 159, 161, 162, 163, 164, 166, 167, 168, 169, 170, 171, 172, 173, 174, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 188, 189, 190, 195, 196, 197, 198, 199, 200, 204, 206, 207, 211, 212, 213, 214, 215, 216, 217, 218, 220, 221, 222, 223, 225, 226, 227, 228, 229, 232, 233, 235, 236, 237, 238, 240, 241, 242, 243, 244, 246, 249, 251, 252, 253, 254, 256, 258, 259, 260, 267, 269, 272, 275, 276, 277, 278, 281, 282, 284, 286, 288, 289, 292, 293, 294, 295, 296, 298, 300, 302, 305, 307, 311, 313, 314, 316, 318, 323, 326, 327, 328, 330, 331, 337, 338, 340, 342, 343, 346, 347, 348, 351, 353, 360, 361, 366, 368, 371, 372, 374, 376, 377, 378, 379, 380, 382, 384, 386, 395, 397, 401, 402, 404, 405, 407, 414, 415, 416, 419, 420, 422, 426, 430, 431, 436, 440, 441, 444, 446, 456, 458, 468, 469, 475, 478, 480, 491, 494, 508, 510, 511, 512, 517, 522, 524, 530, 532, 535, 536, 542, 546, 548, 550, 561, 565, 566, 567, 569, 571, 572, 573, 578, 579, 581, 582, 583, 586, 589, 590, 595, 596, 599, 600, 606, 607, 608, 610, 633, 639, 642, 643, 646, 647, 661, 669, 682, 686, 687, 688, 696, 699, 706, 710, 712, 717, 728, 736, 737, 740, 741, 747, 750, 752, 753, 756, 757, 763, 770, 787, 811, 812, 825, 830, 841, 846, 850, 851, 856, 866, 890, 901, 904, 920, 921, 927, 934, 942, 975, 976, 981, 989, 996, 1005, 1020, 1043, 1049, 1052, 1106, 1109, 1116, 1160, 1260, 1337, 1385
9.8
9.9 $0\mathchar 44\relax\penalty 01\mathchar 44\relax\penalty 02\mathchar 44\relax\penalty 04\mathchar 44\relax\penalty 05\mathchar 44\relax\penalty 07\mathchar 44\relax\penalty 08\mathchar 44\relax\penalty 010\mathchar 44\relax\penalty 011\mathchar 44\relax\penalty 013\mathchar 44\relax\penalty 014\mathchar 44\relax\penalty 016\mathchar 44\relax\penalty 017\mathchar 44\relax\penalty 019\mathchar 44\relax\penalty 020\mathchar 44\relax\penalty 022\mathchar 44\relax\penalty 023\mathchar 44\relax\penalty 025\mathchar 44\relax\penalty 026\mathchar 44\relax\penalty 028\mathchar 44\relax\penalty 029\mathchar 44\relax\penalty 031\mathchar 44\relax\penalty 032\mathchar 44\relax\penalty 034\mathchar 44\relax\penalty 035\mathchar 44\relax\penalty 037\mathchar 44\relax\penalty 038\mathchar 44\relax\penalty 040\mathchar 44\relax\penalty 041\mathchar 44\relax\penalty 043\mathchar 44\relax\penalty 044\mathchar 44\relax\penalty 046\mathchar 44\relax\penalty 047\mathchar 44\relax\penalty 049\mathchar 44\relax\penalty 050\mathchar 44\relax\penalty 052\mathchar 44\relax\penalty 053\mathchar 44\relax\penalty 055\mathchar 44\relax\penalty 056\mathchar 44\relax\penalty 058\mathchar 44\relax\penalty 059\mathchar 44\relax\penalty 061\mathchar 44\relax\penalty 062\mathchar 44\relax\penalty 064\mathchar 44\relax\penalty 065\mathchar 44\relax\penalty 067\mathchar 44\relax\penalty 068\mathchar 44\relax\penalty 070\mathchar 44\relax\penalty 071\mathchar 44\relax\penalty 073\mathchar 44\relax\penalty 074\mathchar 44\relax\penalty 076\mathchar 44\relax\penalty 077\mathchar 44\relax\penalty 079\mathchar 44\relax\penalty 080\mathchar 44\relax\penalty 082\mathchar 44\relax\penalty 083\mathchar 44\relax\penalty 085\mathchar 44\relax\penalty 086\mathchar 44\relax\penalty 088\mathchar 44\relax\penalty 089\mathchar 44\relax\penalty 091\mathchar 44\relax\penalty 092\mathchar 44\relax\penalty 094\mathchar 44\relax\penalty 095\mathchar 44\relax\penalty 097\mathchar 44\relax\penalty 098\mathchar 44\relax\penalty 0100\mathchar 44\relax\penalty 0101\mathchar 44\relax\penalty 0103\mathchar 44\relax\penalty 0104\mathchar 44\relax\penalty 0106\mathchar 44\relax\penalty 0107\mathchar 44\relax\penalty 0109\mathchar 44\relax\penalty 0110\mathchar 44\relax\penalty 0112\mathchar 44\relax\penalty 0113\mathchar 44\relax\penalty 0115\mathchar 44\relax\penalty 0116\mathchar 44\relax\penalty 0118\mathchar 44\relax\penalty 0119\mathchar 44\relax\penalty 0121\mathchar 44\relax\penalty 0124\mathchar 44\relax\penalty 0125\mathchar 44\relax\penalty 0127\mathchar 44\relax\penalty 0128\mathchar 44\relax\penalty 0130\mathchar 44\relax\penalty 0131\mathchar 44\relax\penalty 0133\mathchar 44\relax\penalty 0134\mathchar 44\relax\penalty 0136\mathchar 44\relax\penalty 0137\mathchar 44\relax\penalty 0139\mathchar 44\relax\penalty 0140\mathchar 44\relax\penalty 0142\mathchar 44\relax\penalty 0143\mathchar 44\relax\penalty 0145\mathchar 44\relax\penalty 0146\mathchar 44\relax\penalty 0148\mathchar 44\relax\penalty 0149\mathchar 44\relax\penalty 0151\mathchar 44\relax\penalty 0152\mathchar 44\relax\penalty 0154\mathchar 44\relax\penalty 0155\mathchar 44\relax\penalty 0157\mathchar 44\relax\penalty 0158\mathchar 44\relax\penalty 0160\mathchar 44\relax\penalty 0161\mathchar 44\relax\penalty 0163\mathchar 44\relax\penalty 0164\mathchar 44\relax\penalty 0166\mathchar 44\relax\penalty 0167\mathchar 44\relax\penalty 0169\mathchar 44\relax\penalty 0170\mathchar 44\relax\penalty 0172\mathchar 44\relax\penalty 0173\mathchar 44\relax\penalty 0175\mathchar 44\relax\penalty 0176\mathchar 44\relax\penalty 0178\mathchar 44\relax\penalty 0179\mathchar 44\relax\penalty 0181\mathchar 44\relax\penalty 0182\mathchar 44\relax\penalty 0184\mathchar 44\relax\penalty 0185\mathchar 44\relax\penalty 0187\mathchar 44\relax\penalty 0188\mathchar 44\relax\penalty 0190\mathchar 44\relax\penalty 0191\mathchar 44\relax\penalty 0193\mathchar 44\relax\penalty 0194\mathchar 44\relax\penalty 0196\mathchar 44\relax\penalty 0197\mathchar 44\relax\penalty 0199\mathchar 44\relax\penalty 0200\mathchar 44\relax\penalty 0202\mathchar 44\relax\penalty 0205\mathchar 44\relax\penalty 0206\mathchar 44\relax\penalty 0208\mathchar 44\relax\penalty 0209\mathchar 44\relax\penalty 0211\mathchar 44\relax\penalty 0212\mathchar 44\relax\penalty 0214\mathchar 44\relax\penalty 0215\mathchar 44\relax\penalty 0217\mathchar 44\relax\penalty 0218\mathchar 44\relax\penalty 0220\mathchar 44\relax\penalty 0221\mathchar 44\relax\penalty 0223\mathchar 44\relax\penalty 0224\mathchar 44\relax\penalty 0226\mathchar 44\relax\penalty 0227\mathchar 44\relax\penalty 0229\mathchar 44\relax\penalty 0230\mathchar 44\relax\penalty 0232\mathchar 44\relax\penalty 0233\mathchar 44\relax\penalty 0235\mathchar 44\relax\penalty 0236\mathchar 44\relax\penalty 0238\mathchar 44\relax\penalty 0239\mathchar 44\relax\penalty 0241\mathchar 44\relax\penalty 0242\mathchar 44\relax\penalty 0247\mathchar 44\relax\penalty 0248\mathchar 44\relax\penalty 0250\mathchar 44\relax\penalty 0251\mathchar 44\relax\penalty 0253\mathchar 44\relax\penalty 0254\mathchar 44\relax\penalty 0256\mathchar 44\relax\penalty 0257\mathchar 44\relax\penalty 0259\mathchar 44\relax\penalty 0260\mathchar 44\relax\penalty 0262\mathchar 44\relax\penalty 0263\mathchar 44\relax\penalty 0265\mathchar 44\relax\penalty 0266\mathchar 44\relax\penalty 0268\mathchar 44\relax\penalty 0269\mathchar 44\relax\penalty 0271\mathchar 44\relax\penalty 0272\mathchar 44\relax\penalty 0274\mathchar 44\relax\penalty 0275\mathchar 44\relax\penalty 0277\mathchar 44\relax\penalty 0278\mathchar 44\relax\penalty 0280\mathchar 44\relax\penalty 0281\mathchar 44\relax\penalty 0283\mathchar 44\relax\penalty 0286\mathchar 44\relax\penalty 0287\mathchar 44\relax\penalty 0289\mathchar 44\relax\penalty 0290\mathchar 44\relax\penalty 0292\mathchar 44\relax\penalty 0293\mathchar 44\relax\penalty 0295\mathchar 44\relax\penalty 0296\mathchar 44\relax\penalty 0298\mathchar 44\relax\penalty 0299\mathchar 44\relax\penalty 0301\mathchar 44\relax\penalty 0302\mathchar 44\relax\penalty 0304\mathchar 44\relax\penalty 0305\mathchar 44\relax\penalty 0307\mathchar 44\relax\penalty 0308\mathchar 44\relax\penalty 0310\mathchar 44\relax\penalty 0311\mathchar 44\relax\penalty 0313\mathchar 44\relax\penalty 0314\mathchar 44\relax\penalty 0316\mathchar 44\relax\penalty 0317\mathchar 44\relax\penalty 0319\mathchar 44\relax\penalty 0320\mathchar 44\relax\penalty 0322\mathchar 44\relax\penalty 0323\mathchar 44\relax\penalty 0326\mathchar 44\relax\penalty 0328\mathchar 44\relax\penalty 0329\mathchar 44\relax\penalty 0331\mathchar 44\relax\penalty 0332\mathchar 44\relax\penalty 0334\mathchar 44\relax\penalty 0335\mathchar 44\relax\penalty 0337\mathchar 44\relax\penalty 0338\mathchar 44\relax\penalty 0340\mathchar 44\relax\penalty 0341\mathchar 44\relax\penalty 0343\mathchar 44\relax\penalty 0344\mathchar 44\relax\penalty 0346\mathchar 44\relax\penalty 0347\mathchar 44\relax\penalty 0349\mathchar 44\relax\penalty 0350\mathchar 44\relax\penalty 0352\mathchar 44\relax\penalty 0353\mathchar 44\relax\penalty 0355\mathchar 44\relax\penalty 0356\mathchar 44\relax\penalty 0358\mathchar 44\relax\penalty 0359\mathchar 44\relax\penalty 0361\mathchar 44\relax\penalty 0362\mathchar 44\relax\penalty 0364\mathchar 44\relax\penalty 0379\mathchar 44\relax\penalty 0380\mathchar 44\relax\penalty 0382\mathchar 44\relax\penalty 0383\mathchar 44\relax\penalty 0385\mathchar 44\relax\penalty 0386\mathchar 44\relax\penalty 0388\mathchar 44\relax\penalty 0389\mathchar 44\relax\penalty 0391\mathchar 44\relax\penalty 0394\mathchar 44\relax\penalty 0395\mathchar 44\relax\penalty 0397\mathchar 44\relax\penalty 0398\mathchar 44\relax\penalty 0400\mathchar 44\relax\penalty 0401\mathchar 44\relax\penalty 0403\mathchar 44\relax\penalty 0404\mathchar 44\relax\penalty 0407\mathchar 44\relax\penalty 0409\mathchar 44\relax\penalty 0410\mathchar 44\relax\penalty 0412\mathchar 44\relax\penalty 0413\mathchar 44\relax\penalty 0415\mathchar 44\relax\penalty 0416\mathchar 44\relax\penalty 0418\mathchar 44\relax\penalty 0421\mathchar 44\relax\penalty 0422\mathchar 44\relax\penalty 0424\mathchar 44\relax\penalty 0425\mathchar 44\relax\penalty 0427\mathchar 44\relax\penalty 0428\mathchar 44\relax\penalty 0430\mathchar 44\relax\penalty 0431\mathchar 44\relax\penalty 0433\mathchar 44\relax\penalty 0434\mathchar 44\relax\penalty 0436\mathchar 44\relax\penalty 0437\mathchar 44\relax\penalty 0439\mathchar 44\relax\penalty 0440\mathchar 44\relax\penalty 0442\mathchar 44\relax\penalty 0443\mathchar 44\relax\penalty 0445\mathchar 44\relax\penalty 0448\mathchar 44\relax\penalty 0449\mathchar 44\relax\penalty 0451\mathchar 44\relax\penalty 0452\mathchar 44\relax\penalty 0454\mathchar 44\relax\penalty 0455\mathchar 44\relax\penalty 0457\mathchar 44\relax\penalty 0458\mathchar 44\relax\penalty 0460\mathchar 44\relax\penalty 0461\mathchar 44\relax\penalty 0463\mathchar 44\relax\penalty 0464\mathchar 44\relax\penalty 0466\mathchar 44\relax\penalty 0467\mathchar 44\relax\penalty 0469\mathchar 44\relax\penalty 0470\mathchar 44\relax\penalty 0472\mathchar 44\relax\penalty 0475\mathchar 44\relax\penalty 0476\mathchar 44\relax\penalty 0478\mathchar 44\relax\penalty 0479\mathchar 44\relax\penalty 0481\mathchar 44\relax\penalty 0482\mathchar 44\relax\penalty 0484\mathchar 44\relax\penalty 0485\mathchar 44\relax\penalty 0493\mathchar 44\relax\penalty 0494\mathchar 44\relax\penalty 0496\mathchar 44\relax\penalty 0497\mathchar 44\relax\penalty 0499\mathchar 44\relax\penalty 0502\mathchar 44\relax\penalty 0503\mathchar 44\relax\penalty 0505\mathchar 44\relax\penalty 0506\mathchar 44\relax\penalty 0508\mathchar 44\relax\penalty 0509\mathchar 44\relax\penalty 0511\mathchar 44\relax\penalty 0512\mathchar 44\relax\penalty 0514\mathchar 44\relax\penalty 0515\mathchar 44\relax\penalty 0517\mathchar 44\relax\penalty 0518\mathchar 44\relax\penalty 0520\mathchar 44\relax\penalty 0521\mathchar 44\relax\penalty 0523\mathchar 44\relax\penalty 0524\mathchar 44\relax\penalty 0526\mathchar 44\relax\penalty 0529\mathchar 44\relax\penalty 0530\mathchar 44\relax\penalty 0532\mathchar 44\relax\penalty 0533\mathchar 44\relax\penalty 0535\mathchar 44\relax\penalty 0536\mathchar 44\relax\penalty 0538\mathchar 44\relax\penalty 0539\mathchar 44\relax\penalty 0541\mathchar 44\relax\penalty 0542\mathchar 44\relax\penalty 0544\mathchar 44\relax\penalty 0545\mathchar 44\relax\penalty 0547\mathchar 44\relax\penalty 0548\mathchar 44\relax\penalty 0550\mathchar 44\relax\penalty 0551\mathchar 44\relax\penalty 0553\mathchar 44\relax\penalty 0556\mathchar 44\relax\penalty 0557\mathchar 44\relax\penalty 0559\mathchar 44\relax\penalty 0560\mathchar 44\relax\penalty 0562\mathchar 44\relax\penalty 0563\mathchar 44\relax\penalty 0565\mathchar 44\relax\penalty 0566\mathchar 44\relax\penalty 0571\mathchar 44\relax\penalty 0572\mathchar 44\relax\penalty 0574\mathchar 44\relax\penalty 0575\mathchar 44\relax\penalty 0577\mathchar 44\relax\penalty 0578\mathchar 44\relax\penalty 0580\mathchar 44\relax\penalty 0583\mathchar 44\relax\penalty 0584\mathchar 44\relax\penalty 0586\mathchar 44\relax\penalty 0587\mathchar 44\relax\penalty 0589\mathchar 44\relax\penalty 0590\mathchar 44\relax\penalty 0592\mathchar 44\relax\penalty 0593\mathchar 44\relax\penalty 0595\mathchar 44\relax\penalty 0596\mathchar 44\relax\penalty 0598\mathchar 44\relax\penalty 0599\mathchar 44\relax\penalty 0601\mathchar 44\relax\penalty 0602\mathchar 44\relax\penalty 0604\mathchar 44\relax\penalty 0605\mathchar 44\relax\penalty 0607\mathchar 44\relax\penalty 0616\mathchar 44\relax\penalty 0617\mathchar 44\relax\penalty 0619\mathchar 44\relax\penalty 0620\mathchar 44\relax\penalty 0622\mathchar 44\relax\penalty 0623\mathchar 44\relax\penalty 0625\mathchar 44\relax\penalty 0626\mathchar 44\relax\penalty 0628\mathchar 44\relax\penalty 0629\mathchar 44\relax\penalty 0631\mathchar 44\relax\penalty 0632\mathchar 44\relax\penalty 0634\mathchar 44\relax\penalty 0637\mathchar 44\relax\penalty 0638\mathchar 44\relax\penalty 0640\mathchar 44\relax\penalty 0641\mathchar 44\relax\penalty 0643\mathchar 44\relax\penalty 0644\mathchar 44\relax\penalty 0646\mathchar 44\relax\penalty 0647\mathchar 44\relax\penalty 0652\mathchar 44\relax\penalty 0653\mathchar 44\relax\penalty 0655\mathchar 44\relax\penalty 0656\mathchar 44\relax\penalty 0658\mathchar 44\relax\penalty 0659\mathchar 44\relax\penalty 0661\mathchar 44\relax\penalty 0664\mathchar 44\relax\penalty 0665\mathchar 44\relax\penalty 0667\mathchar 44\relax\penalty 0668\mathchar 44\relax\penalty 0670\mathchar 44\relax\penalty 0671\mathchar 44\relax\penalty 0673\mathchar 44\relax\penalty 0674\mathchar 44\relax\penalty 0676\mathchar 44\relax\penalty 0677\mathchar 44\relax\penalty 0679\mathchar 44\relax\penalty 0680\mathchar 44\relax\penalty 0682\mathchar 44\relax\penalty 0683\mathchar 44\relax\penalty 0685\mathchar 44\relax\penalty 0686\mathchar 44\relax\penalty 0688\mathchar 44\relax\penalty 0692\mathchar 44\relax\penalty 0694\mathchar 44\relax\penalty 0695\mathchar 44\relax\penalty 0697\mathchar 44\relax\penalty 0698\mathchar 44\relax\penalty 0700\mathchar 44\relax\penalty 0701\mathchar 44\relax\penalty 0703\mathchar 44\relax\penalty 0704\mathchar 44\relax\penalty 0706\mathchar 44\relax\penalty 0707\mathchar 44\relax\penalty 0709\mathchar 44\relax\penalty 0710\mathchar 44\relax\penalty 0712\mathchar 44\relax\penalty 0713\mathchar 44\relax\penalty 0715\mathchar 44\relax\penalty 0718\mathchar 44\relax\penalty 0719\mathchar 44\relax\penalty 0721\mathchar 44\relax\penalty 0722\mathchar 44\relax\penalty 0724\mathchar 44\relax\penalty 0725\mathchar 44\relax\penalty 0727\mathchar 44\relax\penalty 0728\mathchar 44\relax\penalty 0757\mathchar 44\relax\penalty 0758\mathchar 44\relax\penalty 0760\mathchar 44\relax\penalty 0761\mathchar 44\relax\penalty 0763\mathchar 44\relax\penalty 0764\mathchar 44\relax\penalty 0766\mathchar 44\relax\penalty 0767\mathchar 44\relax\penalty 0769\mathchar 44\relax\penalty 0773\mathchar 44\relax\penalty 0775\mathchar 44\relax\penalty 0776\mathchar 44\relax\penalty 0778\mathchar 44\relax\penalty 0779\mathchar 44\relax\penalty 0781\mathchar 44\relax\penalty 0782\mathchar 44\relax\penalty 0785\mathchar 44\relax\penalty 0787\mathchar 44\relax\penalty 0788\mathchar 44\relax\penalty 0790\mathchar 44\relax\penalty 0791\mathchar 44\relax\penalty 0793\mathchar 44\relax\penalty 0794\mathchar 44\relax\penalty 0796\mathchar 44\relax\penalty 0799\mathchar 44\relax\penalty 0800\mathchar 44\relax\penalty 0802\mathchar 44\relax\penalty 0803\mathchar 44\relax\penalty 0805\mathchar 44\relax\penalty 0806\mathchar 44\relax\penalty 0808\mathchar 44\relax\penalty 0809\mathchar 44\relax\penalty 0814\mathchar 44\relax\penalty 0815\mathchar 44\relax\penalty 0817\mathchar 44\relax\penalty 0818\mathchar 44\relax\penalty 0820\mathchar 44\relax\penalty 0821\mathchar 44\relax\penalty 0823\mathchar 44\relax\penalty 0826\mathchar 44\relax\penalty 0827\mathchar 44\relax\penalty 0829\mathchar 44\relax\penalty 0830\mathchar 44\relax\penalty 0832\mathchar 44\relax\penalty 0833\mathchar 44\relax\penalty 0835\mathchar 44\relax\penalty 0836\mathchar 44\relax\penalty 0839\mathchar 44\relax\penalty 0841\mathchar 44\relax\penalty 0842\mathchar 44\relax\penalty 0844\mathchar 44\relax\penalty 0845\mathchar 44\relax\penalty 0847\mathchar 44\relax\penalty 0848\mathchar 44\relax\penalty 0850\mathchar 44\relax\penalty 0862\mathchar 44\relax\penalty 0863\mathchar 44\relax\penalty 0866\mathchar 44\relax\penalty 0868\mathchar 44\relax\penalty 0869\mathchar 44\relax\penalty 0871\mathchar 44\relax\penalty 0872\mathchar 44\relax\penalty 0874\mathchar 44\relax\penalty 0875\mathchar 44\relax\penalty 0877\mathchar 44\relax\penalty 0880\mathchar 44\relax\penalty 0881\mathchar 44\relax\penalty 0883\mathchar 44\relax\penalty 0884\mathchar 44\relax\penalty 0886\mathchar 44\relax\penalty 0887\mathchar 44\relax\penalty 0889\mathchar 44\relax\penalty 0890\mathchar 44\relax\penalty 0895\mathchar 44\relax\penalty 0896\mathchar 44\relax\penalty 0898\mathchar 44\relax\penalty 0899\mathchar 44\relax\penalty 0901\mathchar 44\relax\penalty 0902\mathchar 44\relax\penalty 0904\mathchar 44\relax\penalty 0907\mathchar 44\relax\penalty 0908\mathchar 44\relax\penalty 0910\mathchar 44\relax\penalty 0911\mathchar 44\relax\penalty 0913\mathchar 44\relax\penalty 0914\mathchar 44\relax\penalty 0916\mathchar 44\relax\penalty 0917\mathchar 44\relax\penalty 0920\mathchar 44\relax\penalty 0922\mathchar 44\relax\penalty 0923\mathchar 44\relax\penalty 0925\mathchar 44\relax\penalty 0926\mathchar 44\relax\penalty 0928\mathchar 44\relax\penalty 0929\mathchar 44\relax\penalty 0931\mathchar 44\relax\penalty 0937\mathchar 44\relax\penalty 0938\mathchar 44\relax\penalty 0940\mathchar 44\relax\penalty 0941\mathchar 44\relax\penalty 0943\mathchar 44\relax\penalty 0944\mathchar 44\relax\penalty 0947\mathchar 44\relax\penalty 0949\mathchar 44\relax\penalty 0950\mathchar 44\relax\penalty 0952\mathchar 44\relax\penalty 0953\mathchar 44\relax\penalty 0955\mathchar 44\relax\penalty 0956\mathchar 44\relax\penalty 0958\mathchar 44\relax\penalty 0961\mathchar 44\relax\penalty 0962\mathchar 44\relax\penalty 0964\mathchar 44\relax\penalty 0965\mathchar 44\relax\penalty 0967\mathchar 44\relax\penalty 0968\mathchar 44\relax\penalty 0970\mathchar 44\relax\penalty 0971\mathchar 44\relax\penalty 0985\mathchar 44\relax\penalty 0988\mathchar 44\relax\penalty 0989\mathchar 44\relax\penalty 0991\mathchar 44\relax\penalty 0992\mathchar 44\relax\penalty 0994\mathchar 44\relax\penalty 0995\mathchar 44\relax\penalty 0997\mathchar 44\relax\penalty 0998\mathchar 44\relax\penalty 01001\mathchar 44\relax\penalty 01003\mathchar 44\relax\penalty 01004\mathchar 44\relax\penalty 01006\mathchar 44\relax\penalty 01007\mathchar 44\relax\penalty 01009\mathchar 44\relax\penalty 01010\mathchar 44\relax\penalty 01012\mathchar 44\relax\penalty 01018\mathchar 44\relax\penalty 01019\mathchar 44\relax\penalty 01021\mathchar 44\relax\penalty 01022\mathchar 44\relax\penalty 01024\mathchar 44\relax\penalty 01025\mathchar 44\relax\penalty 01028\mathchar 44\relax\penalty 01030\mathchar 44\relax\penalty 01031\mathchar 44\relax\penalty 01033\mathchar 44\relax\penalty 01034\mathchar 44\relax\penalty 01036\mathchar 44\relax\penalty 01037\mathchar 44\relax\penalty 01039\mathchar 44\relax\penalty 01042\mathchar 44\relax\penalty 01043\mathchar 44\relax\penalty 01045\mathchar 44\relax\penalty 01046\mathchar 44\relax\penalty 01048\mathchar 44\relax\penalty 01049\mathchar 44\relax\penalty 01051\mathchar 44\relax\penalty 01052\mathchar 44\relax\penalty 01058\mathchar 44\relax\penalty 01060\mathchar 44\relax\penalty 01061\mathchar 44\relax\penalty 01063\mathchar 44\relax\penalty 01064\mathchar 44\relax\penalty 01066\mathchar 44\relax\penalty 01069\mathchar 44\relax\penalty 01070\mathchar 44\relax\penalty 01072\mathchar 44\relax\penalty 01073\mathchar 44\relax\penalty 01075\mathchar 44\relax\penalty 01076\mathchar 44\relax\penalty 01078\mathchar 44\relax\penalty 01079\mathchar 44\relax\penalty 01082\mathchar 44\relax\penalty 01084\mathchar 44\relax\penalty 01085\mathchar 44\relax\penalty 01087\mathchar 44\relax\penalty 01088\mathchar 44\relax\penalty 01090\mathchar 44\relax\penalty 01091\mathchar 44\relax\penalty 01093\mathchar 44\relax\penalty 01231\mathchar 44\relax\penalty 01232\mathchar 44\relax\penalty 01234\mathchar 44\relax\penalty 01235\mathchar 44\relax\penalty 01237\mathchar 44\relax\penalty 01240\mathchar 44\relax\penalty 01241\mathchar 44\relax\penalty 01244\mathchar 44\relax\penalty 01246\mathchar 44\relax\penalty 01249\mathchar 44\relax\penalty 01250\mathchar 44\relax\penalty 01252\mathchar 44\relax\penalty 01253\mathchar 44\relax\penalty 01255\mathchar 44\relax\penalty 01261\mathchar 44\relax\penalty 01262\mathchar 44\relax\penalty 01264\mathchar 44\relax\penalty 01267\mathchar 44\relax\penalty 01268\mathchar 44\relax\penalty 01271\mathchar 44\relax\penalty 01273\mathchar 44\relax\penalty 01276\mathchar 44\relax\penalty 01277\mathchar 44\relax\penalty 01279\mathchar 44\relax\penalty 01280\mathchar 44\relax\penalty 01282\mathchar 44\relax\penalty 01285\mathchar 44\relax\penalty 01286\mathchar 44\relax\penalty 01288\mathchar 44\relax\penalty 01289\mathchar 44\relax\penalty 01291\mathchar 44\relax\penalty 01294\mathchar 44\relax\penalty 01295\mathchar 44\relax\penalty 01303\mathchar 44\relax\penalty 01304\mathchar 44\relax\penalty 01306\mathchar 44\relax\penalty 01307\mathchar 44\relax\penalty 01309\mathchar 44\relax\penalty 01312\mathchar 44\relax\penalty 01313\mathchar 44\relax\penalty 01315\mathchar 44\relax\penalty 01316\mathchar 44\relax\penalty 01318\mathchar 44\relax\penalty 01321\mathchar 44\relax\penalty 01322\mathchar 44\relax\penalty 01325\mathchar 44\relax\penalty 01327\mathchar 44\relax\penalty 01330\mathchar 44\relax\penalty 01331\mathchar 44\relax\penalty 01333\mathchar 44\relax\penalty 01334\mathchar 44\relax\penalty 01336\mathchar 44\relax\penalty 01354\mathchar 44\relax\penalty 01357\mathchar 44\relax\penalty 01358\mathchar 44\relax\penalty 01360\mathchar 44\relax\penalty 01361\mathchar 44\relax\penalty 01363\mathchar 44\relax\penalty 01366\mathchar 44\relax\penalty 01367\mathchar 44\relax\penalty 01369\mathchar 44\relax\penalty 01370\mathchar 44\relax\penalty 01372\mathchar 44\relax\penalty 01375\mathchar 44\relax\penalty 01376\mathchar 44\relax\penalty 01384\mathchar 44\relax\penalty 01385\mathchar 44\relax\penalty 01387\mathchar 44\relax\penalty 01388\mathchar 44\relax\penalty 01390\mathchar 44\relax\penalty 01393\mathchar 44\relax\penalty 01394\mathchar 44\relax\penalty 01396\mathchar 44\relax\penalty 01397\mathchar 44\relax\penalty 01399\mathchar 44\relax\penalty 01402\mathchar 44\relax\penalty 01403\mathchar 44\relax\penalty 01406\mathchar 44\relax\penalty 01408\mathchar 44\relax\penalty 01411\mathchar 44\relax\penalty 01412\mathchar 44\relax\penalty 01414\mathchar 44\relax\penalty 01415\mathchar 44\relax\penalty 01417\mathchar 44\relax\penalty 01424\mathchar 44\relax\penalty 01426\mathchar 44\relax\penalty 01429\mathchar 44\relax\penalty 01430\mathchar 44\relax\penalty 01433\mathchar 44\relax\penalty 01435\mathchar 44\relax\penalty 01438\mathchar 44\relax\penalty 01439\mathchar 44\relax\penalty 01441\mathchar 44\relax\penalty 01442\mathchar 44\relax\penalty 01444\mathchar 44\relax\penalty 01447\mathchar 44\relax\penalty 01448\mathchar 44\relax\penalty 01450\mathchar 44\relax\penalty 01451\mathchar 44\relax\penalty 01453\mathchar 44\relax\penalty 01456\mathchar 44\relax\penalty 01457\mathchar 44\relax\penalty 01514\mathchar 44\relax\penalty 01516\mathchar 44\relax\penalty 01519\mathchar 44\relax\penalty 01520\mathchar 44\relax\penalty 01522\mathchar 44\relax\penalty 01523\mathchar 44\relax\penalty 01525\mathchar 44\relax\penalty 01528\mathchar 44\relax\penalty 01529\mathchar 44\relax\penalty 01531\mathchar 44\relax\penalty 01532\mathchar 44\relax\penalty 01534\mathchar 44\relax\penalty 01537\mathchar 44\relax\penalty 01538\mathchar 44\relax\penalty 01546\mathchar 44\relax\penalty 01547\mathchar 44\relax\penalty 01549\mathchar 44\relax\penalty 01550\mathchar 44\relax\penalty 01552\mathchar 44\relax\penalty 01555\mathchar 44\relax\penalty 01556\mathchar 44\relax\penalty 01558\mathchar 44\relax\penalty 01559\mathchar 44\relax\penalty 01561\mathchar 44\relax\penalty 01564\mathchar 44\relax\penalty 01565\mathchar 44\relax\penalty 01570\mathchar 44\relax\penalty 01573\mathchar 44\relax\penalty 01574\mathchar 44\relax\penalty 01576\mathchar 44\relax\penalty 01577\mathchar 44\relax\penalty 01579\mathchar 44\relax\penalty 01600\mathchar 44\relax\penalty 01601\mathchar 44\relax\penalty 01603\mathchar 44\relax\penalty 01604\mathchar 44\relax\penalty 01606\mathchar 44\relax\penalty 01609\mathchar 44\relax\penalty 01610\mathchar 44\relax\penalty 01612\mathchar 44\relax\penalty 01613\mathchar 44\relax\penalty 01615\mathchar 44\relax\penalty 01618\mathchar 44\relax\penalty 01619\mathchar 44\relax\penalty 01627\mathchar 44\relax\penalty 01628\mathchar 44\relax\penalty 01630\mathchar 44\relax\penalty 01631\mathchar 44\relax\penalty 01633\mathchar 44\relax\penalty 01636\mathchar 44\relax\penalty 01637\mathchar 44\relax\penalty 01639\mathchar 44\relax\penalty 01640\mathchar 44\relax\penalty 01642\mathchar 44\relax\penalty 01645\mathchar 44\relax\penalty 01646\mathchar 44\relax\penalty 01651\mathchar 44\relax\penalty 01654\mathchar 44\relax\penalty 01655\mathchar 44\relax\penalty 01657\mathchar 44\relax\penalty 01658\mathchar 44\relax\penalty 01660\mathchar 44\relax\penalty 01669\mathchar 44\relax\penalty 01672\mathchar 44\relax\penalty 01673\mathchar 44\relax\penalty 01678\mathchar 44\relax\penalty 01681\mathchar 44\relax\penalty 01682\mathchar 44\relax\penalty 01684\mathchar 44\relax\penalty 01685\mathchar 44\relax\penalty 01687\mathchar 44\relax\penalty 01690\mathchar 44\relax\penalty 01691\mathchar 44\relax\penalty 01693\mathchar 44\relax\penalty 01694\mathchar 44\relax\penalty 01696\mathchar 44\relax\penalty 01699\mathchar 44\relax\penalty 01700\mathchar 44\relax\penalty 01723\mathchar 44\relax\penalty 01726\mathchar 44\relax\penalty 01727\mathchar 44\relax\penalty 01732\mathchar 44\relax\penalty 01735\mathchar 44\relax\penalty 01736\mathchar 44\relax\penalty 01738\mathchar 44\relax\penalty 01739\mathchar 44\relax\penalty 01741\mathchar 44\relax\penalty 01750\mathchar 44\relax\penalty 01753\mathchar 44\relax\penalty 01754\mathchar 44\relax\penalty 01759\mathchar 44\relax\penalty 01762\mathchar 44\relax\penalty 01763\mathchar 44\relax\penalty 01765\mathchar 44\relax\penalty 01766\mathchar 44\relax\penalty 01768\mathchar 44\relax\penalty 01771\mathchar 44\relax\penalty 01772\mathchar 44\relax\penalty 01774\mathchar 44\relax\penalty 01775\mathchar 44\relax\penalty 01777\mathchar 44\relax\penalty 01780\mathchar 44\relax\penalty 01781\mathchar 44\relax\penalty 01790\mathchar 44\relax\penalty 01792\mathchar 44\relax\penalty 01793\mathchar 44\relax\penalty 01795\mathchar 44\relax\penalty 01798\mathchar 44\relax\penalty 01799\mathchar 44\relax\penalty 01801\mathchar 44\relax\penalty 01802\mathchar 44\relax\penalty 01804\mathchar 44\relax\penalty 01807\mathchar 44\relax\penalty 01808\mathchar 44\relax\penalty 01813\mathchar 44\relax\penalty 01816\mathchar 44\relax\penalty 01817\mathchar 44\relax\penalty 01819\mathchar 44\relax\penalty 01820\mathchar 44\relax\penalty 01822\mathchar 44\relax\penalty 01894\mathchar 44\relax\penalty 01897\mathchar 44\relax\penalty 01898\mathchar 44\relax\penalty 01900\mathchar 44\relax\penalty 01901\mathchar 44\relax\penalty 01903\mathchar 44\relax\penalty 01912\mathchar 44\relax\penalty 01915\mathchar 44\relax\penalty 01916\mathchar 44\relax\penalty 01921\mathchar 44\relax\penalty 01924\mathchar 44\relax\penalty 01925\mathchar 44\relax\penalty 01927\mathchar 44\relax\penalty 01928\mathchar 44\relax\penalty 01930\mathchar 44\relax\penalty 01934\mathchar 44\relax\penalty 01936\mathchar 44\relax\penalty 01937\mathchar 44\relax\penalty 01939\mathchar 44\relax\penalty 01942\mathchar 44\relax\penalty 01943\mathchar 44\relax\penalty 01969\mathchar 44\relax\penalty 01970\mathchar 44\relax\penalty 01975\mathchar 44\relax\penalty 01978\mathchar 44\relax\penalty 01979\mathchar 44\relax\penalty 01981\mathchar 44\relax\penalty 01982\mathchar 44\relax\penalty 01984\mathchar 44\relax\penalty 01993\mathchar 44\relax\penalty 01996\mathchar 44\relax\penalty 01997\mathchar 44\relax\penalty 02002\mathchar 44\relax\penalty 02005\mathchar 44\relax\penalty 02006\mathchar 44\relax\penalty 02008\mathchar 44\relax\penalty 02009\mathchar 44\relax\penalty 02011\mathchar 44\relax\penalty 02015\mathchar 44\relax\penalty 02017\mathchar 44\relax\penalty 02018\mathchar 44\relax\penalty 02020\mathchar 44\relax\penalty 02023\mathchar 44\relax\penalty 02024\mathchar 44\relax\penalty 02035\mathchar 44\relax\penalty 02036\mathchar 44\relax\penalty 02038\mathchar 44\relax\penalty 02042\mathchar 44\relax\penalty 02044\mathchar 44\relax\penalty 02045\mathchar 44\relax\penalty 02047\mathchar 44\relax\penalty 02050\mathchar 44\relax\penalty 02051\mathchar 44\relax\penalty 02056\mathchar 44\relax\penalty 02059\mathchar 44\relax\penalty 02060\mathchar 44\relax\penalty 02062\mathchar 44\relax\penalty 02063\mathchar 44\relax\penalty 02065\mathchar 44\relax\penalty 02092\mathchar 44\relax\penalty 02096\mathchar 44\relax\penalty 02098\mathchar 44\relax\penalty 02099\mathchar 44\relax\penalty 02101\mathchar 44\relax\penalty 02104\mathchar 44\relax\penalty 02105\mathchar 44\relax\penalty 02116\mathchar 44\relax\penalty 02117\mathchar 44\relax\penalty 02119\mathchar 44\relax\penalty 02123\mathchar 44\relax\penalty 02125\mathchar 44\relax\penalty 02126\mathchar 44\relax\penalty 02128\mathchar 44\relax\penalty 02131\mathchar 44\relax\penalty 02132\mathchar 44\relax\penalty 02137\mathchar 44\relax\penalty 02140\mathchar 44\relax\penalty 02141\mathchar 44\relax\penalty 02143\mathchar 44\relax\penalty 02144\mathchar 44\relax\penalty 02146\mathchar 44\relax\penalty 02158\mathchar 44\relax\penalty 02159\mathchar 44\relax\penalty 02164\mathchar 44\relax\penalty 02167\mathchar 44\relax\penalty 02168\mathchar 44\relax\penalty 02170\mathchar 44\relax\penalty 02171\mathchar 44\relax\penalty 02173\mathchar 44\relax\penalty 02177\mathchar 44\relax\penalty 02179\mathchar 44\relax\penalty 02180\mathchar 44\relax\penalty 02182\mathchar 44\relax\penalty 02185\mathchar 44\relax\penalty 02186\mathchar 44\relax\penalty 02461\mathchar 44\relax\penalty 02464\mathchar 44\relax\penalty 02465\mathchar 44\relax\penalty 02468\mathchar 44\relax\penalty 02470\mathchar 44\relax\penalty 02482\mathchar 44\relax\penalty 02483\mathchar 44\relax\penalty 02488\mathchar 44\relax\penalty 02491\mathchar 44\relax\penalty 02492\mathchar 44\relax\penalty 02495\mathchar 44\relax\penalty 02497\mathchar 44\relax\penalty 02501\mathchar 44\relax\penalty 02504\mathchar 44\relax\penalty 02506\mathchar 44\relax\penalty 02509\mathchar 44\relax\penalty 02510\mathchar 44\relax\penalty 02522\mathchar 44\relax\penalty 02524\mathchar 44\relax\penalty 02528\mathchar 44\relax\penalty 02531\mathchar 44\relax\penalty 02533\mathchar 44\relax\penalty 02536\mathchar 44\relax\penalty 02537\mathchar 44\relax\penalty 02542\mathchar 44\relax\penalty 02545\mathchar 44\relax\penalty 02546\mathchar 44\relax\penalty 02549\mathchar 44\relax\penalty 02551\mathchar 44\relax\penalty 02650\mathchar 44\relax\penalty 02653\mathchar 44\relax\penalty 02654\mathchar 44\relax\penalty 02657\mathchar 44\relax\penalty 02659\mathchar 44\relax\penalty 02666\mathchar 44\relax\penalty 02668\mathchar 44\relax\penalty 02671\mathchar 44\relax\penalty 02672\mathchar 44\relax\penalty 02707\mathchar 44\relax\penalty 02708\mathchar 44\relax\penalty 02711\mathchar 44\relax\penalty 02713\mathchar 44\relax\penalty 02725\mathchar 44\relax\penalty 02726\mathchar 44\relax\penalty 02731\mathchar 44\relax\penalty 02734\mathchar 44\relax\penalty 02735\mathchar 44\relax\penalty 02738\mathchar 44\relax\penalty 02740\mathchar 44\relax\penalty 02747\mathchar 44\relax\penalty 02749\mathchar 44\relax\penalty 02752\mathchar 44\relax\penalty 02753\mathchar 44\relax\penalty 02767\mathchar 44\relax\penalty 02774\mathchar 44\relax\penalty 02776\mathchar 44\relax\penalty 02779\mathchar 44\relax\penalty 02780\mathchar 44\relax\penalty 02785\mathchar 44\relax\penalty 02788\mathchar 44\relax\penalty 02789\mathchar 44\relax\penalty 02792\mathchar 44\relax\penalty 02794\mathchar 44\relax\penalty 02830\mathchar 44\relax\penalty 02833\mathchar 44\relax\penalty 02834\mathchar 44\relax\penalty 02848\mathchar 44\relax\penalty 02855\mathchar 44\relax\penalty 02857\mathchar 44\relax\penalty 02860\mathchar 44\relax\penalty 02861\mathchar 44\relax\penalty 02866\mathchar 44\relax\penalty 02869\mathchar 44\relax\penalty 02870\mathchar 44\relax\penalty 02873\mathchar 44\relax\penalty 02875\mathchar 44\relax\penalty 02888\mathchar 44\relax\penalty 02893\mathchar 44\relax\penalty 02896\mathchar 44\relax\penalty 02897\mathchar 44\relax\penalty 02900\mathchar 44\relax\penalty 02902\mathchar 44\relax\penalty 02909\mathchar 44\relax\penalty 02911\mathchar 44\relax\penalty 02914\mathchar 44\relax\penalty 02915\mathchar 44\relax\penalty 03028\mathchar 44\relax\penalty 03031\mathchar 44\relax\penalty 03032\mathchar 44\relax\penalty 03035\mathchar 44\relax\penalty 03037\mathchar 44\relax\penalty 03076\mathchar 44\relax\penalty 03077\mathchar 44\relax\penalty 03091\mathchar 44\relax\penalty 03098\mathchar 44\relax\penalty 03100\mathchar 44\relax\penalty 03103\mathchar 44\relax\penalty 03104\mathchar 44\relax\penalty 03112\mathchar 44\relax\penalty 03113\mathchar 44\relax\penalty 03116\mathchar 44\relax\penalty 03118\mathchar 44\relax\penalty 03139\mathchar 44\relax\penalty 03140\mathchar 44\relax\penalty 03143\mathchar 44\relax\penalty 03145\mathchar 44\relax\penalty 03152\mathchar 44\relax\penalty 03154\mathchar 44\relax\penalty 03157\mathchar 44\relax\penalty 03158\mathchar 44\relax\penalty 03199\mathchar 44\relax\penalty 03220\mathchar 44\relax\penalty 03221\mathchar 44\relax\penalty 03224\mathchar 44\relax\penalty 03226\mathchar 44\relax\penalty 03233\mathchar 44\relax\penalty 03235\mathchar 44\relax\penalty 03238\mathchar 44\relax\penalty 03239\mathchar 44\relax\penalty 03260\mathchar 44\relax\penalty 03262\mathchar 44\relax\penalty 03265\mathchar 44\relax\penalty 03266\mathchar 44\relax\penalty 03274\mathchar 44\relax\penalty 03275\mathchar 44\relax\penalty 03278\mathchar 44\relax\penalty 03280\mathchar 44\relax\penalty 04921\mathchar 44\relax\penalty 04927\mathchar 44\relax\penalty 04936\mathchar 44\relax\penalty 04939\mathchar 44\relax\penalty 04963\mathchar 44\relax\penalty 04966\mathchar 44\relax\penalty 04975\mathchar 44\relax\penalty 04981\mathchar 44\relax\penalty 05044\mathchar 44\relax\penalty 05047\mathchar 44\relax\penalty 05056\mathchar 44\relax\penalty 05062\mathchar 44\relax\penalty 05089\mathchar 44\relax\penalty 05098\mathchar 44\relax\penalty 05101\mathchar 44\relax\penalty 05299\mathchar 44\relax\penalty 05305\mathchar 44\relax\penalty 05332\mathchar 44\relax\penalty 05341\mathchar 44\relax\penalty 05344\mathchar 44\relax\penalty 05413\mathchar 44\relax\penalty 05422\mathchar 44\relax\penalty 05425\mathchar 44\relax\penalty 05452\mathchar 44\relax\penalty 05467\mathchar 44\relax\penalty 06151\mathchar 44\relax\penalty 06196\mathchar 44\relax\penalty 06439\mathchar 44\relax\penalty 06520\\$

A.4 Expansion Searches

—c—c—c—c—c—c—c—c—c—c—c—c— Expansion Searches for monomials $M(x)$, with coefficients over $\mathbb{F}_{3^{m}}$.
n $M$ Expansion Terms  
1   2   3   4   5  
m time m time m time m time m time

6 $x^{2}$ 6 0.1s 6 41m 6 535.5h 3 33.7h 2 411.1h
$x^{10}$ 6 1.8s 6 63.1h 3 33.1h 2 29.32h 1 2h
7 $x^{2}$ 7 10ms 7 8m - - - - - -
$x^{4}$ 7 5.6s 7 34.4h - - - - - -
$x^{10}$ 7 2.8m 7 455.7h - - - - - -
$x^{28}$ 7 4.2m 7 589.9 h - - - - - -
8 $x^{2}$ 8 19s 4 21.3m 4 48.4h 2 15.24h 1 3.2h
$x^{4}$ 8 90ms 8 29.6h 4 60.7h 2 45.7h 1 10.4h
$x^{10}$ 8 0.1s 8 5.4h 4 52.9h 2 48.5h 1 4.5h
$x^{28}$ 8 0.1s 8 352.4h 4 323.5h 2 55.6h 1 4.7h
$x^{82}$ 8 36s 8 42.4h 2 148.4h 1 131.7h 1 304.8h
9 $x^{2}$ 1 120ms 1 13.6h 1 42.9h 1 185.2h - -
$x^{4}$ 1 38s 1 24.3m 1 277.5h 1 85.5h - -
$x^{10}$ 1 92s 1 1.1h 1 $>$ 300h 1 205.1h - -
$x^{28}$ 1 8s 1 9.1m 1 87.91h 1 168.6h - -
$x^{82}$ 1 10m 1 1.7h 1 $>$ 300h 1 93.9h - -
$x^{244}$ 1 11m 1 1.9h 1 294.3h 1 89.9h - -