1. Introduction
Fibonacci sequence
defined by
, and
, where
, and Lucas sequence
defined by
, and
, where
, in view of their connections with the golden ratio, are in the center of interest of many researchers and mathematics enthusiasts. These sequences have many interesting interpretations, applications and generalizations. Among numerous generalizations of Fibonacci and Lucas numbers there are generalizations in the distance sens, such as, for example, generalized Fibonacci numbers
introduced by M. Kwaśnik and I. Włoch [
1] given by the formula
, for
and
, with initial conditions
for
and generalized Lucas numbers defined by A. Włoch [
2] as follows:
for
, with initial values
for
. It is worth mentioning that generalizations in the distance sense are usually related to different graph parameters. Applications of Fibonacci-like numbers in graphs was initiated by the paper of H. Prodinger and R. F. Tichy [
3], in which the relationship between Fibonacci numbers and independent sets (i.e., subsets of vertices of a graph being pairwise nonadjacent) was described. Independent sets, and consequently Fibonacci-like numbers play an important role in chemical combinatorics and many types of localization problems. Fibonacci and Lucas sequences, like other recursively defined sequences, are naturally generalized to polynomials. Fibonacci polynomials
are given by the recurrence relation
, for
with initial conditions
, Lucas polynomials are defined by the recursion
, for
, with initial values
. Obviously
, and
. Significant contributions to investigation on properties of Fibonacci and Lucas polynomials have been made by V. E. Hoggatt Jr. and M. Bicknell [
4,
5,
6,
7]. A few newer results on the classical Fibonacci and Lucas polynomials and their applications can be found in [
8,
9,
10]. It is worth noting that Fibonacci and Lucas polynomials are used for determining approximate solutions of many types of integral equations such as for example Cauchy integral equations, Abel integral equations, Volterra-Fredholm integral equations and others (for details see [
11,
12,
13,
14]). Fibonacci numbers and polynomials by their connections with diophantine equations and Hilbert’s 10th problem are also closely related with the so-called Pell surfaces studied by the Shaw prize winner J. Kollar [
15] due to an important algorithmic embeddability problem for algebraic varieties [
16].
The interest in Fibonacci-like polynomials has contributed to the emergence of many generalizations. Most of them are obtained by changing initial terms while preserving the recurrence relation (see References [
17,
18]) or by slight modifying the basic recursion (see References [
19,
20,
21]). Some are obtained in the distance sense i.e., by changing distance between terms of a sequence. In the paper [
22] we have introduced the distance Fibonacci polynomials
given by the following recurrence relation
with initial conditions
for
for integers
,
.
In
Table 1 we present some distance Fibonacci polynomials
for special values of
k and
n.
We have found a direct formula, a generating function, matrix generators and some identities for generalized Fibonacci polynomials
. We have also extended the distance Fibonacci polynomials
to negative integers
n, namely
with initial conditions
for
.
In
Table 2 we present the first few elements of
polynomials for special
k and negative
n.
In this paper, which is a continuation of [
22], based on a graph interpretation of the distance Fibonacci polynomials
we introduce a new generalization of Lucas polynomials in the distance sense. We derive a direct formula, a generating function and matrix generators for these polynomials. We also prove some identities that generalize the classical identities for Lucas polynomials and reveal some Pascal-like relations between coefficients of these polynomials.
2. From the Distance Fibonacci to the Distance Lucas Polynomials
In the paper [
22] we have used a special kind of covering of a graph to obtain a graph interpretation of the distance Fibonacci polynomials
. Let us recall the idea of that covering. Let
G be an undirected, finite graph with the vertex set
and the edge set
,
be an
n-vetrex path (a sequence of distinct vertices
such that
for
) and
C be a set of
x colors, where
. We cover the vertex set
by the subgraphs
and
, with the vertex of a graph
additionally colored with one of
x colors from the set
C. This operation is called
-covering with
-coloring. By
we denote the number of all
-covering with
-coloring of the graph
G.
To illustrate
-covering with
-coloring of a graph
G in
Figure 1 we present
-covering with
-coloring of a graph
(we have surrounded paths
and
by dashed lines, the letter
x below a path
denotes a number of colors we can use for coloring of this path). One can easy check that
.
For a graph we have proved the following theorem.
Theorem 1 ([
22]).
Let be integers. Then Now let us consider -covering with -coloring of an n-vertex cycle (a closed path), where .
Theorem 2. Let be integers. Then
Proof. (by induction on n). Let be integers and be a cycle with the vertex set .
If , for , then we cover the vertices only by subgraphs with coloring by one of x colors. Hence for . If and , then we can cover the vertices of a cycle by k subgraphs which are colored with one of x colors or we can cover such a graph by one path , which can be chosen on k ways. Hence .
Assume that , for , and the theorem is valid for all integers less then n. We will prove that it is true for n. We have to consider two possibilities:
Then a vertex can be colored by one of x colors. By let us denote the number of all -covering with -coloring of a graph with belonging to . Thus, .
- 2.
.
Let denote the number of all -covering with -coloring of a graph with belonging to . Since we have k such paths, hence .
Taking into account both cases, Theorem 1 and induction hypothesis we obtain
Thus, the theorem is proved. □
As a consequence of Theorem 2 we obtain a new graph interpretation of the classical Lucas polynomials .
Corollary 1. Let be integers. Then .
Proof. Let us consider-covering with -coloring of an n-vertex cycle . By Theorem 2 we have . Since then we have the equality and by the well-known identity we obtain . □
Let us denote and call this parameter as the distance Lucas polynomial.
Theorem 3. Let be integers. Then
Proof. We prove this theorem by induction on n. Let be integers.
For
and
, we can easily check that
. Namely, using Theorem 2 and
Table 1 and
Table 2, we get that
. In turn, on the right side we get
.
If and , then we obtain and on the other hand . Thus, for the theorem is true.
Assume that the theorem is valid for all integers
n. We will prove that it is also true for
, i.e.,
is satisfied. Using Theorem 2, recurrence (
1) and induction hypothesis we obtain
Thus, the theorem is proved. □
Based on Theorem 3 we can define the distance Lucas polynomials as follows.
Let
,
be integers. The distance Lucas polynomials
are given by the recurrence relation
with initial conditions
for
Table 3 presents some distance Lucas polynomials
for special values of
k and
n.
It is obvious that for
we have
, and therefore
. For
and
we get
, where
is the
nth generalized Lucas number defined in [
2]. Moreover, by recursion (
3), one can easy check that
By graph interpretation of the distance Lucas polynomials almost immediately follows a direct formula for these polynomials.
Theorem 4. Let be integers. Then For
by the formula (
5) and relation
we get
which is the well-known direct formula for Lucas polynomials.
Using steps method described in [
22] we can observe some Pascal-like relations between coefficients of the distance Lucas polynomials. For a fixed
and each
let us arrange coefficients of the distance Lucas polynomials
in ascending order and form a left-shifted array of these coefficients. Building steps of height
we can see that the sum of elements on steps beginning in a row corresponding to
, where
, is equal to
. Moreover, adding two consecutive elements on steps i.e., an element in the
nth row and the
jth column and an element in the
st row and the
st column we obtain an element in the
st row and the
st column. In
Table 4 and
Table 5 we present cases
and
, respectively.
In
Table 4 we have marked the steps of height 2 starting in rows
(rows are counted from
), adding the elements (red numbers) on the steps we obtain sums
in turn. Analogously, in
Table 5 we have marked the steps of height 3 starting in rows 4 and 8, appropriate sums on the steps are
and
, respectively. We have marked in blue the rule of generating elements in
Table 4 and
Table 5.
3. Generating Function, Extension for Negative Integers and Some Identities
Using the standard method we can derive a generating function for the distance Lucas polynomials .
Theorem 5. Let be integers. The generating function of the distance Lucas polynomials sequence is given by the formula
Proof. Let
. By recurrence relation (
3) we get
Thus,
which ends the proof. □
Note that for by Theorem 5 and the fact that we obtain a function
being a generating function for the classical Lucas polynomials .
The distance Lucas polynomials
can be extended to negative integers
n. Let
,
be integers. Then
with initial conditions
for
.
Table 6 includes the first few elements of
polynomials for special
k and negative
n.
Notice that setting
in (6), we get the well-known extension of Lucas polynomials for negative numbers
For and we obtain the extension of classical Lucas numbers for negative numbers.
Theorem 6. Let be integers. Then
for
for
.
Proof. At the beginning we prove the identity
. Using the recurrence relation (
3) we obtain
Hence, for integers
, we get
Adding these equalities we have
Thus the identity is proved.
Now we prove the identity by induction on n. If , then we have . Hence, the identity is true for . Assume that and the equality is true for an arbitrary n. We will prove that it holds for .
By induction hypothesis and the recurrence relation (
3) we have
Thus the identity is proved.
Analogously we can prove the identity .
To prove the identity
we use the definition of distance Lucas polynomials (
3) by
times. Then we obtain
Hence the identity holds.
Using the recurrence relation (
3) once again we can prove the identity
. Let
be integers. Then
Thus the identity is proved.
The last identity follows from the recursion of the Theorem 2 and the definition of the Fibonacci polynomials.
Thus the theorem is proved. □
Note that for we obtain the identities for Lucas polynomials and for , we obtain well-known identities for Lucas numbers. Moreover, for and , we obtain some new identities for generalized Lucas numbers .
Proving analogously as Theorem 6, we get the following identities for polynomials for negative integers.
Theorem 7. Let be integers. Then
for .