Escolar Documentos
Profissional Documentos
Cultura Documentos
Abstract—In this work, we derive an algebraic formulation imum degree of only 2. Moreover, the form of the equations
for the scalar linear network coding problem as an alternative to has additional structure that can be exploited in several cases.
the one presented by Koetter et al in [1]. Using an equivalence We illustrate this simplification through examples and compare
between network information flow and group-valued circulations,
we derive a system of polynomial equations that algebraically rep- our formulation with the one proposed in [1] in terms of the
resents the scalar linear network coding problem. Surprisingly, number of variables and equations involved.
this system of polynomials has a maximum degree of 2. We An alternative way of viewing our formulation is that we
illustrate our formulation and its advantages through example perform a graphical simplification of the formulation in [1],
networks drawn from the literature. which uses scaling variables on every link. The crux of our
simplification lies in a graph transformation that migrates all
I. I NTRODUCTION
scaling variables to the sources. All the intermediate nodes
The idea of network coding over error-free networks, pio- in the transformed graph simply perform addition and have
neered in [2], has been a subject of active current research. only a single outgoing link resulting in a set of directed trees.
The general idea of linear network coding, where intermediate Finally, we arrive at a set of only linear and degree-2 equations,
nodes linearly combine incoming packets, was explored in relating the scaling variables at the sources, that algebraically
[3]. A simple and effective algebraic formulation of the represents the given network coding problem.
general network coding problem was introduced in [1]. This
established a direct connection between a network information II. T HE N ETWORK C ODING P ROBLEM
flow problem and an algebraic variety over the closure of a The communication network is modelled as a directed,
finite field. acyclic multigraph, G = (V, E), where the node set V
While the multicast network coding problem has been char- represents the terminals and switches in the network and the
acterized almost completely [4], the general network coding edge set E represents the communication links. It is assumed
problem still remains much harder to characterize [5]. The that all communication links are error-free and have unit
algebraic formulation of [1], while being simple and powerful, capacity. For a given edge e = (u, v), we denote u = tail(e)
results in equations that are not readily amenable to easy and v = head(e). For each node v ∈ V , we define
solution in many cases. In this paper, our main result is to I(v) = {e ∈ E : head(e) = v}, the set of input links of
derive an alternative algebraic formulation for the general v and O(v) = {e ∈ E : tail(e) = v}, the set of output
scalar linear network coding problem. Specifically, we show links of v. Let us further assume the following without loss
a correspondence between linear network-coded information of generality: (1) A node v is a source node iff |I(v)| = 0
flow in a given network and group-valued circulations in and all source nodes produce exactly one unit of data per
an equivalent set of directed trees. In a network, a group- unit time. (2) A node v is a sink node iff |O(v)| = 0 and
valued circulation is a mapping from the set of edges to a all sink nodes demand exactly one unit of data per unit time.
group such that the net flow through any node is conserved In cases where a node v produces (demands) more than one
under the group addition operation [6]. One can readily see data symbol, we can add virtual source (sink) nodes that
that linear network-coded information flow should be closely produce (demand) exactly one data symbol, have exactly one
related to group-valued circulations. However, since there is no output (input) link connecting them to v and no input (output)
multiplication operation in groups, we show that the network links. Then, the set of source and sink nodes is defined as
needs to be transformed before a direct relationship can be S = {v ∈ V : |I(v)| = 0} = {s1 , s2 , . . . , s|S| } and
established between linear network coding over a field and T = {v ∈ V : |O(v)| = 0} = {t1 , t2 , . . . , t|T | }.
a group valued circulation. We develop an algorithm for this Let us now assume that the symbols transmitted over the
network transformation, which results in a set of trees. network are elements of a finite alphabet H. For each edge e,
We then use the corresponding flow in the set of trees an edge function is then defined as a mapping fe : H i → H
to derive a system of polynomial equations that provides an where, i = 1 if tail(e) ∈ S and i = |I(tail(e))| otherwise.
algebraic formulation for the network coding problem in the Definition 1: The collection of all the edge functions in a
original network. Surprisingly, this set of equations has a max- given network is defined as a network code. If all the edge
functions are linear maps with respect to a field alphabet H,
in [1] gives the following 8 equations in 10 variables:
then the code is a scalar linear code. α3 + α4 α1 = 1; α4 α2 = 0; α5 + α6 α1 = 0; α6 α2 = 1
Let the data symbol generated at the i-th source node, α7 α2 + α8 = 0; α7 α1 = 1; α9 α2 + α10 = 1; α9 α1 = 0
si ∈ S, be denoted by Xi and the data symbol demanded
by the j-th sink node, tj ∈ T , be denoted by Zj . These also In contrast, our formulation arrives at just 1 equation, a4 b3 =
implicitly define a set of connection requirements, denoted 1, in 2 variables as given in (6).
by C, for the given network G. Given a network G, the set IV. N ETWORK I NFORMATION F LOW AS G ROUP - VALUED
of source nodes S, the set of sink nodes T and the set of C IRCULATIONS
connection requirements C, the network coding problem is to
determine all the edge functions such that all the connection A. Group-valued Circulations
requirements are satisfied. If such a set of edge functions Consider a directed multigraph G = (V, E) and a finite
exists, then the network coding problem is solvable. If a set abelian group H. Then, a group-valued circulation or an H-
of linear edge functions, with respect to a finite alphabet H, circulation is defined as a mapping f : E → H such that the
exists that satisfies all the connection requirements, then the following equation is satisfied at each node v ∈ V :
network coding problem is scalar-linearly solvable. X X
In a scalar linear network coded flow (over a field H), the f (e) = f (e) (1)
P|S|
edge function of an edge e can be written as i=1 ai Xi , where e:head(e)=v e:tail(e)=v
P|S|
ai , Xi ∈ H. We refer to i=1 ai Xi as either the edge function This can be understood as a conservation of flow – the sum
of e or the symbol flowing through e and denote it as a vector of the symbols entering a node is equal to the sum of the
fe = [a1 a2 · · · a|S| ]. symbols leaving it, with addition over H. This is shown in
Fig. 2b.
III. M OTIVATION In a linear network coding scheme, each output link of a
The scalar linear network coding problem was formulated node carries a linear combination of the symbols received by
as a system of polynomial equations in [1]. Our aim in this that node through all its input links. Let us assume a coding
work has been to arrive at a simpler algebraic formulation for scheme over a field F . Let xe ∈ F represent the symbol
the general scalar-linear network coding problem than the one flowing through edge e. Then, the general form of the relation
described in [1]. This advantage of the algebraic formulation between the symbols flowing through one of the output links,
that we will describe in the next few sections can be easily e ∈ O(v), and the set of input links, I(v), of a node v ∈ V
noticed when we compare the two formulations for the case can be written as:
X
of the modified butterfly network shown in Fig. 1 with two xe = ae0 ,e xe0 (2)
sources and four sinks. Note that this network is identical to e0 ∈I(v)
the classic butterfly network under our definition of sources
and sinks. where the coefficients ae0 ,e are elements of F . This is shown
in Fig. 2a. Note that the linear network-coded flow is not
The edge functions under the direct assignment of scaling
immediately a group-valued circulation. The conservation laws
factors (as in [1]) is shown in Fig. 1. The formulation described
of Fig. 2a and 2b are not the same. In this section, we develop a
X1 X2 transformation of the network, which results in an equivalence.
1 2 B. Equivalence for one node
X1 X2
Consider the node shown in Fig. 2a. There are a total of
|O(v)| linear equations that need to be satisfied at this node.
3 Hence, an equivalent graph with group-valued circulations
X1 X2
α1 X1 + α2 X2
must have |O(v)| copies of this node, each satisfying one of
these equations. If ei denotes the i-th output link of the node,
the equation to be satisfied at the i-th copy of the node in the
4
equivalent graph is given by:
α1 X1 + α2 X2 α1 X1 + α2 X2 X
x ei = ae0 ,ei xe0 (3)
e0 ∈I(v)
5 α6 (α1 X1 + α2 X2 )
6 α9 (α1 X1 + α2 X2 )
+α5 X1 +α10 X2 Hence, the node shown in Fig. 2a will have to be replicated
α3 X1
+α4 (α1 X1 + α2 X2 )
α8 X2
+α7 (α1 X1 + α2 X2 )
as many times as the number of its output links along with
7 8 9 10 its entire set of input links. In addition, the flow in the
X1 X2 X1 X2 input links will have to be appropriately scaled. After these
transformations, the linear-network coded flow in one node
Fig. 1. Flow in the butterfly network. shown in Fig. 2a is equivalent to the group-valued circulation
in the graph shown in Fig. 2c.
xe01 n xe01
P xe 1
xe 1 = ae0i ,e1 xe0i
i=1
(a) (b)
n
P
xe0n xe m = ae0i ,em xe0i xe0n m
P n
P xe m
i=1 xe i = xe0j
i=1 j=1
Fig. 2. (a)Picture of a node employing a linear network coding scheme. (b) Picture of a node for a group-valued circulation. (c) Transformation of node
shown in (a).
Fig. 3. (a) The (modified) butterfly network with 4 sinks and 2 sources. (b) The final transformed network in which group-valued circulations correspond to
information flow in the original network. The transformed network has 4 disjoint trees - one for each each sink node. The appropriate scaling factors at the
leaf nodes are denoted by ai ’s and bi ’s
α4 (α1 X1 + α2 X2 ) α6 (α1 X1 + α2 X2 )
α4 (α1 X1 + α2 X2 ) α6 (α1 X1 + α2 X2 )
+α3 X1 +α5 X1
5 5
7 8
X1 X2
(a) 7 (b) 8
X1 X2
α3 X1 α4 α1 X1 α4 α2 X2 α5 X1 α6 α1 X1 α6 α2 X2
1 1 2 1 1 2 α3 X1 α5 X1
1 1
α4 (α1 X1 + α2 X2 ) α6 (α1 X1 + α2 X2 )
3 3
4 4 4 4
5 5 5 5
7 (d) 8 7 (c) 8
X2 X2
X1 X1
1 2 3 1 1 2 2 3 2 3 1 2 1 2 3 2 3
4 5 4 1 2
e1 e4 e1 4
e4 5 e1 e1 4 4 5
6 7 6 e4 5 3 e1 e4
6 6 6
7 7 7
8 9 8 9
e2 e3 e2 8 9 e3
e2 e3
12 10 14 11 13 12 10 11 13
10 14 11 (d)
(a) (b) (c)
Fig. 5. (a) An example network that is solvable only over fields with characteristic 2. There are three sources - 1, 2 and 3 - producing symbols X1 , X2 and
X3 respectively. There are three sinks - 12, 13 and 14 - demanding symbols X3 , X1 and X2 respectively. (b),(c),(d) The final transformed network with 3
trees - one for each sink node.
TABLE I
From equations b3 c4 + c2 = 0 and b3 c4 = c2 , we can derive C OMPARISON OF FORMULATIONS
the relation 2c2 = 0. Substituting c2 = 0 in the above system
leads to the condition 1 = 0 which is not possible. Hence, we
Group-valued Circulations Direct Formulation
must have 2 = 0, which implies that the system is not solvable Example Var.1 Deg. 2 Eqns1 Var. Eqns Deg.
in any field with an odd characteristic. Also, in characteristic
2, setting all variables to 1 in the above equations, is seen to Butterfly 4 6 10 8 2
Fig. 5a 8 15 14 9 3
be a solution. In contrast, the direct forumation of [1] results [1, Fig. 5] 9 5 14 4 4
in a system of 17 equations in 22 variables given in [5]. [9, Fig. 3] 27 45 50 32 3
[5, Fig. 3] 12 30 22 17 3
1 After one iteration of elimination using the linear equations
48 85
50
9
8 35 36
7 66
60 51
59 14
39 64
54
61 22
53
38 52
55
21
40 69
37 27
77 23 3
80 18
28 70
41
2
79 81 1
19
5
75 4
62 63
6 29
57 20 24
26
49 65 13
44
76 25
67 74
12
43
17
42
30
68 16
11
47
15 10 46
45
84 56
31
78
32
33 34 72
58 87
71
73
82
83
Fig. 6. An ISP network over Europe with 87 nodes and 161 edges.
with different number of sources and sinks on this network. variables. Our formulation initially gives 88 linear and
Further, we assumed characteristic 2 in our simplification 11198 degree-2 equations in 632 variables. But on
steps. The five nodes 31, 41, 47, 69 and 82 were set as sources applying the simplification steps, assuming characteristic
in all the following cases. Each sink demands data from one 2, it turns out that the system is not solvable over
of the source nodes. The sinks and their demands were chosen characteristic 2.
at random depending on graph connectivity. Hence, we see that the algebraic formulation of scalar linear
1) 5 sources (all rate 1) and 10 sinks. The direct for- network coding based on group valued circulations appears to
mulation (from [1]) gives a system of 44 equations work better even over large networks with a few sources and
in 30 variables. Our formulation initially results in sinks.
44 linear equations and 3 degree-2 equations in 316
VI. N ETWORK C ODE FROM G ROUP - VALUED
variables. After applying the simplification steps, we
C IRCULATIONS
are left with only 3 degree-2 equations in 7 variables
assuming solution exists in a characteristic 2 field. In We now describe an algorithm to obtain a network code for
fact, setting all the remaining 7 variables to zero results the original network from the group-valued circulations on the
in a valid solution to the three equations (some other transformed network. Note that this completes the proof of the
scaling variables are non-zero). Hence, a solution over sufficiency of edge compatibility conditions. We will briefly
GF(2) is possible. describe the algorithm by example. A formal description with
2) 5 sources (one with rate 2, others rate 1) and 12 sinks. notation can be found in [8].
The direct formulation yields a system of 50 equations in We compute two vectors for every edge e of the graph
40 variables in this case. In comparison, our formulation G = (V, E). The first vector fe that represents the edge
P|S|
initially resulted in 50 linear equations and 34 degree- function or symbol i=1 fe (i)Xi sent over edge e is given
2 equations in 330 variables. But after applying the by [fe (1) fe (2) · · · fe (|S|)]. Suppose e is replicated n times
simplification steps, we are left with only 13 degree-2 to obtain edges e00j , 1 ≤ j ≤ n in the transformed graph
equations in 17 variables assuming solution exists in a G0 = (V 0 , E 0 ). The second vector ce = [c1 c2 · · · cn ] is such
P|S|
characteristic 2 field. Again the all-zero solution is valid that the edge function on e00j ∈ E 0 is cj i=1 fe (i)Xi . Note
for the remaining 17 variables resulting in a network that such a scaling property is guaranteed for all copies of an
code over GF(2). edge by the compatibility conditions. Once the vectors fe are
3) 5 sources (all with rate 2) and 11 sinks. The direct computed for all e ∈ E, the network code in G is completely
formulation yields a system of 88 equations in 180 known.
ce1 = [a1 b1 c1 ] ce2 = [a2 b2 c2 ] ce3 = [a3 b3 c3 ]
A1 A2 A3 a1 A1 a2 A2 a3 A3 b1 A1 b2 A2 b3 A3 c1 A1 c2 A2 c3 A3
a1 A1 + a2 A2 c1 A1 + c2 A2 a1 A1 + a2 A2 b1 A1 + b2 A2 + b3 A3 c1 A1 + c2 A2
+a3 A3 +c3 A3 +a3 A3 = k(a1 A1 + a2 A2 + a3 A3 ) +c3 A3
Suppose fe0 and ce0 are known for all the incoming edges equivalent group-valued circulation network and to arrive at
e0 ∈ I(v) for a node v ∈ V . The vectors fe and ce can be a system of polynomial equations (of maximum degree 2)
computed for the outgoing edges e ∈ O(v) as illustrated for that represents the scalar linear network coding problem. We
a sample case in Fig. 7. have demonstrated the computational advantages of our new
In the figure, a node v ∈ V with I(v) = {e1 , e2 , e3 } algebraic formulation over the traditional approach.
and O(v) = {e4 , e5 } is replicated thrice into v(1), v(2) and The obvious intuitive connection between network coding
v(3) in G0 . The incoming and outgoing links are replicated and group-valued circulations is quite interesting in itself and
as shown. For instance, the edge e1 is replicated thrice as lends a new perspective to the network coding problem. We
e1 (1), e1 (2) and e1 (3). Suppose there are three source nodes have explored one aspect of this connection by arriving at
S = {s1 , s2 , s3 }, P
and fei = [αi1 αi2 αi3 ] resulting in edge an alternative algebraic formulation for the problem. Other
3
functions Ai = j=1 αij Xj for i = 1, 2, 3. The scaling aspects and applications of this connection are yet to be
vectors cei are as shown in the figure. explored.
Using the edge functions and scaling factors on the incom- Since our system of equations is equivalent to the one in
ing edges, the edge function of the copies of ei , i = 1, 2, 3 [1], an alternative way of viewing our formulation is that we
are computed first. For instance, the edge function of e2 (2) is simplify the equations in [1] through a recursive graphical
computed as b2 A2 . Then, the edge function for the outgoing method.
links of v(1), v(2) and v(3) in G0 are computed by simple
R EFERENCES
addition. As shown in the figure, the symbols sent on e4 (1)
and e4 (2) will be scalar multiples. We may then assign [1] R. Koetter and M. Médard, “An algebraic approach to network coding”,
IEEE/ACM Trans. Networking, vol. 11, pp. 782–794, October 2003.
the Psymbol on e4Pin G P to be the symbol on e4 (1) given [2] R. Ahlswede, N. Cai, S.-Y. R. Li, and R. W. Yeung, “Network
3 3 3
by i=1 ai Ai = j=1 ( i=1 ai αij )Xj (assumed nonzero). information flow”, IEEE Trans. Inform. Theory, vol. 46, pp. 1204–1216,
Then, fe4 and ce4 are assigned suitably. Also, the network July 2000.
[3] S.-Y. R. Li, R. W. Yeung, and N. Cai, “Linear network coding”, IEEE
coding coefficients at v for the output link e4 , denoted by Trans. Inform. Theory, vol. 49, pp. 371, February 2003.
αei ,e4 , i = 1, 2, 3, is given by cei (1), i = 1, 2, 3, the first [4] S. Jaggi, P. Sanders, P. A. Chou, M. Effros, S. Egner, K. Jain, and
element of the corresponding vectors (since the symbol on L. Tolhuizen, “Polynomial time algorithms for multicast network code
construction”, IEEE Trans. Inform. Theory, vol. 51, pp. 1973–1982,
e4 (1) was selected to be the symbol that would flow through June 2005.
e4 in G). [5] R. Dougherty, C. Freiling, and K. Zeger, “Linear network codes and
In this manner, all the nodes are processed in a suitable order systems of polynomial equations”, IEEE Trans. Inform. Theory, vol. 54,
pp. 2303–2316, May 2008.
to compute the network code for the original graph from the [6] R. Diestel, Graph Theory (Graduate Texts in Mathematics), Third
group-valued circulation on the transformed graph. Edition, Springer-Verlag, 2000.
[7] Narsingh Deo, Graph Theory with Application to Engineering and
VII. C ONCLUSION Computer Science, Prentice-Hall, Englewood Cliffs, NJ, 1974.
[8] Abhay Subramanian and Andrew Thangaraj, “A simple alge-
In this work, we have established an equivalence between braic formulation for the scalar linear network coding problem”,
network information flows and group-valued circulations and http://arxiv.org/abs/0807.1158.
used the equivalence to arrive at an algebraic formulation [9] R. Dougherty, C. Freiling, and K. Zeger, “Insufficiency of linear coding
in network information flow”, IEEE Trans. Inform. Theory, vol. 51, pp.
for the network coding problem that is different from the 2745–2759, August 2005.
one first proposed by Koetter et al in [1]. Given a network [10] T. Anderson, R. Mahajan, N. Spring, and D. Wetherall, “Rocketfuel
coding problem, we have given algorithms to construct an project”, http://www.cs.washington.edu/research/networking/rocketfuel.