![](/files/happy5.png)
Decomposing eulerian graphs
![$ G $](/files/tex/b8e7ad0330f925492bf468b5c379baec88cf1b3d.png)
![$ P $](/files/tex/b2b0b759db4d5a1b3204c38cdee6d9bd9e0d0dab.png)
![$ G $](/files/tex/b8e7ad0330f925492bf468b5c379baec88cf1b3d.png)
![$ (G,P) $](/files/tex/fd3ab98e5fb37dd2a932a24c976ad96933b74a0e.png)
Definition: Let be an Eulerian graph and for every vertex
, let
be a partition of the edges incident with
. We call
a transition system. If every member of
has size at most
(for every
), then we call
a
-transition sytem. A compatible decomposition of
is a list of edge-disjoint cycles
with union
so that every
contains at most one edge from every member of
.
Let be a graph and let
be the graph obtained from
by replacing each edge
of G by two edges
in parallel. Let
be the 2-transition system of
with
whenever
and
are incident with
. Now,
is an Eulerian graph and every compatible decomposition of
gives a cycle double cover of
. Since the cycle double cover conjecture can be reduced to graphs which are 3-edge-connected, the above conjecture would imply the cycle double cover conjecture.
We define a transition system to be admissable if every member of
contains no more than half of the edges in any edge-cut. It is easy to see that if there is a compatible decomposition of
, then
must be admissable. The converse of this is not true; There is an admissable 2-transition system of the graph
which does not admit a compatible decomposition. Recently, G. Fan and C.Q. Zhang [FZ] have proved that
does have a compatible decomposition whenever
is admissable and
has no
minor. This result imporoved upon an earlier theorem of Fleischner and Frank [FF]. Very recently, I have proved a weak version of the above conjecture, by showing that
also has a compatible decomposition when P is a 2-transition system and G is 80-edge-connected. I'd quite like to see an improvement on this bound. Here is a related conjecture.
![$ W $](/files/tex/48948390b5dc5ab9a52c0afaff3e950050be14a2.png)
![$ G $](/files/tex/b8e7ad0330f925492bf468b5c379baec88cf1b3d.png)
![$ G $](/files/tex/b8e7ad0330f925492bf468b5c379baec88cf1b3d.png)
![$ G $](/files/tex/b8e7ad0330f925492bf468b5c379baec88cf1b3d.png)
![$ F $](/files/tex/bfff269cc7df9bdb7c57d8b6a2a74020d114f24d.png)
![$ W $](/files/tex/48948390b5dc5ab9a52c0afaff3e950050be14a2.png)
![$ F $](/files/tex/bfff269cc7df9bdb7c57d8b6a2a74020d114f24d.png)
If is given by
then we may form a 2-transition system
by putting
in
for every
(working modulo
). Now a compatible decomposition of
is precisely a cycle decomposition of
satisfying the above conjecture. Thus, Sabidussi's conjecture is equivalent to the assertion that
has a compatible decomposition whenever
has no vertex of degree two and
is a 2-transition system which comes from an Euler tour.
Let be a directed Eulerian graph and for every vertex
, let
be a partition of the edges incident with
into pairs so that every in-edge is paired with an out-edge. We define a compatible decomposition to be a decomposition of
into directed circuits so that every directed circuit contains at most one edge from every member of
. Our current techniques don't seem to shed any light on the problem of finding compatible decompositions for Eulerian digraphs. Next I pose a very basic question which is still open.
![$ k $](/files/tex/c450c3185f7285cfa0b88d3a903c54f7df601201.png)
![$ (G,P) $](/files/tex/fd3ab98e5fb37dd2a932a24c976ad96933b74a0e.png)
![$ G $](/files/tex/b8e7ad0330f925492bf468b5c379baec88cf1b3d.png)
![$ k $](/files/tex/c450c3185f7285cfa0b88d3a903c54f7df601201.png)
![$ P $](/files/tex/b2b0b759db4d5a1b3204c38cdee6d9bd9e0d0dab.png)