Generating a Binary Matrix in C

  • Thread starter Thread starter gradnu
  • Start date Start date
  • Tags Tags
    Binary Matrix
Click For Summary
SUMMARY

This discussion focuses on generating a binary matrix in C, specifically an adjacency matrix representing a directed graph with a fixed number of 1s and 0s. Participants suggest using random number generators, such as the C standard library function rand(), and emphasize the importance of ensuring the generated network remains connected. The conversation also touches on the distinction between incidence and adjacency matrices, highlighting the complexity of maintaining connectivity in the generated network.

PREREQUISITES
  • C programming language proficiency
  • Understanding of random number generation techniques
  • Basic knowledge of graph theory concepts
  • Familiarity with matrix representation of graphs
NEXT STEPS
  • Research C standard library functions for random number generation
  • Learn about algorithms for generating connected graphs
  • Study the clique problem and its implications in graph theory
  • Explore methods for validating graph connectivity in generated matrices
USEFUL FOR

Software developers, particularly those working with graph algorithms, data structures, and network simulations, will benefit from this discussion.

gradnu
Messages
21
Reaction score
0
Does anybody know how to generate a random matrix with entries 0 and 1 only(binary matrix). Numbers of 1 and 0 are fixed every time matrix is generated.
Possibly a program in 'C'.

Thanks
 
Technology news on Phys.org
Numbers are not truly random so perhaps you would like to look at random number generators. You could take your seed and square it, then take out the bits. and the square is your new seed. Or think of another mathematical formula for generating random numbers.http://www.cprogramming.com/tutorial/random.html
The number returned by function rand is dependent on the initial value, called a seed that remains the same for each run of a program. This means that the sequence of random numbers that is generated by the program will be exactly the same on each run of the program.

You can change the seed to be the same for the matricies you wish to have the same 1/0.
 
is the size of the matrix random as well?
 
No, size of matrix is finite.
Basically the problem is to generate random networks with fixed number of nodes and fixed number of links. Network is being represented by a binary matrix where the entry '1' means that a link exists between the corresponding row and column and '0' means the link does not exist.
I have to make sure that in generating new network(or matrix) the network should remain connected. It shouldn't break into two or more smaller networks.
I haven't done graph theory so if somebody can help me with some kind of algorithm, it would be great.

Thanks,
 
Are you generating a incidence matrix (matrix is indexed by nodes and vertices; element ij is one if node i is an endpoint of vertex j) or an adjacency matrix (element ij is one if some vertex connects node i to node j)? A network is a directed graph, so the adjacency matrix is asymmetric.

Determining whether a random NxN unitary matrix with M ones represents a digraph is the NP-complete problem called the clique problem.
 
I want to generate random adjacency matrices where element ij is 1 if an edge connects node i to node j. Yes it is asymmetric.
 
gradnu said:
I want to generate random adjacency matrices where element ij is 1 if an edge connects node i to node j. Yes it is asymmetric.

..and you wish to do so with a fixed number of "1"s?

This might not be the most efficient:

initialize A=0

For k ones,
do this k times
{
select randomly i from 1 to n, and j from 1 to n
if A[i,j]=1, go back and get another pair (without changing k)
else set A[i,j]=1
}

If necessary, discard proposed links to oneself (i.e. reject the pair if i=j).
 
Last edited:
Nodes in a network can link to themselves. For example, in finite state machine, it is often best to explicitly represent the time steps during which the machine remains in one state as a transition from that state to that same state.

One problem with this approach is that it most likely will not generate a "network". You could wrap this approach in an outer loop that tests whether the adjacency network represents a single graph. If it doesn't, try again. Determining whether this is the case might take a lot of CPU time (its NP-complete).
 

Similar threads

  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 57 ·
2
Replies
57
Views
6K
  • · Replies 14 ·
Replies
14
Views
2K
  • · Replies 25 ·
Replies
25
Views
3K
Replies
5
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 29 ·
Replies
29
Views
3K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K