You can use the software "Network" (http://www.fluxus-engineering.com/sharenet.htm ) as referenced by Artur. However, Network is designed for non-recombing molecules such as plastid DNA sequence data. You can code microsatellite data to be binary, but that is artificial when dealing with heterzygotes.
I would recommend using EDENetworks (http://becs.aalto.fi/edenetworks/). I generates networks based on various measures of genetic distance at the individual or population level and easily accepts microsatellite
An additional option is that you can use SplitsTree (http://www.splitstree.org). With splits tree you would need to generate a matrix of pairwise genetic distances using a basic genetic program and then input that as your data file.
Can I make similar map that I have attached here using EDENetwork?I tried with my data according to 2.1.6 given in the manual. But it showing x,y data is missing
You can use EdeNetworks to map the location of your samples. They way that you do this is that you include an Auxiliary data file that includes GPS coordinates. The algorithm will not incorporate the coordinates into the analysis, but it will plot each point at the location you have provided in the auxiliary file. This is how the image you attached was generated. These can be very informative if you have clear genetic/geographic structure, but the images can be really messy if you don't. In the auxiliary file you you can also code what color and size you want individual points to be. What I have done in the past is generate a network with GPS coordinates displayed and then I have over-layed that on a map of my choosing.