Hi Catherine,

There is an interesting paper on Energy Efficiency for AODVv2 (DYMO) [1st link].

In the [2nd link] some slides on the implementation on ns2. Could give more help if you are using WSNET. Good luck !

http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=6583447&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D6583447

http://www.slideshare.net/shaymol/simulation-and-performance-analysis-of-aodv-using-ns-2-sam

You need to modify the Routing Agent of AODV protocol which is responsible for selecting next node while forwarding packets. Routing Agent is defined in routing class as a function in the implementation of AODV. You should add another function in the routing class and call the function in the existing Routing Agent function to integrate your suggested approach in AODV next-node selection technique.

I have worked on NetSim (http://www.tetcos.com/) and can see that using two API's you can get the nodes power and position. 

Question is what would you like do with the power & position ?

http://www.tetcos.com/

I have review several articles that propose routing methods based in the residual energy in wifi, and every one of them had the same problem, from the point of view of the engineering, they don’t offer any advantage.

The problem is that in wifi the radio must be always on, and the consumption in reception and transmission is similar, you can avoid the nodes with low energy, but this doesn’t imply that life of the nodes will be extended. The bigger consumption in the radio component is due to the polarization of the electronic, and the electronic must be polarized if you want receive something. It is more, we have measure devices in which the peak of consumption was in idle reception, because in these moment the amplifiers were at maximum. The only form that you really can reduce the consumption is switch of the radio, extract the nodes with low residual energy from the route don’t extend significantly the lifetime of the nodes. Using simulation it is possible to cheat if the reception consumption is significantly lower that the energy in transmission, but with real devices, the consumption is very similar.

Hi

What is your problem? haven't any idea or modifying in AODV's implement in ns2?

Hi,

check this Tutorial: http://arxiv.org/pdf/1007.4065.pdf

I think you can just add a new (class/function) of energy in the AODV agent responsible for selecting the next hop and implement your approach....