Teaching.
OPNET is being/will be used in the following courses:

1. CS 5283. Computer Network Programming.
2. CS 5813. Principles of Wireless Networks.

Research.
OPNET is also being used in the following research projects:

1. Clustering in Ad-hoc Networks

   Over the last few years, there have been a number of applications of mobile ad hoc networks where movement profiles of mobile nodes is not completely random. In this project we develop the Predictive Clustering Algorithm which is a location-aware mobility prediction based framework for organizing the nodes of a mobile ad hoc network into clusters. Our approach utilizes a mobility prediction algorithm to identify nodes which show similarity in movement patterns for some amount of time and group them into a cluster. Such an approach results in long-lived clusters, while simultaneously lowering the overhead associated with cluster maintenance. We investigate the behavior of the proposed clustering scheme through OPNET simulation experiments, we model the movement of the mobile nodes using Gauss-Markov, Random Waypoint, and Reference Point Group mobility models. We analyze the survival times of the clusters obtained and compare them against existing clustering algorithms.
   

2. Topology aggregation and QoS Routing

   QoS routing is the process of routing a connection request based on the connection's resource requirements. One of the key issues in deploying QoS based routing in large networks is to reduce the overhead involved in QoS routing protocols. Topology aggregation is an important technique that reduces the overhead involved in QoS routing by orders of magnitude and makes it scalable. However, the scalability comes at the expense of accuracy as the aggregated network state information is an approximation of the actual network state. This approximation adversely affects the routing decisions and hence the performance of the routing protocol. Using network flow techniques, we propose models that effectively capture the network state in terms of resource availability. Using OPNET, we propose to study the performance of our resource availability models with respect to various traffic parameters on varied topologies.
   

3. Energy Efficient Routing in Delay sensitive Sensor Networks

   Sensor Networks are special purpose ad-hoc networks that have severe constraints on energy resources. Hence, developing routing techniques that minimize energy consumption in these networks is of utmost importance. In this project, we investigate various topology control techniques that work in conjunction with customized routing techniques in order to achieve energy efficent routing for delay sensitive applications such as target tracking. Using OPNET, we aim to experimentally evaulate the effectiveness of the proposed topology control/routing schemes in saving energy.
   

4. Network models for performance analysis

   Current network state models assume that the call arrival pattern in a network is Poisson in nature. However, recent trace studies have shown that, the call arrivals are bursty in nature, and not Poisson. Hence, the current network models do not capture the realistic network behavior. We propose to develop mathematical models that effectively capture the steady-state behavior of multi-class networks with bursty call arrivals. Using OPNET, we propose to study the steady-state behavior of multi-class networks, and characterize the effectiveness of our mathematical models.