|
|
|
|
OPNET
Technologies
OPNET is a
registered |
|
University: Technical University of Catalonia Department:
Computer
Architecture Research using OPNET:
Most of the current topological based routing algorithms used in MANETs treat all the nodes in the network like independent peers, making them not scalable respect to the number of nodes in the network. Since hierarchical routing techniques have been known to afford scalability in large networks (e.g. Internet), and taking advantage of the existence of scenarios in which the nodes of a MANET can be aggregated in a natural manner, we have proposed a two level hierarchical routing scheme for MANETs. However, a trade-off between the # of nodes in the network and the complexity of the system is needed. The main objective of this research is to validate our proposed routing scheme and find out the aforementioned trade off. To reach this objective OPNET Modeler is a relevant tool in our work. Since the MANETs are multi-hop networks, the routing has become one of the main research issues in this area. A number of routing proposals have been developed which can be generally classified in either topology-based routing, or Position-based routing approaches. The latter determines relative or absolute positions (e.g. GPS) of the nodes in the network. On the other hand, the former approach determines the routes according to topological information to achieve connectivity. Two of the main protocols obtained in the routing area are the well-known AODV [2] (reactive) and the OLSR [3] (proactive) topology based protocols. We have performed a comparison study between OLSR and AODV to determine which protocol reacts better with respect to the increment of the number of nodes (scalability). For this study we performed simulations with OPNET models and simulator. The results and conclusions of this study will be available soon in a technical report. A logical solution to improve the scalability of the routing in ad-hoc networks is “to keep, at any node, complete routing information about nodes which are close to it (in terms of hop distance or some other nearness parameter), and lesser information about nodes located further away” [6]. This is the main idea behind the hierarchical routing schemes, like the used in Internet. This can be realized aggregating nodes into sets (subnets) to be handle as a single routing entity. It means that any node will maintain one entry per destination for the closer nodes (other members of its subnet), one entry per subnet for the remote nodes. An example of this kind of efforts is the work presented in [5], which shows improvements in the scalability when address aggregation is used in a static Ad-hoc network with AODV protocol. There are some scenarios in MANETs with heterogeneous environments where some nodes where some nodes have extra capabilities (e.g. processing, transmission range, power supply) or responsibilities (e.g. speed of movement, position in the network, etc), in which the nodes can be grouped following a logical or geographical constraints. Under this kind of scenarios our proposal becomes applicable. Some examples of this kind of scenarios can be found in Emergency Networks [4], and Wireless Mesh Networks WMN [9, 10]. Taking advantage of these scenarios we proposed a two level hierarchical routing scheme for MANETs in [7]. Figure 1 shows an example of the scenario for which our proposal is directed.
Fig. 1. Example of a MANET with nodes grouped in subnets. The “L” denotes the leader nodes (or cluster-heads). The adoption of nodes aggregation in MANETs under such environments envelops the solution of the following challenges: Grouping the nodes (clustering), address allocation and registration, neighborhood and neighboring subnets discovery, the routing into the subnet and inter-subnets, the mobility between subnets (node movement and node location) and finally how to manage the established connections. We would like to emphasize that the majority of the aforementioned challenges (except the inter-subnet routing) are not exclusive of our proposal. They are open issues in wireless ad-hoc communications, and therefore many solutions can be found in the literature. We have chosen the solutions which better adapt to the mentioned environments, or take some basic ideas behind them to build our own solutions. We have proposed 3 functional blocks, which provide mechanisms that solves completely or partially the challenges presented above. These blocks are: the Subnet Formation, the Inter and Intra-subnet routing, and finally the Mobility between subnets. The subnet formation block [8] has two mechanisms, one to group the nodes (forming and removing subnets), and another for address allocation. Additionally, these mechanisms interact with the other blocks in order to optimize our proposal. The Inter and Intra-subnet routing blocks is composed of two mechanisms, one to obtain information about how to get destinations into the same subnet, and another one to get information for reaching other subnets. This block also collaborates with the block that manages the Mobility. The Mobility between Subnets block manages the relative topology location of the nodes, and it deals with the ongoing sessions in case of subnet switching. We performed an analytical evaluation of the entire proposal. Actually, we are developing the OPNET models for each one of the 3 functional blocks in which we have divided our proposal. During the last year we developed some preliminary models to evaluate the Subnet Formation proposal. These models were available in the Contributed Model Depot. We have debugged and updated them recently. The results obtained with these models will be published soon. The job for the next 6 months of license renewal involves: the completion of the models of Inter and Intra-subnet routing functional blocks, performing a performance evaluation of these functional blocks and starting the integration of these models (also integration with OPNET models). References
The CompNet Group (http://www.ac.upc.es/CompNet) used the OPNET modeler for performance evaluation of the Loosely Coupled architecture. Since the Loosely coupled architecture is based on Mobile IP as the mobility management protocol over the WLAN and GPRS networks, and the inter-working between the two networks is achieved by the GGSN (GPRS Gateway Support Node). The model integrates Mobile IP to the existing OPNET GPRS and WLAN models. The main results and the work description of the aforementioned activities are resumed in the following technical reports: “Handoff Latency Performance for the Loosely Coupled GPRS-WLAN Architecture”. Johann Lopez, Jose M. Barcelo, Nik Van den Wijngaert, Christ Blondia. DAC Technical Report UPC-DAC-2004-4. "Impact of the Vertical Handoff on the Transport Layer for the Loosely Coupled GPRS-WLAN Interworking Architecture". Johann Lopez, Jose M. Barcelo, Nik Van den Wijngaert. UPC-DAC-2004-10. |
|
|
|
|
