英语论文网

ed based on received HELLO messages again triggering a recalculation of the MPR set. Finally the MPR selector set is updated according to information received in HELLO messages. Received TC messages triggers updates in the topology set while the MID set is updated upon receiving MID messages. All received messages will also be registered in the duplicate set if not already registered.

When generating HELLO messages, the link set, neighbor set and MPR set is queried. When generating TC messages, the MPR selector set is queried. When forwarding control traffic, the MPR selector set and the duplicate set is used.

Finally, route calculation is based on information retrieved from the neighbor set, the 2 hop neighbor set, the TC set and the MID set.

Chapter 4 OLSR Implementation

With the aim of evaluating the cost-benefit of OLSR Protocol, simulation work was done using the NS-2 network simulator along with the OLSR implementation provided by the Hipercom project, which is called OOLSR. The only modifications made to the all-in-one (NS-2 ver. 2.27 plus OOLSR ver. 0.99.15) source code available for download were: adding packet delay measurement and, a few data outputs to generate the required data files for analysis, therefore, experimentation can be easily repeated. The simulation work was performed following the next steps. A rigorous analysis of Per Packet Delay was performed. Each experimental stage is described in the following sections.

4.1 Scenario without Data Traffic:

As a first stage, simulation was performed over static networks without sending data traffic between nodes. The objective of this stage was to achieve basic understanding on the impact of the proposed strategies. Graphical and numerical analysis was performed. The simulation parameters are listed in Table 4-1.

The metrics that were utilized to measure the performance of the protocol are as follow:

i. TC messages: This metric counts the number of generated TC messages only, it does not count the retransmissions.

ii. TC messages overhead: This metric counts the total amount of bytes composing all the generated TC messages.

iii. Percentage of known links: This metric counts the percentage of known links by each node, over the total amount of existing links. It is averaged over all the nodes in the network.

iv. Percentage of MPRs: This metric counts the number of nodes in the network that have been selected, by any other node, as an MPR.

4.2 Scenarios with Data Traffic:

In a second stage, data traffic was added to the simple scenarios. The objectives this time were to measure the data delivery rate and the impact of the data traffic over the achieved network topology knowledge. The simulation parameters are listed in Table 4-2.

Table 4-2: Simulation parameters for scenarios with data traffic

The same metrics than the ones for scenarios without data traffic were used plus the data delivery rate, which measures the rate and number of data packets that are properly received at the destination node.

4.3 Statistical Analysis:

Several metrics were applied in order to evaluate the performance of the protocol. Most of these metrics are averaged values over a set of simulation scenarios. The PPD metric is a metric th本论文由英语论文网提供整理，提供论文代写，英语论文代写，代写论文，代写英语论文，代写留学生论文，代写英文论文，留学生论文代写相关核心关键词搜索。