Design and simulation procedure 

Introduction

Method followed to create the simulation for this project is given in this chapter and in general there are different steps involved in the simulation process and all these steps are discussed with the corresponding screens in this chapter. As the main aim of this project is to estimate the packet losses across the MANETs under the guidance of different routing protocols three scenarios are created. AODV is used for routing across the first scenario; DSR is used for routing in the second scenario and OLSR is used as the routing protocol for third scenario. Required applications are added to the simulation generate the traffic and from this traffic the actual packet losses are estimated and against the routing protocols and the simulation procedure followed is as given below

 Simulation steps implemented in creating scenario one: AODV routing protocol 

AODV is used as the routing protocol in the first scenario and 20 mobile nodes are used in this simulation to generate the traffic. Performance of AODV is estimated in this simulation against the packet losses and for this purpose a simple MANET model is constructed with 20 mobile nodes and the wireless LAN server. The steps followed in creating the first scenario are explained in this section as below 

Steps in forming the basic network 

MANET is the basic network is used in this project and to estimate the performance of AODV against the packet losses and for this purpose few steps need to be followed in creating the basic network. As discussed 20 mobile nodes are used for the simulation and a single wireless LAN server is also used for setting the traffic on the network and following steps need to be adopted in creating the basic network 

  • Open the OPNET modeler simulation tool and select the new file option to create a new project
  • Set the project name and also the scenario name to create a new project and a new scenario
  • Chose the option create an empty scenario such that a blank scenario is created where the OPNET modeler can be used to load the default scenarios that were inbuilt in the network
  • Different options are available in choosing the type of topology and in this simulation a simple campus network is chosen to act as the required topology.
  • Now the size of the campus need to be set and in this simulation the size of the campus is set to 1500 X 1500 square meters
  • A new network model family is required and for this simulation purpose the network model family used is MANET and this can be chosen from the list of models and the option yes is chosen against the MANET network to set MANET as the basic network in this simulation. With these steps the required basic network is created and now the next step is to use different objects from the object palette and the actual objects used in this simulation process are given in the below points
  • 20 wireless LAN mobile workstations are used as the required mobile nodes and these nodes now act as the source for generating the traffic where the real packet analysis is done
  • A single fixed wireless LAN server is also used to act as the respective application destination and this server will server all the 20 mobile nodes
  • As the main aim of this project is to estimate the packet losses across the MANET, a source of traffic is required and in this context an application is required. In general an application configuration is used to create the application and this application will generate the traffic required for analyzing the packet losses.
  • A profile definition is also created for the application created and this profile is created by using the profile configuration node and it can be dragged from the object palette.
  • Setting up the mobility to the nodes on the simulation is the important step to be implemented and this mobility can be set using the mobile configuration object that was available with the object palette and with this process the complete setup of basic network is done and when these objects are created the first scenario will be created as shown below in the screenshot

From the above screen it is clear that a total of 20 mobile nodes ranging from mobile_node_0 to mobile_node_19 created and in this context a single wireless LAN server is also used to set the traffic definitions. Application configuration, mobile configuration and profile configuration are used set the required definitions respectively. Application configuration settings is the next step in the simulation and the process followed in this context is as given below 

Defining the application configuration settings 

An application is required all the time to set the desired levels of traffic on the network and in general the traffic sent or traffic received across the network mainly depends on the type of application chosen. In general there are different types of applications that can be used across the OPNET modeler simulation and in this simulation Email is used as the application to generate the required FTP traffic and this traffic is now used to analyze the packet sequences on the network. Following steps need to be implemented to analyze the application definition process and they are as listed below 

  • Right click on the application definitions object to edit the required attributes
  • A separate application definitions tab is available and in this section add the required number of rows such that each row represents a single application
  • In this simulation a single application is used and thus a single row is created
  • Name of the application is given as Email and in the description section email is selected for the application
  • Medium load is selected for the application and the corresponding screenshot is as given below

A medium load email application is used in this scenario such that FTP traffic is generated on the network and once the application is selected, OK button is clicked to apply the settings. With this step the required application definitions are done and the next step is to create the corresponding profile definitions and it is explained as below.

Defining the profile configuration settings 

Defining the profile configuration settings is the next step in the simulation and profile definitions are required to process the application requests from the mobile nodes and the server and also the traffic generated will be dependent on the profile definitions created. The steps need to be followed in creating the profile definitions is given below 

  • Profile configuration node is edited to define the profiles and this can be done using the right click option
  • Number of rows are added to the profile definitions and a single row is added in this context to support the single application created in the previous section
  • Few definitions of the profiles like start time and end time are also set and the corresponding screenshot is as shown below 

From the above screen it is clear that email application is used to set the email profile against the single row created for the profile definitions. Start time offset is set to constant value of 100 seconds and the duration of the profile is set to end of the profile. Once the profile definitions are done the actual node level and server settings need to be done and they are explained as below. 

 Node and server level settings done 

Node and server level settings are required to set and define the traffic and packet analysis against these nodes and the detailed configuration followed in this context is as explained in this section. As mentioned there are 20 mobile nodes and a single wireless LAN server, these nodes should be configured and set against the traffic analysis and packet analysis and first step in this context is that setting IP address to all the nodes and the corresponding procedure followed is explained as below 

  • Choose all the mobile nodes and the wireless LAN server
  • Go to Protocols menu -à ApplicationsàIP
  • From IP choose the option Auto assign IPV4 addresses and with this all the selected nodes and the wireless LAN server will be assigned with the corresponding IPV4 addresses and the respective screenshot is as shown below

Auto assign IPV4 addresses is the actual option chosen in this process and this IPV4 address are assigned to all the mobile nodes and the wireless LAN server as shown in the above screenshot. Next step in setting the configuration for mobile nodes and server is that to add the required routing protocol. As mentioned in the previous section, AODV is added as the desired routing protocol and this protocol is applied to all the mobile nodes and the wireless LAN server and the corresponding steps need to be followed are as given below 

  • Setting up the routing protocol option is available with the node level or server level attributes and for this purpose all the mobile nodes and the wireless LAN server are selected
  • Any one of the node or the server is selected and opened in the edit mode by right clicking on the node
  • In the edit mode the attribute like ad hoc routing protocol is available and this will display a list of routing protocols to the users and from this list AODV is used for this scenario and the corresponding screen is as shown below

AODV is selected from the list of routing protocols available as shown in the above screen and with this selection now all the nodes and the server will serve against the AODV routing protocol by checking the option apply to selected objects as shown in the above screen. When the routing protocol is ready to serve the traffic and packet patterns against the routing parameters, now these nodes and server should be set against the application and profile definitions and the corresponding procedure is as shown below. Application level settings to the nodes and the server can be done in either or two ways and the first procedure is given as below 

  • All the mobile nodes are selected from the simulation network and any one of the node is selected and edited to define the applications
  • At the node level attributes there is an option like Application destination preferences and now this tab is expanded to define the required application.
  • Number of rows is given as one as all the nodes need to support the single Email application created and the corresponding screen is as given below

Application destination preferences is set to one rows and then the Email application is selected for this row and the symbolic server name is given as email server such that this server will generate the required TCP traffic and the corresponding packets across the network.  Apart from the destination preferences now the application supported profiles are also set and in this context a single row is added to the section application supported profiles and Email profile is selected for this row as shown in the above screen. With this the node level application and profile definitions are applied to all the mobile nodes and the procedure followed to apply the same for wireless LAN server is given below

  • Select the wireless LAN server from the network and open it in the edit mode
  • There is an option like application supported profiles and expand this option to edit the profile settings for the wireless LAN server and the corresponding screen is as shown below

From this screen it can be observed that email profile is selected as the application supported profile and the corresponding description is set to supported as shown above and with this step now the wireless LAN server is also ready to support the discrete traffic generated from the Email profile and also the server can handle the packets and the corresponding traffic. Another method to apply the application and profile settings for the mobile nodes and the wireless LAN server and this process is known as deployment and the steps need to be implemented in this context are as given below 

  • Go to protocols menuàApplicationsàDeploy defined applications
  • Then a new window is displayed to the users and the deployment can be done at this level
  • All the mobile nodes are submitted to the source side and the wireless LAN server is submitted to the destination side and the corresponding screens are given below

Deploy defined applications is the actual option selected at this level and for this purpose the applications menu from the protocols menu is selected as shown in the above screen. A new window is opened to proceed with the application deployment and the corresponding screen is as shown below

From this screen it can be observed that all the mobile nodes are submitted for the source node of the Email profile and the server is submitted for the destination of the email profile and with this step the deployment is done and the respective consistency of the deployment can also be checked at this new window level. A success message like consistency check pass is displayed to the users if the deployment is done in a right manner. With this the corresponding email application and email profile are applied to the 20 mobile nodes and the single wireless LAN server. 

Setting the mobile definitions to the nodes 

As the main aim of this project is to evaluate the packet losses across the network, the nodes used in the simulation should be set against few mobility patterns and in general this mobility to the nodes can be added using the mobile configuration objects. Mobility will define the direction, source, destination and speed for the mobile nodes and thus always the mobility is required in this context. Following steps need to be followed to set the mobility to the mobile nodes 

  • Edit the mobile configuration object by right clicking on the object
  • There few options are available and from those options chose the option random mobility profiles
  • Now expand the option default random waypoint tab to set this as the required mobility model to the nodes and the corresponding screenshot is as given below

Random waypoint parameters are set in this process and from the above screen it is clear that the speed of the mobile nodes is set to a constant value of 50, pause time is set to a constant value of 0 seconds and start time of the mobile nodes is repeated for every 10 seconds. Stop time of the simulation is set to end of simulation as shown in the above screen. Once the mobility profile is ready this should be applied to all the mobile nodes and this can be done by choosing the option set mobility option from the topology menu and this screen is shown as below 

Set mobility profile is chosen from the random mobility menu from the topology menu as shown in the above screen. From this all the mobile nodes are assigned with the default random waypoint mobility and thus now the nodes can move randomly to process the traffic and the corresponding packets delivered to the destination side. With this setting up the mobility patterns to the network nodes is done and the next step in the simulation is to define the performance DES metrics and the process followed is given as below 

Setting up the DES metrics for performance evaluation 

Analysis the packet losses across the network are the main aim of this project and for this analysis the DES metrics should be chosen and applied for the scenarios. Following steps need to be followed in setting up the DES metrics 

  • Right click on the simulation environment and chose the option like Choose Individual DES statistics
  • With this step a new window is opened where three options like global, link and node level are available and they can be chosen to evaluate the performance of the network
  • Global level attributes are chosen in this simulation and the corresponding screens are as shown below 

From the above screen it is clear that Choose Individual DES statistics option is selected to the set the performance metrics and the corresponding screen displayed after this step is as shown below 

Email performance across the network against the traffic and packet sequences and following are the actual performance metrics used in this context and they are listed as below 

  • Download response time in seconds
  • Traffic received in bytes per sec
  • Traffic received in packets per sec
  • Traffic sent in bytes per sec
  • Traffic sent in packets per sec
  • Upload response time 

Apart from the Email, the wireless LAN metrics are also used and they are as listed below 

  • Data dropped due to buffer overflow
  • Data dropped due to retry threshold exceed limit
  • Delay
  • Load
  • Medium access delay
  • Network load
  • Retransmission attempts
  • Throughput 

Once the required performance metrics for Email application and wireless LAN server the simulation of first scenario is done and even for the second and third scenario similar metrics are used and the actual simulation methodology implemented in this context is as given below

Simulation of Second scenario: DSR 

Simulation of the second scenario is same as the first scenario and the routing protocol set for the mobile nodes and the server is changed. Following steps need to be followed to change the routing protocol in this scenario 

  • All the mobile nodes and the server is chosen from the simulation environment
  • Any one of the mobile node is selected and edited at the attributes level
  • Go to the ad hoc routing protocols menu and change the routing protocol to DSR, where initially there is AODV from the first scenario and the corresponding screen is as shown below 

From above screen it can be observed the AODV routing protocol is changed to DSR routing protocol and the third scenario is explained in the below section.

Simulation of third scenario: OLSR 

Second scenario is duplicated and the routing protocol is changed to OLSR across this scenario and the corresponding screenshot is as shown below 

OLSR is chosen as the required routing protocol for this scenario as shown in the above screen. With this step the simulation of three scenarios is done and next step is to run the simulations against the simulation time. All the three scenarios are run for 1 hour to achieve the results and they are explained in the next chapter.