Mohammad Mirhakkak

Dynamic quality-of-service for mobile ad hoc networks

This paper presents an approach to supporting quality-of-service (QoS) in a dynamic network environment. With this approach, resource reservations represent ranges, and applications adapt to an allocated level of QoS provided by the network at some point within the requested range. To explore this approach, we have implemented a new protocol called dynamic RSVP (dRSVP), which is an extension to RSVP.

Dynamic bandwidth management and adaptive applications for a variable bandwidth wireless environment

This article describes an approach for providing dynamic quality of service (QoS) support in a variable bandwidth network, which may include wireless links and mobile nodes. The dynamic QoS approach centers on the notion of providing QoS support at some point within a range requested by applications. To utilize dynamic QoS, applications must be capable of adapting to the level of QoS provided by the network, which may vary during the course of a connection. To demonstrate and evaluate the dynamic QoS concept, we have implemented a new protocol called dynamic resource reservation protocol (dRSVP) and a new QoS application program interface (API). The paper describes this new protocol and API and also discusses our experience with adaptive streaming video and audio applications that work with the new protocol in a testbed network, including wireless local area network connectivity and wireless link connectivity emulated over the wired Ethernet. Qualitative and quantitative assessments of the dynamic RSVP protocol are provided.

Modeling and Simulation of Haipe

The high assurance Internet protocol encryptor (HAIPE) is a critical component of the global information grid (GIG) that will enable secure communications to support information assurance (IA) in the evolving network-centric architecture of the Department of Defense (DoD) and Intelligence Community networks. We developed a HAIPE model, including a HAIPE discovery approach proposed by the GIG routing working group (GRWG), to assess the impact of HAIPE overhead on the performance and scalability of the GIG. This paper describes our HAIPE model, presents the operation of the HAIPE-related protocols including HAIPE discovery, and defines future work

Verification and validation of the QualNet JTRS WNW and SRW waveform models

OTC, together with the JTRS JPEO and PEO-I MSO, funded a Small Business Innovative Research (SBIR) Phase 2.5 Technology Transition effort, with SNT as the contractor, to develop the JTRS Network Emulator (JNE) to support GMR LUT and MOTE test activities. The accuracy of the JNE, to a large extent, depends upon the accuracy of its constituent models. Consequently, OTC funded an effort to verify and validate the performance of the QualNet WNW and SRW waveform models. The purpose of this paper is to report on the status and results generated to date for the QualNet WNW and SRW V&V activities that are scheduled through the end of Fiscal Year 2012.

A Performance Evaluation of Transport Mechanisms in Hybrid Networks

This paper evaluates the performance of several alternative reliable unicast transport mechanisms in a hybrid network. Options investigated include end-to-end TCP (different flavors), end-to-end space communications protocol standards-transport protocol (SCPS-TP), and performance enhancing proxies (PEPs) (also called transport layer proxies). Our approach is to analyze these options in a specific scenario using modeling and simulation (M&S). We describe this scenario and the corresponding OPNET network model, our experiment plan, and the results obtained. Finally, we identify several areas for further analyses

Optimizing Route Formation Algorithm to Reduce Simulation Run-Time for Large Tactical Networks

Mobile ad-hoc networks (MANETs) are a major component of all DoD tactical networks. MANETs must be carefully analyzed to ensure integrated DoD networks perform properly on an end-to-end basis. Simulation is generally used to evaluate these networks before their actual deployment. We have identified the routing table calculation to be a major contributor to the excessive run-time and have employed an algorithm based on the breadth-first search to improve the simulation run-time. Our results show that employing the new algorithm will result in a lower run-time for the simulation

Modeling and Simulation of GIG Networking Scenarios Using Efficient HAIPE Discovery

In an earlier paper, we described a proof of concept and a simulation model for HAIPE peer discovery using the Border Gateway Protocol as a transport mechanism for exchanging mapping information between peering HAIPEs. In this paper we are expanding our HAIPE discovery approach to support some DoD specific requirements, such as network mobility, and to reduce its technology-insertion cost. The new approach conforms to the standard HAIPE v3.1 client specification. The purpose of this paper is to discuss our HAIPE peer discovery approach, its operation, and the OPNET model that we used to demonstrate its correct operation. In addition, this paper reports on some results that can be used to quantify the impact of HAIPEs and to allow conducting the Analysis of Alternatives for various GIG design approaches.

Addressing Runtime Performance Issues in Ad-Hoc Network Simulations

Quantifying the end-to-end performance of evolving DOD communication networks is highly desired by the component and network designers during all phases of the development process. Analytical techniques, in-lab testing and field demonstrations are all necessary toward this end but all have limitations in addressing this need leaving simulation as a primary method with which to generate end-to-end performance. However, simulation often results in unacceptably long runtimes for these types of networks. The purpose of this paper is to report on the results generated in the second year of a 2-year IR&D program to investigate methods to improve simulation runtime performance of mobile ad-hoc communication networks

Quantifying network performance of Mobile Ad-hoc Networks

The rapid deployment requirements, limited infrastructure, and mobile nature of tactical edge networks have led the Department of Defense (DoD) to investigate and implement mobile ad-hoc networks (MANETs) to support its mission needs. MANETs rely on spectrum as the transmission medium, and their performance depends heavily on the electromagnetic environment (EME) where they operate. Traditional methods of assessing MANET performance have been focused on link capacity and network throughput, without adequately accounting for the effects of the EME. The joint spectrum center of the defense spectrum organization (DSO/JSC) has developed the spectrum simulation testbed to adequately account for spectrum impacts on MANET performance. As part of the DSO/JSC spectrum simulation testbed development, a number of capability gaps were identified, specifically in areas of quantifying the relationship between spectrum requirements and MANET performance. The purpose of this paper is to report the results of a survey to identify the current capabilities to address MANET performance within the context of accounting for available spectrum and to describe two capabilities that were developed to help bridge the analysis gap in the area relating spectrum requirements to system performance predictions.

An Approach for Real-Time Monitoring and Control of Tactical Network Simulations

With the conventional modeling and simulation (M&S) of networks, simulations are usually executed from beginning to the end without interruption, and results are available only at the end of the simulation runs. These results typically consist of various statistics for some measures of performance (MOPs) that attempt to quantify the networks performance. Although these results can capture the timing of significant events and performance transitions, they generally capture neither their causes nor the transient behavior that leads to their occurrence. This paper presents an approach to operate a simulated network similar to an actual network where the simulated network can be monitored in real-time and better, and, if appropriate, be stopped for explorations of the network state. A console equipped with a GUI interrogates the simulated network during the run and displays the relevant characteristics of the network, which allows experimenters to monitor the networks behavior and performance. Experimenters can use the console to stop and query the simulation to obtain detailed information from each node or protocol layer within each node. Examples of such information are route forwarding tables at a node, node position, or some protocol attributes configured at a node. Further, experimenters can use the console to change network configuration during the simulation run and observe its impact on the performance of the network. Our solution is built using OPNET Modeler as the simulation engine and its cosimulation facility to interconnect the experimenters console with the simulation.