OPNET Technologies
7255 Woodmont Avenue
Bethesda, MD 20814
Tel: 240-497-3000
Fax: 240-497-3001
E-mail: university@opnet.com
Web: http://www.opnet.com/

OPNET Technologies, Inc.
OPNET and OPNET Technologies Inc.
are registered trademarks of OPNET.
All trademarks are the property of their
respective owners and are used herein
for identification purposes only.

University: The University of Pisa, Italy
Name of sponsoring Professor: Stefano Giordano
Department: Department of Information Engineering, Telecommunication Networks Group (TLCNETGRP)

Research:

1. Capacity Evaluation of Resource Allocation Strategies in the GPRS System

A fundamental issue in GSM/GPRS networks planning is the choice of an appropriate resource allocation strategy at MAC layer to allow data and voice users to share the common transport media. Our goal is to quantitatively evaluate the convenience, in terms of efficient radio resources utilization, of reserving a few timeslots in each TDMA frame to GPRS traffic rather than adopting a "complete-sharing" strategy. Consequently, a single-carrier GPRS system scenario has been accurately modeled under realistic working conditions, including physical layer errors, RLC blocks retransmissions and MAC allocation procedures.

2. Shared Risk Link Groups for Optimal Lightpath Restoration in OPNET™

Selectable strategies for lightpath restoration are basic requirement for the near future Optical Transport Networks (OTN) with a GMPLS control plane. Routing of lightpaths between GMPLS equipment benefits of new specific information, as the Shared Risk Link Group (SRLG). SRLGs refer to the physical topology of the network and they explicit the correlation between the failures of disjoint links due to the same risk event.A 14-nodes network topology has been set up in OPNET™ to analyze the effects of SRLGs handling in case of two possible restoration strategies for the lightpaths: the pre-planned local repair and the on-the-fly end-to-end.

3. An adaptive credit based flow control mechanism

A credit based flow control algorithm was applied within a hardware multiprocessor platform in order to allow co-existence of traffic with Real Time (RT) characteristics (e.g. voice and video) and Best Effort (BE) traffic. It adaptively reacts against congestion situations limiting throughput of BE sources when Round Trip Time (RTT) for RT traffic grows above a maximum threshold. The credit control mechanism works between ingress point and egress point of a flow crossing a generic network. At system start up, a credit pool is assigned to each flow. This credit amount is managed by the ingress point. Each flow starts to transmit at its maximum allowed rate, consuming credits at the same rate. Credits can be replenished only by credit return packets sent back from egress point. A credit return packet is generated each time egress point crosses a minimum threshold of bits forwarded downstream.

4. Evaluation of scheduling and AQM techniques in a DiffServ environment

Principal aim of this work is to simulate a real diffserv environment and to achieve fair buffer resource allocation through the combination of two different kind of  tecniques : scheduling and discarding. We tried to carry out a real environment and so we used TCP/IP client  and server (taken from OPNET libraries) to simulate several kind of traffic such as : file transfer , videoconferencing , mail and so on. The scheduling discipline used in this work is a WF2Q, a GPS approximating discipline with high fairness properties and relatively low implementation complexity (the scheduling algorithm was built from scratch and implemented in a process model). The scheduler grants access to the link to each of the queues wich group several different clients in 3 separate class (expedited forwarding , assured forwarding and best effort forwarding). The Active Queue management is a congestion avoidance mechanism wich takes advantage of TCP's congestion control mechanism (by randomly dropping packet prior to periods of high congestion the discarding tecniques tell the packet source to decrease its transmission rate). The discarding tecniques used in this work are RED (Random Early Detection) WRED (Weighted RED)and GRED (Generalized RED); all of the mentioned AQM disciplines were implemented . We studied different aggregation strategies of traffic and their impact on the scenario in words of queing delay and throughput, moreover we are tried to study is if there is a relationship beetweeen the treshold defined in discarding disciplines and the traffic mix wich our network is going to substain.

5. Protection and restoration on optical networks using MultiProtocol Lambda Switching

We are working on protection and restoration techniques in optical networks using Multiprotocol Lambda Switching as the forwarding scheme. MPLS provides some highly efficient mechanisms to perform protection and fast re-routing upon a link failure. Our work aims to study the effects of those schemes on an optical domain. Optical Cross Connects will be the core elements of the network. A physical optical network will be simulated by connecting some OXC models; a virtual topology will be constructed by OxC configuration and Label Switched Paths will be deployed according to Constraint Based Routing rules; protection paths will be set up. Failures will be simulated and network response will be observed. New models will be built to simulate simple Optical Cross Connects behavior and schedule failure. Performance will be evaluated in terms of packet loss and restoration time.

6. Resource channel allocation scheme in a GPRS system

Simulative study of the performance of a mobile network supporting both GSM and GPRS services. Construction of a simulation model using OPNET V.6, through which the quality of service can be estimated. The simulation model consist of a scenario containing one SGSN, one BSC, one BTS, GSM mobiles and GPRS mobiles that behave independently of each other. The BTS contains a number of TRX (4 or 6), 200 kHz carriers, and the channels are shared between GSM and GPRS. Three of the available channels are dedicated to signalling purposes and the leftover resources consist of common channels that can be configured dynamically either to GSM or to GPRS according to the concept of "capacity on demand". The traffic model is of decisive importance. The GSM mobiles generate call requests for circuit switched calls as in ordinary telephony by a Poisson process. This is no longer true for data transmission services and especially for packet data services. In these simulations we have used three traffic models: one to model e-mail traffic, one to model WWW services and one to model WAP services. The simulation scenario is downlink: voice + e-mail + WWW + WAP.

7. Impact of various error models in wireless channels on TCP performance

The purpose of this work is to evaluate the impact of various error models on TCP performance.
TCP is a protocol developed for wired transmission and it expects a channel with a low error rate. When a high error rate occurs (typical of wireless connections) TCP activates anti-congestion procedures as Slow Start and Congestion Avoidance; in that way throughput in wireless channel falls down. OPNET was used to simulate a FTP transfer between a server and a workstation connected by a link that destroy packets with a high rate. There were three channels with different error models:
1) a channel without memory with a Poissonian loss process
2) a channel with a two-state Markovian loss process
3) a channel with a two-state Pseudo-Paretian loss process.
Two state channels (case 2 and 3) have a state with an high loss probability (“bad state”) and a state with a low loss probability (“good state”); global loss probability of the three channels is the same. OPNET simulations demonstrate that performance of models with memory and without memory are  strongly dependent by channel delay; there are two relevant cases:

• Little delay (it means very lower than average time spent in bad state): two state channel shows a throughput lower than one-state channel’s one.
• Great delay (it means similar to average time spent in bad state): two state channel shows a throughput higher than one-state channel’s one.

8. IEEE 802.11b Performance Evaluation: Convergence of Theoretical, Simulation and Experimental Results

The paper presents the analysis of the evaluation of the maximum goodput achievable in a IEEE 802.11b Basic Set Service (BSS) performed by means of theoretical, simulative and experimental approaches. With respect to previous works on this topic, the presented study focuses on the comparison of the results obtained using the different approaches, highlighting their convergence in the simple case of ideal radio channel. The assumption of an ideal channel, although not always verified in real scenarios, is useful to evaluate the upper bound of the network capacity offered by this technology in absence of losses and performance degradations due to factors not directly related to the protocol. Furthermore, the adopted simulation tool has been preventively validated versus theoretical results under varying working conditions and for different physical layers. It is noteworthy that the results obtained by each one of the three approaches are in accordance with those already presented in literature, although they are derived by means of different tools, thus further confirming the accuracy of such results.

9. Admission Region for Multimedia Services in IEEE 802.11e Systems

The paper presents the simulation analysis aimed at evaluating the admission region of a IEEE 802.11e draft 4.0 standard, adopting the EDCF (Enhanced Distributed Coordination Function) access control mechanism. In particular, the study gives an estimate of the number of QoS aware services, namely videoconference and Voice over IP (VoIP), that can be admitted to the transport service offered by the EDCF, while their QoS requirements are satisfied. In the study, the ideal radio channel model is assumed. Furthermore, the traffic sources are obtained from traffic data acquired by means of measurement campaigns made during emulation of VoIP and videoconference services, based respectively on G.723.1 and H.261 codec. Finally the QoS parameters experienced by a mix of VoIP, videoconference and TCP traffic transmitted according to the EDCF mechanism are compared with those obtained when the DCF mechanism is adopted. The result of this comparison clearly highlights the efficiency in the traffic differentiation of the EDCF mechanism.

10. Configuration of WiMAX Networks supporting Data and VoIP traffic

Due to its large coverage area, low cost of deployment and high speed data rates, WiMAX is a promising technology for providing wireless last-mile connectivity. Physical and MAC layer of this technology refer to the IEEE 802.16e standard, which defines 5 different data delivery service classes that can be used in order to satisfy Quality of Service (QoS) requirements of different applications, such as VoIP, videoconference, FTP, Web, etc. The main aim of the paper is to examine a case of QoS deployment over a cellular WiMAX network. In particular, the paper compares the performance obtained using two different QoS configurations differing from the delivery service class used to transport VoIP traffic, i.e. UGS or ertPS. Results indicate that for delay-sensitive traffic that fluctuates beyond its nominal rate, having the possibility to give back some of its reserved bandwidth, ertPS has the advantage to permit the transmission of BE traffic.

11. Experimental and Simulation Study of a WiMAX system in the sea port scenario

The paper presents a measurement campaign carried out on a 2.5 GHz Mobile WiMAX test-bed in a sea port scenario, aimed at evaluating the system performance. The analysis of acquired data suggested the definition of a 2-ray pathloss model, which is able to fit the actual behavior of the observed marine channel. In order to evaluate the effects of the proposed pathloss model on system performance the paper presents a simulation study carried out by means of OPNET simulator. In particular, the simulation study is aimed at comparing the throughput at IP level obtained using different pathloss models implemented in OPNET, with that obtained when the proposed 2-ray model is considered. Further, the throughput at IP level obtained by simulation is compared with the one measured in the actual test-bed. This comparison points out that the WiMAX OPNET model is able to reproduce in most cases the system performance observed during the measurement campaign, when the proposed 2-ray pathloss model is used. However, the simulator is not able to predict the behavior of actual system in all situations. Indeed, in the areas where the pathloss presents peaks, the breakdown of system performance, mainly caused by synchronization problems, is not predicted by the simulator.

Authored papers

Rosario G. Garroppo, Stefano Giordano, Davide Iacono, "Experimental and Simulation Study of a WiMAX system in the sea port scenario", in IEEE International Conference on Communications (ICC 2009), Dresden, Germany, June 14-18, 2009.

Iwan Adhicandra, Rosario G. Garroppo, Stefano Giordano, "Configuration of WiMAX Networks supporting Data and VoIP Traffic", OPNETWORK 2008, August 25-29, Washington, DC, USA

R. G. Garroppo, S. Giordano, S. Lucetti, "Admission Region for Multimedia Services in IEEE 802.11e Systems", Networks 2004, vol. 1, pp. 411-416, Wien 2004

R. G. Garroppo, S. Giordano, S. Niccolini, G. Procissi, "Experimental Study on QoS Provisioning to Heterogeneous VoIP Sources in DiffServ Environment", in Networks 2004, vol. 1, pp. 111-116, Wien, Austria, 2004

R. G. Garroppo, S. Giordano, S. Lucetti, F. Russo, "IEEE 802.11b Performance Evaluation: Convergence of Theoretical, Simulation and Experimental Results", in Networks 2004, vol. 1, pp. 405-410, Wien, Austria, 2004

R. G. Garroppo, S. Giordano, S. Lucetti, "Capacity Evaluation of Resource Allocation Strategies in the GPRS System", OPNETWORK 2002, vol. 1, pp. 100-106, Washington, D.C., USA, 2002

G. Carrozzo, S. Giordano, F. Russo, "Shared Risk Link Groups for Optimal Lightpath Restoration in OPNET", OPNETWORK 2002, Washington, DC, USA, 2002

R. G. Garroppo, S. Giordano, S. Lucetti, "TCP performance in the GPRS system adopting the Channel Quality Aware Scheduler", IEEE GLOBECOM 2001, vol. 1, pp. 564-569, San Antonio, Texas, USA, 2001

R. G. Garroppo, S. Giordano, S. Niccolini, F. Russo, "A Simulation analysis of aggregation strategies in a WF2Q+ schedulers network", II IP-Telephony Workshop, pp. 102-107, New York, USA, 2001

D. Adami, R.G. Garroppo, S. Giordano, F. Russo, "Experimental investigation of the impact of bursty losses on the TCP performance", European Wireless Conference, pp. 181-186, Dresden, Germany, 2000