Implementation of an IEEE 802.11 Multi-Channel Multi-Interface Wireless Ad-Hoc Network

Presentation slides, demo slides, and final report (PDF files).

The IEEE 802.11 standard defines multiple non-overlapping channels at the physical layer in the 2.4 GHz and 5 GHz spectrums. However, most wireless ad-hoc networks today are configured to operate under a single channel in order to ensure connectivity of all their nodes. Hence, the aggregate bandwidth provided by the radio spectrums is not fully-utilized. In order to meet the high throughput demands, it is essential to use all available spectrums. Several past research proposals exploit multiple channels and multiple interfaces to increase the network capacity by having multiple simultaneous transmissions without interference. In this project, I plan to implement one of the multi-channel multi-interface approaches (to be determined) using NS-2 or OPNET and demonstrate the effectiveness of the approach.

References:
[1] P. Kyasanur and N. H. Vaidya, "Routing and Interface Assignment in Multi-Channel Multi-Interface Wireless Networks," Wireless Communications and Networking Conference (WCNC) 2005, vol. 4, pp. 2051-2056, Mar. 2005.
[2] A. Raniwala, K. Gopalan, and T. Chiueh, "Centralized Channel Assignment and Routing Algorithms for Multi-Channel Wireless Mesh Networks," Mobile Computing and Communications Review (MC2R) 2004, vol. 8, no. 2, pp. 50-65, Apr. 2004.
[3] A. Raniwala and T. Chiueh, "Architecture and Algorithms for an IEEE 802.11-based Multi-channel Wireless Mesh Network," INFOCOM 2005, vol. 3, pp. 2223-2234, Mar. 2005.
[4] R. Maheshwari, H. Gupta, and S. R. Das, "Multichannel MAC Protocols for Wireless Networks," Sensor and Ad Hoc Communications and Networks (SECON) 2006, Reston, VA, vol. 2, pp. 393-401, Sept. 2006.
[5] S. Wu, C. Lin, Y. Tseng, and J. Sheu, "A New Multi-Channel MAC Protocol with On-Demand Channel Assignment for Multi-Hop Mobile Ad Hoc Networks," International Symposium on Parallel Architectures, Algorithms and Networks (ISPAN), 2000, Dallas, TX, pp. 232-237, Dec. 2000.

Multimedia Content Delivery Over Wireless Links

Presentation slides, demo slides, and final report (PDF files).

Over the last decade there has been a significant increase in demand for delivery of multimedia content to the end user ? from voice to video. At the same time, the growth of popularity of wireless access technologies has prepared the consumer to expect the increase in bandwidth and increase in the quality of user experience over wireless links every year. As a result, with the popular acceptance of personal wireless devices in everyday life, there is a strong need for reliable multimedia content transmission, both to and from mobile devices. This project will concentrate on studying and simulating multimedia streaming over a wireless link. Existing models will be taken and modified to support multiple clients via multicast or broadcast techniques. As a first step, downlink streaming to a client terminal will be simulated over WLAN. Depending on model availability and time, the uplink streaming capabilities of a client will be examined, introducing an application of video conferencing. The study can be taken further replacing a physical layer with UMTS/HDSPA combination for video conferencing over a cellular network. Current problems and suggestions for future work will be given at the end.

References:
[1] Domoxoudis S., Kouremenos S., Drigas A., Loumos V., "Frame-based modeling of H264 constrained videoconference traffic over an IP commercial platform," IEEE TridentCom, 2006.
[2] Freytes M., Rodriguez C.E., Marques C.A., "Real-time H.263+ video transmission on 802.11 wireless LANs," in Proceedings of Information Technology: Coding and Computing, 2001, pp 112-114.
[3] Minghua Chen, and Avideh Zakhor, "Rate Control for Streaming Video over Wireless," IEEE Conference Proceedings, INFOCOM, 2004.
[4] Abhik Majumdar, Daniel Grobe Sachs, Igor V. Kozintsev, Kannan Ramchandran, and Minerva M. Yeung, "Multicast and Unicast Real-Time Video Streaming Over Wireless LANs," IEEE Transactions on Circuits and Systems for Video Technology, vol. 12, no. 6, June 2002.
[5] Salah K., "Assessing readiness of IP networks to support desktop videoconferencing using OPNET," J. Network Comp. Appl., 2007, doi:10.1016/j.jnca.2007.01.001.

Streaming Video Content over WiMAX Broadband Access

Presentation slides, demo slides, and final report (PDF files).

WiMAX, also known as a WirelessMAN and more formally as IEEE 802.16, is an evolving set of wireless broadband standards focused on delivering high bandwidth wireless access service to subscribers over the "last mile". Wired broadband access such as xDSL technologies are an expensive venture that also impose unnecessarily short maximum connection distances to the subscriber, collectively hindering access deployment in many urban and suburban areas. As WiMAX standards advance, it has become a dominant forerunner both in broadband access and point-to-point backhaul solutions.

The objective of this project is to simulate bandwidth intensive, delay sensitive, video traffic representative of video streaming, IPTV, and other video-rich emerging applications targeted at fixed and mobile subscribers. These video streams are typically encoded using MPEG-2 or MPEG-4 codecs and while these streams are marginally loss-tolerant, their performance is inherently a function of available link bandwidth and delay characteristics. Consequently, this project will examine various performance factors including throughput, goodput, packet loss, packet delay, and packet jitter of a subscriber station over a WiMAX access network.

Recognizing Opnet's widespread industry adoption as a commercial-grade network modeling tool with integrated support for WiMAX, I have chosen Opnet as the tool of choice for this project.

References:
[1] M. Chatterjee, S. Sengupta, and S. Ganguly, "Feedback-based Real-Time Streaming over WiMax," IEEE Wireless Communications Magazine, vol. 14, no. 1, pp. 64-71, Feb. 2007.
[2] D. Niyato, E. Hossain, and J. Diamond, "IEEE802.16/WiMAX-Based Broadband Wireless Access And its Application For Telemedicine/E-Health Services," IEEE Wireless Communications Magazine, vol. 14, no.1, pp. 72-83, Feb. 2007.
[3] WiMAX Network Architecture [Online]. Available: http://santos.ee.ntu.edu.tw/mobile/Speech/WiMAX%20Network%20Architecture.pdf.
[4] F. Retnasothie, M. Ozdemir, T. Yucek, H. Celebi, J. Zhang, and R.Muththaiah, "Wireless IPTV over WiMAX: Challenges And Applications," Proc. IEEE WAMICON 2006, Clearwater, FL, Dec. 2006, pp. 1-5.
[5] F. Yousaf, K. Daniel, C. Wietfeld, "Performance Evaluation of IEEE 802.16 WiMAX Link With Respect to Higher Layer Protocols," Proc. IEEE ISWCS 2007, Trondheim, Norway, Oct. 2007, pp. 180-184.
[6] L. Betancur, R. Hincapie, and R. Bustamante, "WiMAX Channel - PHY Model in Network Simulator 2," Proc. ACM WNS2 2006, Pisa, Italy, Oct. 2006, vol. 202.
[7] H. Juan, H. Huang, C. Huang, and T. Chiang, "Scalable Video Streaming over Mobile WiMAX," Proc. ISCAS 2007, New Orleans, Louisiana, May 2007, pp. 3463-3466.

Congestion Control in TCP Wireless Network

Presentation slides and final report (PDF files).

Many congestion control algorithms have been proposed in TCP protocol and they work satisfactory in wired networks. However, due to high bit error rates and data losses in wireless network, the congestion control algorithms in TCP do not perform efficiently and degrade usage of network bandwidth and throughput of the network. There are many congestion control algorithms have been proposed such as M-TCP and TCP-ADALR in order to improve TCP performance in wireless/wired networks and broadband GEO satellite networks respectively. Our aim is to find a better solution for congestion control algorithm which increases and improves throughput and good put in wireless networks. Also, we intend to simulate the related algorithms in Opnet Modeler and compare them as a result.

References:

Evaluation of TCP congestion control mechanisms using the OPNET simulator

Presentation slides, demo slides, and final report (PDF files).

TCP is the major protocol that provides reliable delivery of packets suitable for various application protocols such as FTP, HTTP, SMTP and SSH. TCP provides error-free data transfers and proper control of data flow and congestion management. However, TCP requires enhancements in order to reliably handle loss, minimize errors, and successfully manage congestion in high-speed network. Various mechanisms have been developed to avoid congestion collapse, such as Tahoe, Reno, Vegas, New Reno, Hybla, BIC, and CUBIC. This project deals with the implementation of TCP Reno and SACK algorithms in the simulated network using OPNET and evaluation of the performance on these algorithms in case of congestion in wired and wireless network.

References:
[1] R. Paul and Lj. Trajkovic, ``Selective-TCP for wired/wireless networks,'' Proc. SPECTS 2006, Calgary, AL, Canada, Aug. 2006, pp. 339-346.
[2] W. G. Zeng and Lj. Trajkovic, ``TCP packet control for wireless networks,'' Proc. IEEE Int. Conf. on Wireless and Mobile Computing, Networking and Communications (WiMob 2005), Montreal, Canada, Aug. 2005, pp. 196-203, vol. 2.
[3] Z. Chen and M.M. Ali, ``The performance of TCP congestion control algorithm over high-speed transmission links,'' IEEE Canadian Conf., Department of Electrical and Computer Engineering, Concordia University, Montreal, Canada, May 2004, pp.1371 - 1374 Vol.3.
[4] G. Holland and N. Vaidya, ``Analysis of TCP performance over mobile ad hoc networks,'' Proc. ACM/IEEE Int. Conf. on Mobile Computing, Networking,, Seattle, Washington, United States, 1999, pp. 219-230.
[5] RFC [online]. Available: http://www.ietf.org/rfc.html.

Examination of Routing Algorithms in Distributed Hash Tables (DHTs) for Peer to Peer (P2P) Networks

Presentation slides and final report (PDF files).

P2P networks, when built in the truest sense of the word, are fully distributed. With no central co-ordinating authority or entity, such as a server or central database, they must make use of other, sometimes novel approaches in order to locate nodes (peers) that hold the information being searched for. Networks such as Gnutella made use of a 'flooding query' methodology, where the entire network was flooded with requests, in the hope that the peer holding the data will receive the request and respond. However, such a methodology is significantly less efficient than a central lookup service that would be provided by a index database hosted on a server, such as was implemented by the Napster service. In an effort to devise a routing and organizing methodology that would provide the fully distributed advantages of P2P and the efficiency and guaranteed results of a central index server, the concept of using Distributed Hash Tables (DHTs) was introduced.

DHTs take the concept of hash tables, which generate values, given input keys and give this capability to every node in the network, with each node given responsibility for a particular subset of keys in the overall keyspace. Thus, the nodes are given the responsibility of maintaining the mapping from names to values. In effect, this capability allows P2P networks using DHTs to scale to extremely large numbers of nodes, while also being extremely dynamic in nature with frequent node arrivals, departures and failures (also known as 'churn'). Finally, more complex services can function on top of such network infrastructure, such as distributed file systems, p2p file sharing, content distribution, multicasting, domain name services and instant messaging. BitTorrent is a prominent and well-known P2P network that makes use of DHTs.

This project aims to investigate the routing capabilities and performance of P2P networks using DHT protocols. Such networks, while aiming to be distributed, scalable and fault tolerant must also deal with challenges such as load balancing, data integrity and performance. This last challenge involves network operations such as routing of requests to peer nodes, storage and retrieval. DHT protocols such as CAN, Chord, Kademlia, Pastry and Tapestry fall within the scope of this investigation. The model of such a network that will be built and simulated aims to answer the following questions:
1. How does the 'geometry' of the DHT selected affect routing ?
2. When node arrival, departure or failure occurs (i. e. churn), how much effort must be made by the peer nodes to adapt?
3. When churn occurs, how quickly can new links and alternative routing options be put into place among existing peers?
4. How is traversal time of a request (and associated packets) from source to destination and back affected by the above scenarios?
5. What possible improvements could be made to available DHTs in order to progress towards the objective of being a fully distributed p2p network with the efficiency and guaranteed results of a central index server?

References:
[1] Ratnasamy, Shenker, Stoica - Routing Algorithms for DHTs: Some Open Questions
[2] Manku - Routing Networks for Distributed Hash Tables
[3] Yang - Measuring the Performance and Reliability of Routing in Peer to Peer Systems
[4] Borisov - Anonymous Routing in Structured Peer-to-Peer Overlays
[5] Ghodsi - Distributed k-ary System: Algorithms for Distributed Hash Tables
[6] Balke, Siberski – DHT Algorithms (2007)
[7] Akella – DHTs and P2P Systems
[8] Risson, Moors – RFC 4981 Survey of Research towards Robust Peer-to-Peer Networks: Search Methods
[9] Ledlie, Mitzenmacher, Seltzer, Pietzuch - Wired Geometric Routing
[10] Marti, Ganesan, Garcia-Molena - DHT Routing Using Social Links
[11] K. Gummadi, R. Gummadi, Gribble, Rantasamy, Shenker, Stoica - The Impact of DHT Routing Geometry on Resilience and Proximity
[12] Aberer, Datta, Hauswirth - Route Maintenance Overheads In DHT Overlays
[13] Tati, Voelker - ShortCuts: Using Soft State To Improve DHT Routing
[14] Li, Stribling, Gil, Morris, Kaashoek - Comparing the Performance of Distributed Hash Tables Under Churn
[15] Li, Stribling, Gil, Morris, Kaashoek - Bandwidth-efficient Management of DHT Routing Tables
[16] Dabek, Li, Sit, Robertson, Kaashoek, Morris - Designing a DHT for Low Latency and High Throughput
[17] Rhea, Geels, Roscoe, Kubiatowicz - Handling Churn in a DHT, Proceedings of the USENIX Annual Technical Conference, June 2004.
[18] Xu, Min, Hu - Reducing Maintenance Overhead in DHT Based Peer-to-Peer Algorithms
[19] H. Shen, C. Z. Xu - Elastic Routing Table with Probable Performance for Congestion Control in DHT Networks
[20] Myrmic: Secure and Robust DHT Routing
[21] T-DHT: Topology Based Distributed Hash-Tables
[22] Distributed Hash Table: http://en.wikipedia.org/wiki/Distributed_hash_table, last modified Jan 22nd, 2008