FuNLab Projects

Vehicular Networks

Lead: Fei Ye

Throughout the years, the vehicular safety systems have evolved from passive systems (seatbelt, airbag) to active systems (ABS, EPS). Recent researches show that leveraging the state-of-art wireless networking technology is the next logical step towards a pro-active vehicular safety system. Such pro-active system will provide reliable safety services while concurrently support commercial data service. Vehicle-to-vehicle ad hoc communication and vehicle-to-infrastructure communication are involved. Currently we are investigating:

  • Emergency Warning Message (EWM) propagation MAC protocol in vehicle-to-vehicle ad hoc networks for collision avoidance applications. The stringent latency and reliability requirements are satisfied at all costs.
  • Efficient data dissemination protocol at layer-2 and layer-3. Commercial data service will greatly accelerate the development and deployment of vehicular networks. The success of commercial data service relies on harmonious integration of data acquisition through vehicle-to-infrastructure communication and data sharing through vehicle-to-vehicle communication.

For more information see the VANET Project Homepage

[top]

Collaborative Wireless Networks

Lead: Wei Shi

Cooperative communications and networking allows different users or nodes in a wireless network to share resources to create collaboration through distributed transmission/processing; each user's information is sent out not only by the user, but also by collaborating users. Cooperation promises significant increase of capacity and reliability in wireless networks, by counteracting fading channels with cooperative diversity and by exploiting the broadcast nature of wireless. Currently we are investigating:

  • The cooperative schemes for more practical and complex relay networks, such as multi-hop relay networks. The cooperative schemes include relaying strategies and protocols, relay selection and resource allocation.
  • In conventional multi-user networks, only a limited number of the degrees of freedom is used by each transmitting node. Our research looks at novel ways to increase the degrees of freedom used.

Network Coding

Affiliated Students: Hamed Firooz and Linda Bai

Linear Network coding has received considerable attention in recent years for its potential for achieving the theoretical upper bound (max-flow) of network resource utilization via the introduction of coding concepts at the network layer. It has been shown that with simple distributed linear coding in place of the usual forwarding at intermediate nodes, system throughput can be increased in several canonical network topologies. In this project we investigate novel use of network coding for a different purpose - to locate congested link(s) inside a network through end-to-end measurements at (external) boundary nodes. Current link monitoring schemes suffer from identifiability problems; i.e. they are unable to infer link status in many canonical network topologies. We show that network coding offers the promise of being able to identify congested links in any `logical' network.

For more information see the Network Coding Homepage

[top]

Cognitive Heterogeneous Networks

Lead: Chittabrata Ghosh

Due to the rapidly increasing demands of wireless bandwidth, available spectral resources become scarce. Moreover, current fixed spectral allocation suffers from underutilization as in the case of TV bands. These are the motivations of Cognitive Radio (CR) network, which is defined as a radio that makes secondary users actively search for the idle channels NOT used by the primary users by scanning the whole spectrum. Our current research mainly focuses on channel modeling, search algorithm analysis and channel detection in CR network. Currently for CR, we are investigating the following:

  • We are studying efficient dynamic spectrum access strategies using Markov Chain and Queuing Theory. The adaptive strategy of scanning for spectral opportunities in a dynamic scenario, where the usage pattern of primary users change is also included in potential directions to extend our work. The modelling of channel state information using correlated Markov model is expected to be utilized to design such a strategy.

For heterogeneous networks:

  • Efficient vertical hand-off strategies between various networks to support user QoS requirements

For more information see the Cognitive Radio Project Homepage

[top]

Adaptive Mesh Networks

Lead: Rohit Gupta, Hui Ma, Fei Ye

802.11 wireless LANs for broadband wireless access constitute a growing success story. Their deployment in single-cell (i.e. single AP) scenarios (homes, small business and hotspots) is well-supported by current .11 technology. However scaling such networks to the enterprise environment to serve a large number of simultaneous users with voice and data services while providing coverage remains a challenge. A promising architectural solution consists of a two-tier multi-cellular, multi-hop approach to WLAN network design whereby an AP mesh provides the infrastructure for communication between mobile clients. Implicit in this is a direct wireless inter-connection between mesh nodes which all route traffic (only some of which are APs with associated clients, and a very small fraction act as gateways to the wired Internet).

Our research seeks to advance the state-of-art of such .11 based multi-hop mesh networks by undertaking an integrated cross-layer approach to innovations at Layers 1-3 (joint PHY/MAC/Network). Optimizing of such networks will require on-line tuning of key protocol parameters at various layers (hence leading to an Adaptive Mesh). Currently we are investigating

  • The impact of Physical Carrier Sensing (determined by carrier sensing and/or receiver sensitivity threshold) on MAC performance as a function of network topology
  • The impact of Multiple Available Channels and Multi-radio Mesh Nodes on network performance. Appropriate utilization of multiple channels is the key to scale the capacity of wireless mesh networks. This in turn places a premium on channel-to-link assignment algorithms for interference mitigation. We are investigating the cross-layer information aided channel assignment as well as the joint channel assignment and routing.
Our approach combines protocol/algorithmic innovation supported by OPNET and NS3 simulations and experimental results from a StarEast based MESH network in a laboratory setting.

For more information check http://commnet.ee.washington.edu/funlab/
For more information on MRMC WMN research, please check MRMC WMNs Homepage

[top]

Underwater Acoustic Networking

Lead: Nathan Parrish, Leonard Tracy

Underwater acoustics has been a topic of research for decades. However, the idea of deploying networked teams of underwater vehicles for both deep and shallow water ocean exploration is a more recent topic of interest. The Seaglider, developed at the University of Washington, is one such vehicle. FUNLab along with the Applied Physics Lab (APL) are exploring physical and MAC layer protocols to provide robust, low power, efficient networking solutions to the Seaglider.

The underwater acoustic channel has properties that make it a very difficult medium for communications. For instance, the long propagation delay of sound, multi-path spread of the medium, frequency selective attenuation, shadowing zones, and other factors make this channel extremely hard to characterize. A commonly used approach for determining the acoustic propagation of sound in the underwater channel is to use ray tracing techniques based on Snell's Law. Members of the FUNLab are investigating ways to statistically characterize the underwater channel using techniques similar.

MAC protocols in the underwater environment must be designed with different considerations than those in the terrestrial environment. The long propagation delays of sound make carrier sensing and acknowledgment packets impractical. Additionally, autonomous underwater vehicles (AUVs) are extremely energy-constrained. These and other design considerations, including the lack of position information from GPS, necessitates new MAC design for AUV deployment, which is also an ongoing topic of research between the FUNLab and the APL.

For more information see the UAN Projects page

[top]

ns-3

Principal Investigators: Tom Henderson (Boeing), Sumit Roy (UW), George Riley (Georgia Tech), and Sally Floyd (ICSI, Berkeley).

Contributor: Illango Purushothaman (UW)

ns-2 is a discrete-event network simulator that is being used extensively in network research circles. It has been funded by a number of previous research projects, but none directly as an infrastructure project since 2000. A clear need exists for additional focused development work to occur on ns. The ns-3 project seeks to do just that and is the next major revision of the ns-2 simulator.

ns-3 is a managed software development program to comprehensively re-design, enhance and maintain the popular ns-2, to address research and educational challenges for next generation of data networks. This four-year project will

        i) refactor the simulator's architecture

        ii) develop new networking protocol models for wireless

        iii) provide new opportunities for software encapsulation, and

        iv) integrate the tool with virtual network testbeds.

For details on this project, check http://www.nsnam.org/.

To download the latest 802.11 infrastructure mode patch for ns-2, Click here.. The changes made to the ns-2 module are described in this report.

 

[top]

Advanced Imaging Approaches for Detecting Obscured Objects

Principal Investigator (PI): Sumit Roy, Co PI: Yasuo Kuga

Investigator: Sermsak Jaruwatanadilok and Thomas Chan

This research aims to develop new methods based on the synergy between electromagnetic wave imaging and detection and network techniques to solve a difficult problem of detection and imaging of targets-of-interest in the presence of complex, unstructured, and obscured environments. We are currently investigating:

- Detection and imaging of a target on a random rough surface using correlation function including the analysis on probability of detection and probability of false alarm performance

- Imaging of a target in cluttered environments using passive coherent radar and angle-of-arrival algorithm

For further details, please refer to this link http://www.borders.arizona.edu/.

For more information see the Advanced Imaging of Obscured Objects Project Homepage

[top]