Sensor Array and Multipath Signal Processing Lab

The Sensor Array and Multipath (SAM) Signal Processing Lab is concerned with the extraction of information from signals carried by acoustic or electromagnetic waves propagating in complex media. A sensor array is a configuration of devices, each of which converts energy into an electrical response which can then be represented by a sequence of numbers in a digital computer. SAM processing is used in sonar, radar, biomedical imaging, seismology, wireless communications, radio astronomy, and oceanography, to name a few application areas. Historically, array processing methods were developed assuming single path propagation models due to their analytic and computational simplicity. This is despite the fact that in many problems, reception of signals via multiple propagation paths between the source and receiver is a dominant feature. Multipath signal processing thus refers to methods which explicitly account for realistic propagation in more complex environments. By virtue of the uncertainties in modeling such environments, as well as the presence of noise, interference, and clutter, SAM signals are typically characterized as multichannel random processes. 

The key to successful SAM information extraction is the integration of robust physical phenomenology into the design of statistical signal processing techniques. Intuitively speaking, we try to design statistical signal and array processing techniques which exploit or account for physical model-based phenomena which we can see in real data. Examples of this are:

1. Aircraft height finding with radar that uses the direct and sea-surface reflected multipath returns.

2. Underwater broadband acoustic source localization which exploits waveguide-induced relationships between modal phase and group velocity differences.

3. Skywave HF radar clutter mitigation that compensates for abruptly time-varying multi-modal ionospheric propagation.

4. Sonar surveillance that uses dynamically reconfigurable distributed sensor arrays which account for changes in the underwater acoustic noise and propagation environment.

These are just some examples of the sensor array and multipath signal processing methods being developed in our group. A distinguishing characteristic of our work here at Duke is that we process real data and thus have impact on real systems. Please explore our website to learn about our current and past projects as well as opportunities to join us here at Duke.

Reflectivity Image