SWIFT: Interference Mitigation using Spatial and Frequency Nulling for Wideband mm-Wave Transceivers

Currently deployed fifth generation (5G) solutions operate in frequency bands, mostly below 6 GHz, and to a limited extent at 28/39 GHz, and expansion to other emerging frequency bands is under consideration. When operating at 28 GHz or 39 GHz, hundreds of antennas are needed to boost the power of the transmitted signal to form a focused beam to increase the communication range.

On the receiver, there is a unique opportunity to take advantage of the many antenna elements to also cancel out or attenuate interference from unwanted directions. This proposal seeks to understand the most power efficient and optimal means of achieving interference cancellation. Furthermore, as the transition of high-speed applications starts to occur from sub-6 GHz frequency bands to higher frequencies, especially at 28/39 GHz, traditional means of cancelling unwanted interference operating in other frequency bands (other “channels”) using high quality acoustic resonators does not seem viable.

As such, this project will explore the application of filters that are electronically tunable and take advantage of the switching properties of modern digital devices as a means of overcoming these limitations. While addressing the coexistence issue is a major hurdle for commercial radios, many other radios are also in danger of losing functionality and sensitivity if steps are not taken to address the problem of interference.

Many important sensors, such as weather radars and radio astronomy telescopes, may cease to function if interference levels increase as expected. It is therefore imperative to protect such radios by directly collaborating with the radio astronomy community. This research will also lead to the training of students in the engineering of modern communications systems and wireless communications.

Future generation of radios utilize broader band millimeter-wave (mm-wave) front-ends, higher channel bandwidths (1 GHz or more), and beamforming. Multi antenna array signal processing techniques naturally provide some spatial filtering of unwanted signals. By placing nulls in the antenna pattern to purposefully “zero-out” interference, one can improve signal-to-distortion ratio, achieve better spectrum utilization, and realize more robust radios.

While all these benefits can be realized using digital signal processing, the wide dynamic range requirements on the analog-to-digital converter make the hundred element array radio high power and even unfeasible. Furthermore, out-of-band interference will likely pose an issue when the mm-wave spectrum is more crowded. Traditional ways of removing interference using sharp filters based on high-Q resonators is not viable above 10 GHz.

To address these issues, interference cancellation will be explored in several locations in the receiver, at the radio frequency itself (using tunable electronic notch filters), at the boundary between radio frequency blocks and analog blocks (spatial notch), as well as inside the analog-to-digital converter itself. In this way, interference will be rejected before (or during) quantization, which will reduce the dynamic range requirements of the receiver greatly.

This is especially important at wider channel bandwidths proposed in 5G and future generation radios. Cancellation of interference, though, requires knowledge of the location and frequency of the interfering signal. Digital and baseband tracking loops will be explored to identify the properties of the interfering signal and feedback loops will then allow interference cancellation to be performed in the various forms.

Application of these techniques for the protection of passive radio sensors, such as radio astronomy telescopes, will be studied and techniques for interference tracking from the radio astronomy community will be investigated.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.