Top Ten Most Wanted Wireless Innovations

RF Environment Modelling Techniques for Heterogeneous Network Management
Tawachi Nyasulu, University of Strathclyde

To meet the growing demand for wireless connectivity, there is a regulatory shift from static to dynamic spectrum assignment (DSA), from homogeneous to heterogeneous access and usage of the same spectrum band. Furthermore, general authorised access (GAA) is being advanced to spur wireless innovation and universal internet access. It is therefore anticipated that there will be multiple network operators, multiple Radio Access Networks (RANs), and multiple Radio Access Technologies (RATs) that are required to coexist in the same radio frequency environment. In order to represent such a complex radio environment using near-accurate and efficient data models for automated network management functions such as coexistence management and spectrum coordination, there is need to investigate new mathematical tools.

While similar research has been concentrated on coexistence of multiple RATs in mobile cellular networks, there is more work that needs to be done for dynamic shared spectrum, such as Citizen Broadband Radio Service (CBRS), which is governed by regulatory requirements that are different from those of mobile cellular network spectrum bands. The challenge is distinctive in dynamic shared spectrum due to the dynamic nature of the availability of spectral resources and in GAA spectrum where multiple operators have equal spectrum access rights. Due to the high complexity of such environments, one can envision the need for novel modelling techniques for sufficiently-accurate representation of the properties of the RF environment using data structures that, for example:

a)       Enable efficiency in exchange of coexistence-related information for coexistence discovery, 

b)      Reduce complexity of computation of coexistence decisions and spectrum assignments,

c)       Support efficient algorithms for desired network management outcomes,

d)      Facilitate implementation of Artificial Intelligence/Machine Learning (AI/ML) databases and algorithms for optimising network performance and predicting spectrum availability.

Spectrum Monitoring for Spectrum Sharing
Robert W. Stewart and David Northcote, University of Strathclyde, Glasgow, UK.

The idea that new approaches to spectrum management are needed to enable better reuse of the RF spectrum is gaining traction. Regulators are beginning to consider moves away from the fixed frequency allocation models that have been used up until now, exemplified by the introduction of shared bands in the UK.

New technologies now offer the enabling platform to develop more dynamic spectrum management methods. In this talk, we will present a proof-of-concept innovation that overlays regulator spectrum allocations onto a spectrum analyser tool, fusing these two sources of information to enable better decisions for shared spectrum access. This is a single-chip solution, based on the Xilinx RFSoC (which features multi-GHz ADCs) and delivered via a browser-based interface, with system development leveraging the PYNQ software/hardware framework. A live demonstration will be given as part of the talk, highlighting in particular the ability to probe and visualise the spectrum at remote locations from a standard web browser.

High Q Tunable RF Filters For Next Generation Radio Systems
Thomas F Raschko, Anlotek

Anlotek, a member of the Wireless Innovation Forum, has developed a very versatile and effective tuneable RF filter.  The method has been fully proven in simulation and prototype hardware.  Filter characteristics are excellent with:

-          Predictable, uniform response as the filter is tuned throughout its range

-          Deep rejection, on the order of -50 to -60dB, with no spectral regrowth

-          No insertion loss as the filter is active

-          Fully, independent, tuneable bandwidth with the ability to provide very narrow bandwidth

-          An ability to form Gaussian, Butterworth, or Chebyshev filters with the same circuit

Anlotek’s circuit has been demonstrated as a tracking filter for super-Nyquist DDS outputs, allowing a very clean DDS output well above the clock frequency.  It has also been shown as a well behaved, effective receive filter for direct conversion.  The next step for Anlotek is to produce the circuit on an IC.  Goals include development of the IC in such a manner that it is simple to operate, self-calibrating and the RF engineer is able to determine the frequency range of the part with the addition of a few passives.