Feasibility Study for Embedded Software Control of Flexible RF Filters,
Inclusive Dates: 12/21/12 – 04/21/13
Background — As part of a Defense Advanced Research Projects Agency (DARPA)-funded project, Southwest Research Institute (SwRI) previously teamed with a small business to create a controller algorithm for a new type of tunable radio frequency (RF) filter. During this work, the need to develop a rapid-prototyping capability for embedded control systems was recognized. The purpose of this internal research was to improve these capabilities internally, and to investigate possible implementation approaches for the control algorithm resulting from the previous project.
Approach — A hybrid feed-forward/feedback controller model was developed in a graphical modeling environment. This modeling environment enabled the controller algorithm to be deployed to a microprocessor or field programmable gate array (FPGA) on an accompanying hardware platform. This setup provided the flexibility necessary to investigate multiple deployment strategies without requiring significant rework of the model code.
Using models derived from a hardware characterization, the feed-forward portion of the control algorithm was implemented in a microprocessor. This approach was sufficient to demonstrate basic control capability, enabling a user to select a desired center frequency and tuning the filter appropriately. This level of control provided reasonable results for static frequencies at ambient temperature.
Due to the early prototype status of the customer-provided hardware, SwRI did not receive a functioning feedback sensor early enough to include it in this feasibility study. By including a feedback sensor in the filter control setup, it should be possible to greatly increase the stability of the system across a variety of operating conditions.
Accomplishments — The experience gained from this research should improve SwRI’s rapid-prototyping capability for embedded control systems. Although hardware delays impacted the intended scope of this project, the prototyping setup was sufficiently flexible to allow parts of the system to be simulated in the absence of hardware.
Follow-on work has already been awarded based partly on results from this research, and work will be continued in collaboration with the small business and a major defense contractor. This technology has broad applicability in areas such as electronic warfare, radar systems, and cognitive radio, and other entities at SwRI have expressed interest in applying this technology to their own projects.