Extending the Bandwidth of a Low-speed Arbitrary Waveform Generator to Synthesize Ultra-wideband Rec
Abstract
We propose to develop an ultra-wideband compact transceiver that has software-defined pulse generation, digital predistortion, and bandwidth reconfigurability. The transceiver consists of a wideband software-defined transmitter enhanced by digital pre-distortion algorithms and a high sampling-rate programmable receiver. The proposed transmitter architecture includes a compact arbitrary waveform generator (AWG), a non-linear transmission line (NLTL) to sharpen pulses to expand the bandwidth of input signals from the AWG, a high-speed switch network to support multiple antennas, and timing circuitry for operation with a receiver. In this architecture, we use a single transmitter source and effectively switch for multiple antenna channels at high frequency. We will develop a non-linear transmission line with minimal distortion in the compressed output pulse. To further minimize distortion, we will develop digital predistortion (DPD) algorithms to pre-compensate for the NLTL’s unwanted dispersion. To assist the development of our DPD algorithms, we plan to develop a transfer function for the NLTL using advanced X-parameters. The transfer function will capture accurate phase and amplitude information at fundamental and harmonic frequencies over 10 GHz bandwidth. Furthermore, we plan to incorporate machine learning to improve DPD algorithms. Machine learning is predicted to lead to a more precisely corrected output by leveraging previously-known DPD results. Using NLTL, we aim to achieve the reduction of the output pulse’s full-width-at-half-maximum (FWHM) from an AWG by a factor of 3 which will triple the transmitter’s bandwidth. The transmitter will emit FWHM pulses with widths in the range of 100-320 ps and voltage swings up to 20 V peak-peak. In parallel, we propose to develop a novel stepped-pulse transceiver that will up-convert trains of pulses from a low-frequency AWG to cover a 10 GHz bandwidth. Preliminary results have shown that this type of transmitters offers increased signal-to-noise ratio (SNR) over single pulses. The proposed architecture is a low-bandwidth (< 2.5 GHz) AWG followed by a high-bandwidth low-distortion mixer fed by a field-programmable gate array (FPGA) controlled RF synthesizer. The receiver will associate each pulse with the synthesizer’s frequency and classify it accordingly. In a digital signal processing block, the receiver will recombine the pulses with optimal weighting coefficients according to a general recombination algorithm. We expect that this transceiver will offer a ??? (??-pulses) increase in SNR. A key factor in the transceiver’s novelty is its on-the-fly reconfigurability. We can program the FPGA’s control to focus RF energy on a specific bandwidth to increase power spectral density, potentially creating dramatic improvements in SNR. Both the stepped-pulse transmitter and NLTL-based software-defined transmitter require a programmable receiver. On the receiving side, we propose to develop a time-equivalent FPGA-based sampling receiver with an 8:1 switch network for 8 receive antennas, timing circuitry for operation with a software-defined transmitter, a bandwidth of larger than 10 GHz, and a selectable equivalent sampling rate of up to 100 GS/s. This receiver is required to support the novel reconfigurability of both proposed transceivers. We propose to develop a novel cavity-based Vivaldi antenna to achieve high and flat gain in a compact size. We will incorporate an innovative resistor loading technique to significantly reduce the antenna’s size. The cavities are holes drilled into the antenna’s arm structures to provide localized current density enhancement. Electrically, the cavities will enhance the antenna’s gain while the total weight of the antenna will be reduced because bulk dielectric material is removed.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Feb 08, 2022
- Source ID
- N000142212167
Entities
People
- Anh-Vu Pham
Organizations
- Office of Naval Research
- United States Navy
- University of California, Davis