Beyond phased arrays—design principles for an imaging radar

Looking farther into the future, an Atmospheric Imaging Radar (AIR) is proposed. An imaging radar uses much of the same technology as phased array radars, and is one in many respects, but with a different signal distribution architecture. An imaging radar uses digital beamforming techniques to image many locations simultaneously. At a minimum an imaging radar needs multiple subarrays, each with its own receiver. Advanced designs may utilize multiple transmitters as well. Initial designs of an atmospheric imaging radar at the University of Oklahoma have focused on using a single, monolithic transmitter to illuminate a volume of space. The transmitter can be one used for a conventional dish-based radar, if desired. The difference occurs with the antenna the transmitter is attached to. A wide radiation pattern, referred to as a spoiled or spotlight beam, is used. This transmitter and antenna combination was chosen to save cost and complexity of the initial model. The decision is technically viable since the same transmit waveform can be sent to the whole volume at once.

When a monolithic transmitter is used, a separate receive array must be used. The receive subarrays differ from a traditional phased array in that the received signal is not summed in analog, but summed digitally. Following the RF chain on each subarray, a digital architecture will be designed to support a next-generation digital receiver. As analog-to-digital converters (ADC) become faster, this will allow them to become closer to their intended sensor. This will foster an environment that will continue to allow a paradigm shift in which digital systems replace analog ones, thus mitigating many non-ideal effects, while reducing weight and lowering economic cost. Thus, the IF signal from each subarray is followed by an ADC. The digitized signal is then filtered and processed to yield digital in-phase (I) and quadrature (Q) signals that are sent to a dedicated computer to extract atmospheric information. This allows the signal to be split to n beamformers without incurring a 1/n signal loss. This allows the full signal strength to be used for each beam formed. To make this possible, the digitization of the received signal must occur at each subarray. This requires a receiver for each subarray, instead of one per radar for traditional radars. To make this possible several obstacles must be overcome. A few of these are array calibration, sample alignment, data transfer and receiver cost. Another challenge in the design of an atmospheric imaging radar is the array pattern design. With the subarrays spaced wavelengths apart, grating lobes in the array pattern must be overcome. Several techniques are available to do this. Digital beam forming techniques are applied to weight the individual subarray beams to form the composite array beam. Also available is spatial weighting of the array. This spatial weighting is also known as a sparse array. With the correct pattern selection from individual element, to subarray, to array it is possible to effectively mitigate the effect of grating lobes from the array spacing.

The full conference paper will provide the trade-offs discussed previously. The design decisions will be defined in the context of the weather radar equation and the impacts to radar performance. Examples of the trade-offs and future work will be given as well.