Impact Study for Direction-Finding Antenna Count Reduction, 16-9226Printer Friendly Version
Inclusive Dates: 11/01/00 - 03/01/01
Background - The next generation of shipboard signal exploitation system for the U.S. Navy will incorporate super-resolution direction-finding (DF) algorithms such as MUSIC, a radio direction-finding algorithm that requires a separate receiver and signal processor for each antenna response in the DF system. Because each wideband receiver and associated processor is a significant cost driver, competitive pressures will demand that the number of antenna outputs be minimized without significant loss of performance. SwRI's AS145 antenna has been cost effective because 1) the AS145 antenna design is less expensive to manufacture and to install than the antennas supplied by SwRI's previous competitor, and 2) a sequenced receiver system has been exploited, requiring only two channels of receivers and associated processors regardless of the number of antenna responses. However, sequenced receivers do not support the MUSIC DF algorithm, so that total cost of all the expensive wideband receivers and associated processors will depend directly on the number of antenna responses being used.
Tradeoffs between DF system performance and number of contributing antennas have been well established for textbook scenarios describing arrays of identical antennas laid out on uniformly spaced grids without interactions from scattering surfaces (such as ship superstructures). In contrast to the textbook scenarios, the constraints of the shipboard installation result in a group of antennas having highly distorted patterns that differ significantly from one location to another within a highly irregular array geometry. The highly complex responses associated with the shipboard installation scenario do not lend themselves to straightforward theoretical trend analyses and must be empirically evaluated and summarized.
Approach - During this internal research project, a processing approach was developed and exercised to examine the impact of array decimation on high-frequency DF performance. This approach used calibration responses measured from the original, complete array to create the array manifold that would actually be measured for any selected subset of elements within the array. This sequence of analysis produced 1) a statistical characterization of DF performance specific to the particular ship being analyzed, and 2) identification of the best- and worst-performing subarrays for a specific value of antenna count.
Accomplishments - This analysis capability provides a first-time statistical summary of DF performance as a function of antenna count and signal-to-noise ratio for DF arrays installed in cluttered sites. More importantly, it shows the range of performance available from specific antenna counts and identifies which subarrays yield the best DF performance and which subarrays yield the worst DF performance, all on a band-by-band basis. Such information provides critical guidance in selecting subarrays for operation under constraints of limited receiver channels.