Design of Accurately Modeled Direction-Finding Antenna Installations, 16-R9679

Principal Investigator
Jason Polendo

Inclusive Dates:  01/01/07 – 07/01/08

Background - Shipboard radio Direction Finding (DF) systems developed by SwRI are critically dependent on a stored table-of-characteristic-antenna responses as functions of signal arriving direction and frequency. This table-of-antenna responses (dubbed the array manifold) has traditionally been generated by at-sea calibration measurements performed at extremely high cost. SwRI has been a pioneer in developing shipboard DF array manifolds from electromagnetic numerical models as a basis for reducing at-sea calibration measurements.

An apparent problem with numerically modeling these electromagnetically cluttered DF sites is the sensitivity of the installed antenna responses to inaccurate structural details. These structural details, geometric in nature, deviate from ship drawings and specifications even though such deviations may be within construction tolerances. Small deviations in nearby conducting structures can have a major impact on the response of the installed antenna. Installing antennas at locations having reduced sensitivity to deviations in structural details will result in actual responses that better match the modeled responses, thereby reducing array manifold errors and improving DF performance.

Approach - The capabilities developed within this internal research project enable the design of distributed shipboard antenna arrays whose actual antenna responses maintain a high probability of agreement with those predicted by a numerical model even under the uncertain structural variability. Through the use of a developed surrogate numerical ship model, an iterative perturbation analysis was completed to not only determine which geometrically uncertain ship structures are most influential to the individual antenna responses, but also to quantify the explicit impact structural variability has on each specific antenna site.

Accomplishments - The methodology developed in this project can be applied as a platform-specific technique or in a heuristic manner by applying the generalities stemming from the project analyses to candidate antenna locations on another platform. Employing these practices will yield DF antenna locations, and thus a DF array, that is robust to structural differences in platform details, thereby minimizing the DF error attributed to array manifold mismatch between the measured antenna responses and modeled array manifolds, or cross-decked manifolds. Such minimization efforts are particularly crucial to DF performance in instances where the modeled array manifold is utilized as the calibration database against groundwave and/or skywave signals.

2008 Program Home