Enhanced Signal Copy in High Noise Environments, 16-9436Printer Friendly Version
Inclusive Dates: 10/01/03 10/01/05
Background - Enhanced signal copy approaches utilize the diversity of an N-element antenna array connected to a set of N receivers to concurrently enhance a signal-of-interest (SOI) while suppressing interfering signals. If an enhanced copy system is located in a noisy environment, suppressing unwanted interference can present an enormous challenge. For example, on a building rooftop there may be RF emitters present such as emergency services transmitters, cellular transmitters, PCS transmitters, and paging transmitters. If the total number of signals impinging the antenna array is greater than the number of array elements, then it becomes very difficult to separate the SOI from the interfering signals. Recently, researchers have discovered that array processing based on higher order statistics yields a concept known as the virtual array. A virtual array is one that has virtual sensors in addition to the physical array elements. This concept has two significant implications. First, the virtual elements extend the aperture of an array, producing lower sidelobe levels and better signal copy performance. Second, the virtual array allows enhanced signal copy to take place when there are more signals than physical array elements present. Both of these factors have the potential to reduce the cost of signal copy (and direction finding) systems by requiring fewer antenna elements, fewer receivers, and less real estate.
Approach - The approach to this problem consisted of four main efforts. First, we investigated blind enhanced signal copy approaches that exploit second and higher order statistical processing. Second, the impact of the virtual array concept on signal copy performance was studied. Third, a data collection exercise to capture on-air signals that were affected by co-channel interference and noise was performed. Fourth, copy algorithm throughput optimizations were investigated.
Accomplishments - We have studied and derived the theoretical background that supports the virtual array concept. Using an internally developed array response simulator, we applied second- and fourth order blind copy approaches to separate simulated co-channel BPSK and FSK signals. We tested both second-order and fourth-order blind signal copy algorithms on HF co-channel signals collected by a L-shaped antenna array connected to an N-channel HF receiver. We were able to separate multiple combinations of on-air SSB voice signals, FM voice signals, and digitally modulated signals such as OOK, BPSK, and FSK. In addition, the estimated blind copy weight vectors were used to perform copy-directed two-dimensional (i.e. azimuth and elevation) direction finding.