|
|
 |
Identification Algorithms
In addition to the design of detection algorithms matched to observable signals and sensing physics, APS has developed algorithms to automatically identify signals as belonging to specific classes of targets. The automated identification algorithms include feature-based statistical classifier systems and Markov sequence identification using hidden Markov model analysis. APS utilizes a rich set of industry and custom analysis tools to identify the key parameters that, when properly estimated, will provide the optimum separation of detected classes of signatures for identification.
Examples of identification algorithms and applications developed at APS include:
| |
 |
Identification of acoustic events associated with artillery range operations for noise monitoring and abatement programs |
| |
|
Identification of AC and DC motor signatures in ELF Magnetic and Electric field sensing apparatus |
| |
|
Identification of humans from RF estimation of walking-gait parameters and statistical classification from radar scattering features |
| |
|
Identification and characterization of submarine acoustic sources measured at noise monitoring facilities |
| |
|
Active sonar classification of scattering objects using a feature-based statistical classifier to separate submarine echoes from discrete boundary reflection |
 Top of Page
|
|
