Sonar uses echolocation to measure range to objects at low cost and computational effort. 
From range readings to landmark classification is large step, although bats and dolphins 
are successful at it. These biosonars move while scanning and we employ this strategy 
to classify objects. Our sonar generates a random pulse sequence related to echo waveform amplitude, 
which we term pseudo-action potentials (PAPs) because of their similarities to biological spikes. 
We employ neuromorphic elements, such as delays, threshold,
coincidence detection, short-term memory, and massive parallelism to classify objects from their echoes.  
When the sonar moves along a linear trajectory, a small object produces hyperbolic range readings parameterized 
by the passing range. Estimates of the passing range form a temporal coincidence of the PAP arrival times. 
This talk demonstrates that features of range and passing-range data are useful for object classification.