Hi,
Dave's idea of using SigmaDSP can really help with your application since you're not exactly sure of what response might work the best. You can tweak a SigmaStudio design to your heart's content -- something that's difficult with multi-section active filters. Two SigmaDSPs come to mind -- the ADAU1772 and the ADAU1701. Both have on-chip converters for analog-in, analog-out simplicity, and both can boot themselves from an external E2PROM which SigmaStudio can program for you. The -1772 is low-power, has on-chip mic inputs and a headphone driver -- a nearly perfect match for your application. Its DSP resources are comparatively limited, but there's more than enough horsepower to build a filter like the one shown below:
I got the gain, center frequency, and Q for each section from the Filter Wizard mentioned above, typed them into Sigma Studio's drag-and-drop filters, did a little manual tweaking, and ended up with the predicted response shown. You can download SigmaStudio and play with response curves right away. With an ADI evaluation kit you can hear the results as well.
The AUAU1701 DSP has more algorithms in its toolbox, which may allow you to design in extra features -- such as blinking an LED when it recognizes pinging. For this chip you'll need to build a mic preamp, and its higher power usage -- about 110 mA at 3.3V -- means that with a good switching regulator you could power it from a 9V battery, but not for very long.
Best regards,
Bob