Can we observe an aircraft launch and recovery on a carrier with ADS-B?

I've been on a quest, scouring for good ADS-B data to analyze carrier launch and recovery. A breakthrough came with a E-2 Hawkeye flight late last month on ADS-B Exchange. While the launch data was sparse, the recovery insights were promising.

I knew my usual data processing workflow, which involves several layers of data smoothing, wouldn't capture the recovery (also known as the trap). Smoothing tends to "smear" events over time, so instead of a normal 2-3 seconds, the trap stretched to almost 25 seconds. I decided to run the converted earth-fixed position data (I only get latitude, longitude, and EGM-96 altitude from ADS-B Exchange) through an alpha-beta-gamma filter to see the results. These filters are typically used in tracking systems to provide predictions and smoothed states, including derivatives.

Now, the indicated airspeed is still smoothed, but I have more control over the smoothing process and can balance noise level with response. Compared to my standard approach, the noise increased, but the trap event duration is much shorter, now about 8 seconds.

I think this is probably the best that can be done with the low update rate ADS-B data we have at the moment. Given the constraints and limitations of the data, this seems to be the optimal solution. However, one of these days, I might try implementing a Kalman filter, though. It could be an interesting experiment to see if it can improve the results or provide any additional insights.