Reply to thread

With an accurate measurement up to 4m TOF is OK. But if you want a reliable low cost motion security sensor, I would not recommend it for its cost and complexity.

However you need to define objects and range for test purposes. One purpose is to define the minimum object size, distance and speed range to define for a) false positive errors and true negative errors (missed alarm). Without these reasonable specs no system can be perfect.  i.e. Do you want to detect a child but ignore a pet?

I believe the best solution already exists with motion controlled exterior lights (however they work(?)

My thoughts are as follows for a design spec depend on if you are designing an Occupancy Sensor or an Intrusion Sensor.

Active IR Diffuse Reflection Motion Sensors using a high current 10%d.f.  pulsed emitter in the range of  10kHz to 50 kHz not commonly used by remote controllers.

Then use an IR Rx similar to the 38 kHz digital remotes with  a BPF Q=10 + AGC except with a stable log amplifier detect any small change in the diffuse Rx signal with a high CNR carrier ratio  and high SNR of demodulated motion signal such that it can detect a person moving slowly at xx m distance.

The carrier f, and Q of BPF improves the range or sensitivity or SNR but also improves immunity from interference.

I would try the top two frequencies and detect then detect changes in amplitude with a suitable detector for motion.

[URL unfurl="true"]https://www.vishay.com/docs/82489/tsop322.pdf[/URL]  These were formerly designed and built by Sharp , the leader in this optical design.

For this to work the Emitter and Rx must be optically isolated. Then design like above digital ones except analog out and detect change in reflection.

You will need a shield and very clean DC power.