Typically you use a pulse train and filter your train from the noise

If one lidar hits another, it will result in at most one bad reading (perhaps a bad column?). This can likely be filtered, or a bad scan (360deg) can be altogether rejected and the data predicted using models based on past sensor readings.

I guess phase and timing sensitivity help a lot, because it's unlikely that another emitter will perfectly match your emission/detection duty cycle. It's also hard to get hundreds of cars at one intersection, because cars are very big.

The key terms in your literature/patent search should probably be "Crosstalk" and "multi-LIDaR".