Just to explain something what I know.
The way PDoA was done in DW1000 Beta PDoA kit using Sync is just to clock devices from the same source, and because DW1000 was not able to do AoA out of 1 chip. So, you sync 2 chips to run synchronously; then on RX each delivers its portion of diagnostic information, based on Ipatov sequence. Out of that information, and because DW1000 were clocked from the same source, Beta PDoA Kit provided the method to calculate the AoA between 2 antennas.
The DW3000 chip is different in this respect, it does the calculation of AoA based on the STS sequence. So for DW3000 chip AoA to operate, you must have both Ipatov and STS, i.e. the sts_mode shall be set to 1 (SP1 packets) or 3 (SP3 packets), but both have STS.
As well there are 2 PDoA modes, they do PDoA differently:
PDoA_1 → the phase measured between the Ipatov and the STS (can use plen=64 and sts_len=64).
PDoA_3 → the phase measured between the first half of STS and the second half of STS, works only when STS length is > 256, so use PLEN = 128, STS=512, PDoA_Mode = 3 in this case.
Each DW3000 chip calculates phase in between ports ant_1 and ant_2, the antennas should be placed in 1/2 wave len from each other.
In your experiment, my understanding is that you want to place 4 antennas in the square as below:
chip_1_ant_1 chip_2_ant_3
chip_1_ant_2 chip_2_ant_4
So in this experiment, you can get phases from chip_1_ant1_vs_ant2 and phase from chip_2_ant3_vs_ant4 ← no synchronization of DW3000 chips is needed. you just would get this information as a part of standard in-chip calculateion.
But now you want to get phases as chip_1_ant_1_vs_chip_2_ant_3 and chip_1_ant_2_vs_chip_2_ant_4.
I do not quite sure if this information is available or can be extracted from DW3000 chips in the same manner it was available on DW1000.
So you need to experiment and try to systematically research the topic…
Sorry but I cannot help here more. Good luck and if you solve it, please let others know.