I have read the article about the BSS Coloring, seems it didn't solve the co-channel interference problem.
Could someone explain to me what exactly BSS Coloring work?
As the topolocy above, when STA1 and STA2 are sitting together, even with different BSS color, the RSSI on AP1 should be similar, so, how to aviod co-channel interference? For my point of view, the STA1's signal would be corrupted by the STA2's signal, and cause retransmission.
One way to look at it, is that you are "playing games" with RSSI, SNR, and other roaming algorithm parameters. Both from the AP and the client perspectives.
I am not out there configuring large networks these days, but I have been involved in Wi-Fi for over seventeen years and do know the PHY and majority of the MAC level issues involved.
From my perspective, coloring is only a SUB-optimization technique. It should NOT be considered a primary method. In other words, the basic CCI level, BSS overlap, and other time honored considerations, understood for at least the last 10 years still have to be observed for coloring to have any significant effect.
As usually happens, future increases in CPU power, and algorithm refinement in AP's and clients could help "coloring" become SLIGHTLY more effective.
Coloring, may help you squeeze out that last bit of localized BSS bandwidth, but if you have a poorly designed RF network, have over-subscribed resources, etc. coloring is not going to help.
Theoretically coloring shouldn't hurt performance, but we've all seen how poorly designed or implemented algorithms can hurt just about any system. Just look at RMM, which definitely cannot make up for a poorly designed physical configuration. Or the "tweaks" that Apple has used over the years. Apple products themselves may have been helped, but often to the detriment of other companies products (there are lots of "holes" in the 8021.11 specifications) on the network.
Many fully adopted 802.11 "ideas" have been dropped over the years - even after years of discussion and rule making, simply because they did not work or help in the long run.
Others, 802.11T for example, have been abandoned because of political bickering.