Since there are very large number of antennas at the transmitter and receiver, we may need to estimate large number of channels, correspondingly we may need very large length pilot sequence to ensure orthogonality among pilot signals used for different transmission paths.
In my opinion if first the somehow the pilots are optimized and then channel estimation is performed. this would get better solution. By optimization i mean to use few pilot sequences compared to total number of massive Antennas but use fewer than that but those would be representing most of the channel information. This would eventually reduce complexity at the BS. Secondly a hybrid combination of both can also be used. that is with pilot and blind.
If you send pilots from the many base station antennas to the users, then you need many pilots (at least one per antenna). The original idea with massive MIMO is to send the pilots in the uplink direction only. Then you only need to have one pilot per user in the cell!
There are more details on this in my overview paper "Massive MIMO: 10 Myths and One Grand Question" (http://arxiv.org/abs/1503.06854)
I would say that pilot transmission is necessary to ensure reliable channel estimation, but blind techniques can be used on top of this (also called semi-blind in this context) to mitigate interference.
Blind techniques have the advantage of not requiring the insertion of pilots and hence a loss of spectral efficiency. However, Pilot-based techniques achive a much higher reliability in the channel estimation and much lower complexity. And the spectral efficiency loss due to the pilots has been minimized in recent systems like LTE so that its less than 1%.
So in the end, its a complexity-efficiency tradeoff but I would say pilot-based techniques would still be prefered.
Massive MIMO systems are to support lots of user transmission simultaneously. So the exact channel information is critical. Besides, because of the large-scale antennas in BS, the complexity of computation is another challenge. So based on above two reasons, blind estimation is not a good choice for future massive MIMO systems. For the cost of pilots resource, massive MIMO is always based on TDD mode and the cost is proportion to the uses, which will not take much resources.
I am trying to choose my topic in massive Mimo can anyone help ?i am reading about channel estimation .any suggestion special from Emil Björnson
- Badges
- Science topic