Channel estimation time for fdd is longer than tdd due to non reciprocity of channels in fdd. Why then does this time depends on number of antennas and independent of antennas for tdd.
To estimate a MIMO channel, you need to send one pilot per transmit antenna. The number of uplink pilots are then equal to the number of (single-antenna) users and the number of downlink pilots is equal to the number of base station antennas. By using TDD and exploiting channel reciprocity, you only need to send pilots in the uplink.
You can read more on this in the TDD/FDD part of the paper:
Emil Björnson, Erik G. Larsson, Thomas L. Marzetta, “Massive MIMO: Ten Myths and One Critical Question,” IEEE Communications Magazine, vol. 54, no. 2, pp. 114-123, February 2016.
But m is greater than k users. What would be the extra antennas used for in the uplink
You use the antennas to achieve an array gain and to better separate the users in the spatial domain. Have a look at Eq. (1) in the paper and you will see how the SINR is proportional to M.
Dear @Emil can you please explain in simple terms what is Channel Heardening in Massive Mimo setting and what are the implications benefits and pros and cons
Channel hardening means that a fading channel behaves as if it was deterministic, when we have applied the uplink detector or downlink precoder. This makes the system more reliable, simplifies the estimation of the effective channel, and simplifies the resource allocation since the effective channel is equally good in every coherence interval.
Dear @Emil what challenges we could expect in massive mimo channel estimation in uplink for high speed users. Because precoder and accurate beamforming is only possible with a suitable coherence time and estimation window. For high speeds the channel may become outdated. Can you point out some work done in this regards.
- Badges
- Science topic
- Similar topics
- Psychology