Reality Check: LTE-Advanced Business Impacts
Limits of traditional transmission schemes have been reached a while ago and multiple antenna system evolution is needed to cater for future traffic growth. Multiple antenna (MIMO) techniques allow for exploitation of the spatial domain (complementing the time, frequency and code domains ) for providing higher bit rates. To get all MIMO benefits, the number of antennas need to be increased from one to multiple (two is just a pair).
Nowadays sites having only one transmission antenna are mostly special sites, proof-of-concept sites or special sites with low number of users. 1TX is simple and affordable to implement in UE side and most of the mobile devices today only use one antenna to transmit. On base station side however, the 1TX use cases are rather rare.
Typical installations today with cross polarised TX-antennas, and 2RX antennas on mobile devices exploit 2×2 MIMO. 2×2 gives decent improvements in mobile communications using transmission and reception diversity, but the true gains are coming from the spatial multiplexing. In good radio conditions two data streams can be transmitted instead of one, effectively doubling the bit rates. This is how the baseline LTE can reach 150 Mbps.
Next big thing in mobile networks is going to be four TX-antenna deployments, namely 4×2 MIMO. This allows effective use of closed loop MIMO basically meaning higher S(ignal) with smaller I(nterference) and thus more bits per second. The downside with closed loop MIMO is the dependency on the accuracy of the channel feedback from UE. This means that the closed loop MIMO (LTE transmission mode 4, TM4) is feasible in low velocity cases only. On the other hand, this makes deployment decisions much easier. Multi-user MIMO, MU-MIMO, are also usable with 4 TX due tomuch larger precoding codebook compared to 2 TX.
Development has not stopped to 4TX antennas, as 8TX configurations are specified in 3GPP specifications as well. However, to fully exploit 8TX antennas, more than two RX antennas are needed in the UE side. That is, the gain of 8×2 MIMO is somewhat limited compared to 4×2. Or at least maybe not worth the added complexity. Once the technology is mature enough, it would be more beneficial to target for 8X4 and 8×8 MIMO configurations.
Next in the pipeline is massive MIMO with very large number of antennas, meaning antenna arrays with hundreds of antennas serving tens of users. This allows taking beamforming beyond the next level and having ultra-narrow beams towards users. Unfortunately today or in near future such smart antenna arrays are more of an academic interest than implementable systems.
More antennas naturally require more space. One needs to add another antenna panel, add necessary cabling, and very likely also more PA, RRU, RRH or BB equipment. Fortunately the network vendors have developed their systems to need less and less space. What used to be one full rack is nowadays 1U rack device and a tower mounted module. On top of the space requirements, there are also zoning laws, existing lease contracts, tower capabilities, wind loads, etc. which need to be addressed and accounted for.
Certain features and upgrades are feasible in specific situations and locations. For example closed-loop MIMO requires accurate channel feedback and is best in low mobility cases such as pedestrian and CPE.
As the mobile networks evolve and roadmaps are planned, antennas are one significant part of the puzzle. Considering the overall complexity and cost associated to increasing the amount of BTS antennas, and remembering that the gains of different MIMO schemes are highly cell-specific, mobile operators really need to carefully plan their technology roadmaps.
Opportunities to miss are legion. Choosing the best paths requires careful and thorough analysis – As is always the case when maximising customer experience and minimising network cost.