Base Station Antenna considerations for 5G 18 th September 2018 David Barker CTO
Where we started from (2005) Quintel Antennas Vodafone s Base Station QinetiQ company era. - 5x3G Ops in 1xAntenna - Ind. Tilt per Operator - Large Combiner Unit, but - Avoided site redevelopment -Disruptive idea but very complex Business Model. -We don t do these anymore -Very different company -We value these early years and customers O2 Cabinet 3 s Base Station 2 Quintel Combiner Base Unit
Where we are today 6 Ports 8 Ports 10 Ports 12 Ports 45 Deg HBW Optimized SPR & C/I Smallest Form Factor / Lowest Wind-loading Independent Tilt per Band Maximum Service Optimization Best Quality Base Station Antenna Supplier to T1 US Operators HQ in Rochester, NY, USA R&D in San Jose, CA Acquired by/merged with Cirtek Aug 2017 17K Antennas 2017. 50% Growth YoY since 2013. Always adding innovation to our products Design for Manufacture and Deployment Thought leadership and value-add consulting Leaders in High-Value, Multiport Antennas 3
C/I Better Interference Suppression Optimized for Radiation Patterns Comparison of 13 Vendors Overlap regions causes reduced C/I Vendor M Antenna J Worst Coverage Antenna F Best Coverage Best performing antennas are Quintel (8ft & 6ft) Best performing antennas are toward upper right 25dB -1dB Antenna P Best Interference Suppression Antennas K, H,& M Worst Interference Poorest performing antennas are from Vendors H, K and M Better In-Cell Coverage (Gain) T1 Operators Have Verified Quintel s Antenna Performance >30% Capacity Increase on Average 4
Innovation for Deployment Optimized Wind Loading Antennas have a profile designed to minimize drag. Hurricane Maria, Puerto Rico Our Antennas held out exceptionally well. Credit to Mounts, Brackets, Integrity and Wind-Load of Antennas. Simulations and now wind tunnel tests have demonstrated this Highly Optimized Frontal Wind Loading 5
A brief history of the Base Station Antenna 1986 1992 1998 2006 2012 2018 G900 G900 G900 G1800 G900 G1800 U2100 + RET G900 U900 G1800 L1800 U2100 All RET L800 2T2R G900 U900 L1800 4T4R U2100 L2600 4T4R All RET 6
10-Port 6-Port NR1900 Massive MIMO 6-Port 6-Port 6-Port US Operator Example 2018 L700 and L850 4T4R L700 4T4R CDMA850 L1900 4T4R L2100 4T4R 1900 B25 700/850 B13/B5 AWS1 B66 850 B5 AWS3 L2100 4T4R (AWS3) 3x Positions 2021 As above, but also Independent Tilt for L700 & L850 5G/NR850 4T4R 5G mmimo Antenna Re-farm 2GHz: 1900 or AWS AWS3 B66 1900 B25 700/850 B13/B5 AWS1 B66 Possible 3.7-4.2GHz Both 2GHz and 4GHz 3x Positions 7
More complexity Wider Distribution of UE s in Elevation Plane ntnr 8T8R Passive 16T16R AAS 32T32R AAS 64T64R AAS 8
US Passive Small Cell Variants for 4G/5G 9
Small Cells Operator Host Only Operator Host Lease only Neutral Host Single Op Neutral Host Multi-Operator 1 2 3 4 Operator owns everything Operator owns site Operator owns all hardware Op leases site from CCI Operator owns active hardware Op leases site, Antenna and Fiber from CCI Each Operator owns active hardware Each Op leases site, Antenna and Fiber from CCI 10
AAS Passive + AAS Passive Antenna Category Fragmentation Macro Cell Small Cell mmwave Multi-Band N/A Single-Band AAS Multi-Band Passive Single-Band Open Interfaces, e-cpri and ORAN vital for evolving eco-system Not considering Indoor DAS and Speciality Solutions (Multi-Sector, etc) 11
The Future of the Base Station Antenna in 5G 2018 2024 2030 Continued innovation and development for Passive Antennas for next few years Complex multi-array environments Many challenges PIM, Isolation, beam integrity Size, weight, wind-load Massive MIMO AAS predominately Single Band (2300, 2600, 3500). AAS needs to meet TCO targets. L800 2T2R G900 U900 L1800 4T4R U2100 L2600 4T4R NR700 4T4R L800 4T4R L900 4T4R L1500 4T L1800 4T4R L2100 4T4R L2600 4T4R NR3500 16T16R mmimo NR700 16T16R NR800 16T16R NR900 16T16R L800 8T8R NR1500 32T NR1800 64T64R NR2100 64T64R NR2600 64T64R L1800 16T16R NR3500 64T64R mmimo Evolution to 2030 takes us to using all of the aperture for all of the bands In 2030 there still could be a need for hybrid Passive and AAS arrays Will Open RAN interfaces prevail? Will OEM s take over the antenna function. New players (e.g. Blue Danube), New partnerships? 12
Some Market Predictions Annual Antenna capex MIMO Installed by Platform CBRS Small Cell Growth Source: Mobile Experts IA/AAS Penetration Source: ABI Research Source: ABI Research Source: ABI Research 13
Antenna Design Innovations Optimized Rooftop Antenna No Rooftop Diffraction With Rooftop Diffraction Conventional Antenna Optimal for Rooftop 14
PIM Interference Engineering Bidding for Optimal new spectrum blocks Identifying Optimal RF combining options (existing Spectrum) Troubleshooting Interference issues (existing Spectrum) Quintel PIM Tool can be used for many applications; Strategic to Tactical to Troubleshooting 15
Precoding Massive MIMO Massive MIMO for Next Generation Wireless Systems; IEEE Communications Magazine, Feb 2014 mmwave propagation needs classic beamforming for raw gain and radio channel is more reflective rather than dispersive. Massive MIMO with TDD at sub 6GHz Spectrum allows higher order MIMO to be much more efficiently exploited Massive MIMO can be one of various array structures or distributed, or indeed a composite 96 H 1,1 96 H 1,96 Several 100 s Antennas MRC/MRT is best Many 10 s of Antennas ZF/MMSE is best 16
MIMO Techniques Classic Array with ZF Beamforming -50dBm -100dBm Tx Precoding for UE 2 means UE 1, 3 -UE 10 will be in deep nulls Tx Precoding for UE 1 means UE 2 -UE 10 will be in deep nulls Example above illustrates that data intended for one User can be precoded such that 9 Nulls can also be created where the other (co-channel) users are. 17
MIMO Techniques Distributed Massive MIMO 5.5G? -50dBm -100dBm Tx Precoding for UE 1 means UE 2 -UE 10 will be in deep nulls 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 5m 2 0-10-9-8 -7-6 -5-4 -3-2 -1 0 1 2 3 4 5 6 7 8 9 10 Distributed Massive MIMO produces highly interferometric radiation patterns. ZF Beamforming would also ensure UE s are in nulls. 18
C/I How cell optimization has evolved 2004 2010 2016 2022 2028 25dB -5dB Inter-cell Interference Minimized Inter-cell Interference Increases Exploiting Inter-cell Interference Move to LTE Radio Planning Tilt Settings ACP SON ICIC eicic CoMP C-RAN As more cells are added, more interference seen (sub-6ghz) C-RAN architecture provides a means of interference suppression or coherent signal addition; the notion of the cell blurs or even disappears. C-RAN needs significant fronthaul capacity 19
When wireless is pushed to the limit the answer isn t more of the same The answer is more innovation 20