LTE and Wi-Fi SON: Opportunities and Lessons Learned
Steven GlapaSteven GlapaSenior Director of Field Marketing
About Ruckus
Optimized packet-by-packet selection from among 2n
possible patterns
Large number (n) of small, inexpensive antenna elements with directional and polarization diversity
BeamFlex optimization engine on host processor
Off-the-shelf 802.11 chipset and RF front end
digital switch
2
7 years
2,500,000 units shipped in carrier, enterprise Wi-Fi
~100% annual growth
2x real-world performance gain over conventional Wi-Fi
350 people, 200 engineers
By the numbers:
80 countries
100,000-node 3G offload network underway in Tokyo
45,000-node mesh network in India
#1 outdoor Wi-Fi share
What LTE SON can learn from Wi-Fi
0ddt
(attribute)
3
ddt
> 0(attribute)
Why not just use Wi-Fi?
Small-cell Infrastructure Capex, US$/Mbps/km2 and Availability
0 1,000 2,000 3,000
HSPA
LTE
Now
2012?
4
Source: operator and TEM benchmarking, Ruckus back-of-the-envelope analysis.
LTE
802.11n
802.11ac
2012?
Now
Year-end 2012
Dynamic Authentication in Secured Wireless Networks (4085US/EP/CN & also see 4264US)
Provisioned Configuration for Automatic Wireless Connection(4086US/EP/CN)
Multicast to Unicast Conversion Using Distributed AP Control(5100US)
Omni Planar Antenna w/Selectable Elements(2823US/7,292,198 2823CN/TW/EP)
Minimized Antenna w/Selectable Elements(2868US/7,362,280)
Improved Data Throughput in Communications Networks(2880US/7,505,447 2880EP/1759543)
Circuit Board w/Peripheral Antennas(2894US/7,193,562 [REX] 2894CN/TW/EP)
Wireless LAN Device(3325US/D532,001 3325TW/D113194)
Antenna w/Vertical & Horizontal Polarization(3346US/7,358,912 [REX])
Pin Diode Network for Multiband RF Coupling(3441US/TBD 3441EP/CN)
Multiband Omni Planar Antenna with Selectable Elements
Details will necessarily be limited
5
Multiband Omni Planar Antenna with Selectable Elements(3442US 3442EP/CN/TW)
Selectable Antenna Elements w/Phase Shifting(3443US/7,498,999 3443TW)
Antennas w/Polarization Diversity(3529US/7,498,996 3529CN/EP/HK)
Multiple-Input Multiple-Output Wireless Antenna(4080US)
Peripheral Antenna Apparatus(4159US/7,525,486)
Wireless Access Point(4335US/D596,628)
Wireless Access Point Base(4351US/D596,168)
Wireless Access Point Dome(4352US/D595,705)
Minimized Antenna w/Selectable Elements(4393US/7,511,680)
Wireless Access Point(4601US/D596,635 4601CN/TW)
a word from our patent lawyer...
A wealth of useful control
Classic Beamforming Ruckus BeamFlex (1,000s of these)
6
W-CDMA or LTE small-cell application
Autonomously
Organize
Adapt to long-term changes
Null
7
Null
Adapt = f (time of day)
Baseline demand distribution
8
Deploy at available sites, self organize
9
New traffic trend
10
Add cells, adapt
11
Null
interference source
12
Traffic = f (time of day)
e.g. park at lunch hour
13
Adapt temporally
14
Customers report huge TCO advantages
80% Less technical staff time50% Lower cash costs
Conventional Alternative
~50% less capex and ongoing cash maintenance costs
Conventional Alternative
~80% reduction in technical staff time required for WLAN
15
Ruckus
Annual (3x)
1x Capex
Average results from sample of 40 Ruckus case studies
Ruckus
Maintain
Deploy
required for WLAN installation & maintenance
Troubleshoot
Representative Hospital Case Study (Midwest Surgical)
Simulation results: 1.5 to 2x gains
6 0.1 -6 5 .1
5 5 .1 -6 0.1
5 0.1 -5 5 .1
4 5 .1 -5 0.1
40.1-45.1
35.1-40.1
30.1-35.1
25.1-30.1
20.1-25.1
15.1-20.1
10.1-15.1
5.1-10.1
0.1-5.1
-4.9-0.1
DL Mbps
6 0.1 -6 5 .1
5 5 .1 -6 0.1
5 0.1 -5 5 .1
4 5 .1 -5 0.1
40.1-45.1
35.1-40.1
30.1-35.1
25.1-30.1
20.1-25.1
15.1-20.1
10.1-15.1
DL Mbps
Macro
Macro + Omni
16
6 0.1 -6 5 .1
5 5 .1 -6 0.1
5 0.1 -5 5 .1
4 5 .1 -5 0.1
40.1-45.1
35.1-40.1
30.1-35.1
25.1-30.1
20.1-25.1
15.1-20.1
10.1-15.1
5.1-10.1
0.1-5.1
-4.9-0.1
DL Mbps
10.1-15.1
5.1-10.1
0.1-5.1
-4.9-0.1
Macro + BeamFlex
Hetnets key challenge3G Yesterday
3G/4G Tomorrow
17
Size matters
Mass, kg Antenna gain, dBi
8
10
12
14
16
18
20
Company A2
Company W
Company M
Company C
Company A1
18
Longest dimension, cm(includes external antennas)
Source: Product datasheets.
0
2
4
6
8
0 20 40 60 80 100
RuckusZoneFlex 7762
Company W
Company B
Company A1
24 cm(9.4 in)
20 cm(7.9 in)
The real world of small cells is messy
dB
SNR in multipath environment
19
~0.5m
+ many cells = plentiful co-channel interference
Hint: optimize on outcomes
Signal strength is not a reliable indicator of potential radio resource utility
d(SINR)/dx and /dt can be very high
20
very high
Optimizing on immediately realizable utility can have 3x better results in saturated spectrum
Wish list
Operators: support multivendor hetnets for best-in-class performance
TEMs: lets talk...
UEs: fix Wi-Fi handoff, add IPSec+IKEv2
21
Thanks.
More questions?More questions?