bringing life to dead spots grace woo pouya kheradpour, dawei shen, and dina katabi
DESCRIPTION
Poor Coverage Is Not No Coverage! X X Can recover a correct packet if we combine the correct bits from these receptions Spatial Diversity APs are unlikely to have same bit error Loss Currently High Bit Error Persistent Loss Dead SpotTRANSCRIPT
Bringing Life to Dead SpotsGrace Woo
Pouya Kheradpour, Dawei Shen, and Dina Katabi
Many APs But Still Poor Coverage
Problem increases with mobility and low power devices
mit1
mit4
mit5
Poor Coverage Is Not No Coverage!
010101011111X 011101011011X
Can recover a correct packet if we combine the correct bits from these receptions
Spatial Diversity APs are unlikely to have same bit error
Loss Loss
Currently High Bit Error Persistent Loss Dead Spot
But Which AP Got the Right Bit?
• Clearly can’t have per bit checksum
• Prior work (MRD) tries all block combinations to satisfy checksum
• Exponential Complexity• Works for a few bit errors But not dead spots
First bit is “0”First bit is “1”
SOFT• Recovers a correct packet from its
faulty receptions at APs
• Leverages physical layer hints to identify correct bits
• SOFT’s delivery rate is up to10x higher than current WLANs and MRD
SOFT Architecture
APs leverage high-speed Ethernet to combine their receptions
Internet
AP1 AP2
SOFT Combiner
APn
Wired Ethernet
But which bits are correct?
010001001111X010101001011X
010001001011
Physical Layer Knows More!
PHY already estimates a confidence in its 0-1 decision Soft Value
PHY 01011001
0.0
0.2
0.4
0.6
0.8
1.0
-1.0 -0.5 0.0 0.5 1.0
Measured Soft Values
PDF of per bit soft values
Larger absolute soft values More confidence in bit
Soft Value < 0 “0”
Soft Value > 0 “1”
We Use the Soft Values• SOFT changes the PHY interface to expose the soft values to higher layers• SOFT combines the soft values of a bit to decode it correctly
• The combiner forwards the decoded packet if it satisfies the 802.11 checksum
AP1AP2
SOFT Combiner
Soft packet Soft packet
010110111
How Do We Combine Soft Values?
0.4 - 0.1 - 0.2
How do we decode the bit?• Maximum soft value Bit is “1”• Majority vote Bit is “0”• Average Bit is “1”
Say for a particular bit, we got
Different Combining Methods Different Answers!
SOFT Combining Algorithm
"0" else "1" is Bit 02 jS
ii
ij
2iLet be the noise variance on the channel to
APi
For AWGN and dead spots rule is proven optimal.
Let Sij be the soft value of bit j reported by APi
SOFT decision rule:
Intuitively, we want to favor less noisy channels
But, How Does SOFT Get the Noise Variance?
Randomness in soft values is caused by channel noise
Estimate from the PDF of the soft values in packet2i
0.0
0.2
0.4
0.6
0.8
1.0
-1.0 -0.5 0.0 0.5 1.0
Measured Soft Values
PDF of per bit soft values
How About Overhead? • PHY soft values can be 32-bit float
Excessive Ethernet trafficSolution• Invoke SOFT only when associated
AP can’t decode• Quantize soft values (we used 3 bits)
What About the Downlink?
Use Time Diversity Combine a packet with its retransmission
010001001011X010101000011X
Performance
SOFT Implementation• Software – GNURadio
codebase• Hardware – USRP frontend• GMSK and DBPSK
modulations• Soft values are inputs to
the slicer• Poor Coverage:
• SNR 5 – 12 dB• BER about 10-3
Experimental Setup• 13 GNURadio nodes• Compared
– Current 802.11 WLAN (user associates with best AP)
– MRD– SOFT
• Each Experiment– 3 random APs – Random source – Transmit 500 packets
Does SOFT Help?CDF of 100 experiments
Packet Delivery Rate
Does SOFT Help?CDF of 100 experiments
Packet Delivery Rate
Does SOFT Help?CDF of 100 experiments
Packet Delivery Rate
Does SOFT Help?CDF of 100 experiments
Packet Delivery RateSOFT’s delivery rate can be 10x higher
10x
Packet Delivery Rate
Bit Error Rate
Performance with Increasingly Poor Coverage
Packet Delivery Rate
Bit Error Rate
Performance with Increasingly Poor Coverage
Current Approach
Packet Delivery Rate
Bit Error Rate
Current ApproachMRD
Performance with Increasingly Poor Coverage
Packet Delivery Rate
Bit Error Rate
Current Approach
SOFTMRD
SOFT Addresses Dead Spots
Performance with Increasingly Poor Coverage
0.00
0.20
0.40
0.60
0.80
1.00
0 5 10 15 20 25 30
Current Approach
CDF over 50,000 packets
Number of Retransmissions Until Correct Packet
SOFT on Downlink
0.00
0.20
0.40
0.60
0.80
1.00
0 5 10 15 20 25 30
Current ApproachSOFT
CDF over 50,000 packetsSOFT on Downlink
17 ReTx
Number of Retransmissions Until Correct PacketMuch Higher Throughput!
Combining Method Is Important
CDF of 100 experiments
Packet Delivery Rate
MAX
SOFTMajority
SOFT Outperforms MAX and MAJORITY
Effect of Quantization SOFT Average Delivery Rate
3 Bits 32 Bits
2 Bits
All presented results are for 3-bit quantization!Overhead on Wired Ethernet is Acceptable
Related Work• Soft and softer handoff in cellular
networks• Theoretical Maximum Ratio Combining
(MRC) [Brennan55,Yang99]• H-ARQ & Chase Combining [ASX03]• Partial Packet Recovery [Jam07]
Conclusion• WLAN can have better coverage if the
interface to the PHY exposes soft values
• Delivery rate can be up to 10x higher
• Ethernet overhead is acceptable
• The new architecture, SOFT, can co-exist with unmodified 802.11 cards and APs