Outline Introduction M ulti-Channel Reliability and Spectrum - - PDF document

SLIDE 1

M ulti-Channel Reliability and Spectrum Usage in Real Homes

Empirical Studies for Home-Area Sensor Networks M o S ha, Gregory Hackmann, Chenyang Lu

Department of C

• mputer Science and Engineering

Outline

• Introduction
• Experimental methodology
• Empirical study in homes



Spectrum study of existing wireless signals



802.15.4 link reliability in all 16 channels

• Conclusion

2

Smart Grid

• Home Area Network



Power meters, smart thermostats, home appliances.

• Enables both wired and wireless communication between

utility companies and household devices

3

Wireless Sensor Networks

• Advantage



Do not require wired infrastructure.



Easily and inexpensively retrofit existing homes.



Energy efficiency

• Reliability challenges



Crowded 2.4 GHz IS M band



Unpredictable environment

4

Outline

• Introduction
• Experimental methodology
• Empirical study in homes



Spectrum study of existing wireless signals



802.15.4 link reliability in all 16 channels

• Conclusion

5

M ethodology

• Spectrum usage between 2.400 GHz and 2.495 GHz
• Wi-Spy 2.4x spectrum analyzer



Sweep across the 2.4 GHz spectrum



Sampling period: 40 ms



Signal strength reading on each of 254 discrete frequencies

• Traces over 7 days in 6 apartments and Bryan Hall



Normal daily activities



15,120,000 readings for each of the 254 frequencies



2.5 GB of data per location

• Convert signal strength readings to binary values based on

a threshold



Communication theory:



0: idle channel



1: busy channel

6

SLIDE 2

Spectrum Usage Traces

• Collected from the 2.4 GHz spectrum in six apartments and an office

7

Questions

1. Is there a channel free in all apartments?

• No. There is no “golden” channel.

2. Do homes have similar spectrum usage patterns as offices?

• No. Test in lab is not enough.

3. Does spectrum usage change over time?

• Yes. Channel configuration won’t work.

4. Is 802.11 the dominant interferer in homes?

8

Is Wi-Fi the dominant user of the spectrum?

9

Is Wi-Fi the dominant user of the spectrum?

10

Is Wi-Fi the dominant user of the spectrum?

• While Wi-Fi is a major source of interference, others can

be non-negligible contributors to spectrum occupancy. Channels overlapping with Wi-Fi channel Channels not overlapping with Wi-Fi channel

11

Is Wi-Fi the dominant user of the spectrum?

• While Wi-Fi is a major source of interference, others can

be non-negligible contributors to spectrum occupancy. Channels overlapping with Wi-Fi channel Channels not overlapping with Wi-Fi channel

12

SLIDE 3

Outline

• Introduction
• Experimental methodology
• Empirical study in homes



Spectrum study of existing wireless signals



802.15.4 link reliability in all 16 channels

• Conclusion

13

M ethodology

• Platform



Tmote Sky and TelosB motes



IEEE 802.15.4 compliant Chipcon CC2420 radio



16 channels (11—26) in 5 MHz steps



TinyOS2.1 using default CSMA/ CA M AC layer

• Packet Reception Rate (PRR) of all 802.15.4 channels



10 apartments, 24 hours per apartment



A node broadcast 100 packets per channel to multiple receivers, cycling through all 16 channels in 5 minutes



Receivers recorded the PRRs in onboard Flash

14

Questions

1. Is there a persistently reliable channel? 2. If a good channel cannot be found, are retransmissions sufficient to deal with packet loss? 3. If no single channel can be used for reliable operation, can we exploit channel diversity to achieve reliability? 4. Do channel conditions exhibit cyclic behavior over time? 5. Is reliability strongly correlated among different channels?

15

Is there a persistently reliable channel?

Different links within a same apartment

• Link reliability varies among apartments and links.

Different apartments

16

Is there a persistently reliable channel?

Different links within a same apartment

• Link reliability varies among apartments and links.

Different apartments

17

Is there a persistently reliable channel?

Different links within a same apartment

• Link reliability varies among apartments and links.

Different apartments

18

SLIDE 4

Is retransmission sufficient?

• No, due to burstiness of transmission failures.

19

Is retransmission sufficient?

• No, due to burstiness of transmission failures.

10 % of time, consecutive drops larger than 60

20

• Yes!

Optimal channel hopping schedule Single best channel

Is channel hopping effective?

21

How often needs a link switch channel?

Number of channel hops required under an optimal schedule (one link selected randomly per apartment)

22

• Only a small number of channel hops per day.

Questions

1. Is there a persistently reliable channel? 2. If a good channel cannot be found, are retransmissions sufficient to deal with packet loss? 3. If no single channel can be used for reliable operation, can we exploit channel diversity to achieve reliability? 4. Do channel conditions exhibit cyclic behavior over time?

• No. The channel behavior is not cyclic.

5. Is reliability strongly correlated among different channels?

• Yes. Avoid using adjacent channel.

23

Findings

• Home environments are much more complicated than offices



numerous and diverse Wi-Fi APs and others.

• There is no channel that works for all homes



no channel works for all.

• Channel reliability changes dynamically



cannot pre-select channels.

• Channel hopping is effective



enhance reliability with a few channel switches per day.

SLIDE 5

ARCH: Practical Channel Hopping for Reliable Home-Area Sensor Networks

M o S ha, Gregory Hackmann, Chenyang Lu Department of C

• mputer Science and Engineering

Outline

• Introduction
• Related work
• Protocol design



Design Insights



ARCH Protocol Outline



Coordinated Channel Hopping



Handling Channel Desynchronization

• Evaluation
• Conclusion

26

ARCH Protocol Outline

27

• Receiver-oriented protocol
• M onitor channel condition



M aintain a sliding window of ETX values of incoming links



M ark channel unreliable if ETX values exceed threshold

• Blacklist current channel when channel condition degraded
• Switch to a new picked channel



There is strong correlation among adjacent channels



Uses a probabilistic scheme



Generate a random number



If q falls into the range then, switch to channel i

Coordinated Channel Hopping

28

• Upon selecting a new channel, nodes notify their neighbors
• f this change. Neighbors update their neighbor tables.
• M ulti-hop problem
• M ulti-sender problem

Handling Channel Desynchronization

29

• Default channel



No data transmission, only for resynchronization



Senders use when reach maximum retransmissions



Receivers use when reach maximum waiting time

• False detection



Channel is too noisy



Exchange previous channels when resynchronizing



Blacklist this channel and pick a new one

Outline

• Introduction
• Related work
• Protocol design
• Evaluation



Simulator-Based M icrobenchmarks

* Channel Selection Scheme * Channel Quality Estimator 

Real-World M acrobenchmarks

* Single-Hop Data Collection * M ulti-Hop Data Collection

• Conclusion

30

SLIDE 6

Simulator-Based M icrobenchmarks

31

• Channel Selection Scheme

Simulator-Based M icrobenchmarks

32

• Channel Selection Scheme

Simulator-Based M icrobenchmarks

33

• Channel Quality Estimator

Simulator-Based M icrobenchmarks

34

• Channel Quality Estimator

Real-World M acrobenchmarks

35

• Single-Hop Data Collection

Real-World M acrobenchmarks

36

• Single-Hop Data Collection

SLIDE 7

Real-World M acrobenchmarks

37

• M ulti-Hop Data Collection

Real-World M acrobenchmarks

38

• M ulti-Hop Data Collection

Real-World M acrobenchmarks

39

• M ulti-Hop Data Collection

Real-World M acrobenchmarks

40

• M ulti-Hop Data Collection

Conclusion

• ARCH has the following salient features that distinguish it

from existing channel diversity schemes:



Adaptively channel selection



Distributed approach



Lightweight and robust



M inimal communication overhead

• Results



42.3% decrease in packet retransmissions



17% increase in the proportion of links with perfect delivery rates



2.2X increasing in the minimum delivery rate for the most challenging

• f links



31.6% average reduction in radio usage