adaptive cleaning for rfid data streams

26
Adaptive Cleaning Adaptive Cleaning for RFID Data for RFID Data Streams Streams Presented by Willie and Abhishek Shawn Jeffery Minos Garofalakis Michael Franklin UC Berkeley Intel Research Berkeley UC Berkeley Disclaimer: The slides are taken from Jeffrey’s talk at VLDB ‘06

Upload: juliet

Post on 25-Feb-2016

22 views

Category:

Documents


1 download

DESCRIPTION

Shawn Jeffery Minos Garofalakis Michael Franklin UC Berkeley Intel Research Berkeley UC Berkeley . Adaptive Cleaning for RFID Data Streams. Presented by Willie and Abhishek. Disclaimer: The slides are taken from Jeffrey’s talk at VLDB ‘06. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Adaptive Cleaning for RFID Data Streams

Adaptive Cleaning for Adaptive Cleaning for RFID Data StreamsRFID Data Streams

Presented by Willie and Abhishek

Shawn Jeffery Minos Garofalakis Michael Franklin UC Berkeley Intel Research Berkeley UC Berkeley

Disclaimer: The slides are taken from Jeffrey’s talk at VLDB ‘06

Page 2: Adaptive Cleaning for RFID Data Streams

04/22/23

RFID: Radio Frequency RFID: Radio Frequency IDentificationIDentification

Page 3: Adaptive Cleaning for RFID Data Streams

04/22/23

RFID data is dirtyRFID data is dirtyShelf 0 Shelf 1

RFIDReaders

StaticTags

Mobile Tags

15ft1.5ft

3ft9ft

3ft

3ft

3ft

A simple experiment:•2 RFID-enabled shelves•10 static tags•5 mobile tags

Page 4: Adaptive Cleaning for RFID Data Streams

04/22/23

RFID Data CleaningRFID Data Cleaning

Time

Raw readings

Smoothed output

• RFID data has many dropped readings• Typically, use a smoothing filter to

interpolateSELECT distinct tag_idFROM RFID_stream [RANGE ‘5 sec’]GROUP BY tag_idBut, how to set the size

of the window?

Smoothing Filter

Page 5: Adaptive Cleaning for RFID Data Streams

04/22/23

Window Size for RFID Window Size for RFID SmoothingSmoothing

Fido moving Fido resting

Small windowRealityRaw readings

Large window

Need to balance completeness vs. capturing tag movement

Page 6: Adaptive Cleaning for RFID Data Streams

04/22/23

Truly Declarative Truly Declarative SmoothingSmoothing

• Problem: window size non-declarative• Application wants a clean stream

of data• Window size is how to get it

• Solution: adapt the window size in response to data

Page 7: Adaptive Cleaning for RFID Data Streams

04/22/23

ItineraryItinerary

• Introduction: RFID data cleaning• A statistical sampling perspective• SMURF

• Per-tag cleaning• Multi-tag cleaning

• Ongoing work• Conclusions

Page 8: Adaptive Cleaning for RFID Data Streams

04/22/23

A Statistical Sampling A Statistical Sampling PerspectivePerspective

• Key Insight: RFID data random sample of present tags

• Map RFID smoothing to a sampling experiment

Page 9: Adaptive Cleaning for RFID Data Streams

04/22/23

RFID’s Gory DetailsRFID’s Gory Details

Epoch TagID ReadRate0 1 .90 2 .60 3 .3

Tag 1

Tag 2

Tag 3

Tag 4

Antenna & readerTags

E1 E2 E3 E4 E5 E6 E7 E8 E9E0

Read Cycle (Epoch)

(For Alien readers)

Tag List

Page 10: Adaptive Cleaning for RFID Data Streams

04/22/23

RFID Smoothing to SamplingRFID Smoothing to Sampling

RFID SamplingRead cycle (epoch) Sample trialReading Single sampleSmoothing window Repeated trialsRead rate Probability of inclusion

(pi)

Now use sampling theory to drive adaptation!

Page 11: Adaptive Cleaning for RFID Data Streams

04/22/23

SMURFSMURF• Statistical Smoothing for Unreliable RFID

Data• Adapts window based on statistical

properties• Mechanisms for:

• Per-tag and multi-tag cleaning

Multi-tagCleaning

SMURFPer-tag

Cleaning

raw RFID streams

cleanedcount readings

cleanedper-tag readings

Application(s) Application(s)

Page 12: Adaptive Cleaning for RFID Data Streams

04/22/23

Per-Tag Smoothing: Per-Tag Smoothing: Model and BackgroundModel and Background

• Use a binomial sampling model

Time (epochs)

pi

1

0

Smoothing Window

wi Bernoulli trials

piavg

Si

(Read rate of tag i)

E1 E2 E3 E4 E5 E6 E7 E8 E9E0

Page 13: Adaptive Cleaning for RFID Data Streams

04/22/23

Per-Tag Smoothing: Per-Tag Smoothing: CompletenessCompleteness

• If the tag is there, read it with high probability

Want a large window

pi

1

0

Reading with a low pi

Expand the window

Time (epochs)E1 E2 E3 E4 E5 E6 E7 E8 E9E0

Page 14: Adaptive Cleaning for RFID Data Streams

04/22/23

Per-Tag Smoothing: Per-Tag Smoothing: CompletenessCompleteness

Expected epochs needed to read

With probability 1-

Desired window size for tag i

1ln*1

avgi

ip

w

Page 15: Adaptive Cleaning for RFID Data Streams

04/22/23

Per-Tag Smoothing: Per-Tag Smoothing: TransitionsTransitions• Detect transitions as statistically

significant changes in the data

pi

1

0

Statistically significant difference Flag a transition and

shrink the window

The tag has likely left by this point

Time (epochs)E1 E2 E3 E4 E5 E6 E7 E8 E9E0

Page 16: Adaptive Cleaning for RFID Data Streams

04/22/23

Per-Tag Smoothing: Per-Tag Smoothing: TransitionsTransitions

# expected readings Is the difference

“statistically significant”?# observed

readings

)1(**2|*||| avgi

avgii

avgiii ppwpwS

Page 17: Adaptive Cleaning for RFID Data Streams

04/22/23

SMURF in ActionSMURF in ActionFido moving Fido resting

SMURF

Experiments with real and simulated data show similar results

Page 18: Adaptive Cleaning for RFID Data Streams

04/22/23

Multi-tag CleaningMulti-tag Cleaning

• Some applications only need aggregates• E.g., count of items on each shelf Don’t need to track each tag!

• Use statistical mechanisms for both:• Aggregate computation • Window adaptation

Page 19: Adaptive Cleaning for RFID Data Streams

04/22/23

Aggregate Aggregate ComputationComputation

• –estimators (Horvitz-Thompson) • Count:

• P[tag i seen in a window of size w]:

Use small windows to capture movementUse the estimator to compensate for lost

readings

wSiwN

1

wavgii p )1(1

Page 20: Adaptive Cleaning for RFID Data Streams

04/22/23

Window AdaptationWindow Adaptation

• Upper bound window similar to per-tag

• “Transition” based on variance within subwindows

1ln*1

avgpw

CountNw

Nw’

Time (epochs)E1 E2 E3 E4 E5 E6 E7 E8 E9E0

'VarVar2ww NN

Page 21: Adaptive Cleaning for RFID Data Streams

04/22/23

Multi-tag ScenarioMulti-tag Scenario

Page 22: Adaptive Cleaning for RFID Data Streams

04/22/23

Ongoing Work: Spatial Ongoing Work: Spatial SmoothingSmoothing

• With multiple readers, more complicated

Reinforcement

A? B? A U B? A B?Arbitration

A? C? All are addressed by statistical framework!

U

A

B

C

D

Two rooms, two readers per room

Page 23: Adaptive Cleaning for RFID Data Streams

04/22/23

Beyond RFIDBeyond RFID

• -estimator for other aggregates Use SMURF for sensor networks

• Use SMURF in general streaming systems (e.g., TelegraphCQ)

Remove RANGE clause from CQL

Other sensor data

Other streaming data

Page 24: Adaptive Cleaning for RFID Data Streams

04/22/23

Related WorkRelated Work

• Commercial RFID middleware• Smoothing filters: need to set smoothing

window• RFID-related work

• Rao et al., StreamClean: complementary• Intel Seattle, HiFi, ESP: static window size

• BBQ, MauveDB• Heavyweight, model-based• SMURF is non-parametric, sampling-based

• Statistical filters (digital signal processing)• Non-linear digital filters inspired SMURF design

Page 25: Adaptive Cleaning for RFID Data Streams

04/22/23

ConclusionsConclusions

• Current smoothing filters not adequate• Not declarative!

• SMURF: Declarative smoothing filter• Uses statistical sampling to adapt window size

Page 26: Adaptive Cleaning for RFID Data Streams

04/22/23

Thanks!Thanks!

Questions?