Enabling third-party apps in home gateways for and - - PowerPoint PPT Presentation

▶
enabling third party apps in home gateways for and
SMART_READER_LITE
LIVE PREVIEW

Enabling third-party apps in home gateways for and - - PowerPoint PPT Presentation

Aug 20, 2022 385 likes •855 views

Enabling third-party apps in home gateways for and (through virtualization, edge computing, SDN, and RF management) Suman Banerjee (suman@cs.wisc.edu) 1 IoT and smart home apps IoT and smart home apps Should the cloud manage

slide-1
SLIDE 1

1

Enabling third-party apps in home gateways for and

(through virtualization, edge computing, SDN, and RF management) Suman Banerjee (suman@cs.wisc.edu)

slide-2
SLIDE 2

IoT and smart home apps

slide-3
SLIDE 3

IoT and smart home apps

Should the cloud manage your home thermostats?

slide-4
SLIDE 4

IoT and smart home apps

Should the cloud manage your home thermostats?

slide-5
SLIDE 5

IoT and smart home apps

Should your home security feed be uploaded to the cloud?

slide-6
SLIDE 6

IoT and smart home apps

Should your home security feed be uploaded to the cloud?

slide-7
SLIDE 7

IoT and smart home apps

Can we watch 4K streaming movies?

slide-8
SLIDE 8

IoT and smart home apps

Can we watch 4K streaming movies?

Network variability

slide-9
SLIDE 9

What is ParaDrop?

slide-10
SLIDE 10

What is ParaDrop?

Edge computing in the extreme Move the cloud right into your home

slide-11
SLIDE 11

What is ParaDrop?

Edge computing in the extreme Providing a cloud-like abstraction in your home WiFi router

slide-12
SLIDE 12

3rd party apps/services drop into your home WiFi router

  • n-demand

WiFi router has low latency to home devices, and is always-on

slide-13
SLIDE 13

ParaDrop cloud manager SeeCam cloud service SeeCam

How it works?

slide-14
SLIDE 14

ParaDrop cloud manager SeeCam cloud service SeeCam EnvSense EnvSense cloud service

How it works?

slide-15
SLIDE 15
  • Transcode video to adapt to

wireless channel conditions

ParaDrop examples

  • Cache movies in router from

head of instant queue

  • Vehicular: Passenger tracking

and engagement

  • Home vent control
slide-16
SLIDE 16

More examples …

  • Kids router
  • Framily router
  • TOR in your router
  • Monitor your ISP (a la BisMark)
  • <Your idea here …>
slide-17
SLIDE 17

Let us know if you need one! (We can ship you one or more) Check out www.paradrop.io

slide-18
SLIDE 18

ParaDrop smart router

  • A programmable substrate
  • Virtualization framework
  • “Chutes”
  • Isolated and proprietary
  • Multiple wireless interfaces
  • WiFi, Bluetooth, ZigBee
  • Cloud-based RF management
slide-19
SLIDE 19

19

Commodity AP Hardware

Cloud-based Wireless Management

Overall architecture

SDN

Virtualization and edge computing

slide-20
SLIDE 20

20

Commodity AP Hardware

Cloud-based Wireless Management

Overall architecture

SDN

Virtualization and edge computing

slide-21
SLIDE 21

Problem statement

Wireless users

Experience: “seems to work well mostly” “flaky”, “connection keeps dropping”

“Is anybody else finding the network to be slow/flaky? I want to sanity check before I tell lab admin ” “It was flaky for me yesterday. Restarting the laptop seemed to solve the problem”

[thread from user mailing list]:

Expectation: wire-like performance

slide-22
SLIDE 22

What is the cause of interference?

Problem statement

slide-23
SLIDE 23

Problem statement

“We continued to hear reports of problems, but could not reproduce the problems "on demand" nor do we have enough details to identify the real issues.”

[email from wireless network admin]:

Network administrators

Problem: issues are transient, lack of details and tools

“Is my wireless network operating well enough?”

slide-24
SLIDE 24

A “blood test” for the network

(Comment made by Pravin Bhagwat: CTO, Airtight Networks)

slide-25
SLIDE 25

Basic question

Can we provide a real-time,

  • verall health score for the

entire network?

Access points Clients

WLAN score BAD GOOD

L1 L2 L3 L4 L5 L6

slide-26
SLIDE 26

Basic question

WLAN score BAD GOOD

Goodness of the link Links

BAD GOOD

L1 L2 L3 L4 L5 L6

Identify why certain links are under-performing

weak signal Impact: 0.7 Impact: 0.3 Impact: 0.1 Impact: 0.2 L6

slide-27
SLIDE 27

RF Management goals

Key questions:

  • Detect: Who are the real interferers?
  • Quantify: How much interference do they cause?
  • Locate: How do we locate these non-WiFi devices?
  • Mitigate: What configuration needed to mitigate them?

Cloud controllers

slide-28
SLIDE 28

(use WiFi-only hardware!)

  • COLLIE (Collision Inferencing Engine)

Real-time Interference estimators

Our solution

WiFi to WiFi Interference

  • PIE (Passive Interference Estimator)

Non-WiFi to WiFi Interference

  • Airshark
  • WiFiNet
  • 1. quantify interference impact
  • 2. pin point device location
  • 1. detect non-WiFi devices
slide-29
SLIDE 29

Spectrum at a university cafe

High powered non-WiFi devices share the spectrum with WiFi devices

29

slide-30
SLIDE 30

Is non-WiFi interference a real problem?

More than 50% throughput drop and, in some cases, throughput drops to zero! Across locations, many devices frequently appeared with a high signal strength How severe is the impact? How prevalent are the devices?

30

slide-31
SLIDE 31

Analog Cordless Phone (Uniden EXP4540) Video Camera (Pyrus Electronics) FHSS Cordless Phone (Panasonic KT-TG2343) Bluetooth ACL/SCO (iPhone, iPod touch, Jabra headsets) Game controllers (Wii, PS3) Audio Transmitter (GoGroove PurePlay) Microwave Ovens (Whirlpool, Daewoo, Sunbeam)

High duty devices Frequency hoppers Broadband interferers

ZigBee (Jennic JN 5121)

Non-WiFi interferers

slide-32
SLIDE 32

32

Time Frequency

22 MHz 312 kHz Sub-carriers O(103) to O(104) samples/sec

Coarse-grained

  • freq. resolution

Coarse-grained time resolution Limited spectrum view RSSI/power per sub-carrier

Use a coarse-grained WiFi lens

slide-33
SLIDE 33

Spectral Sampling Pulse Detection Feature Extraction

Extract Features Decision Tree Classifier Training / Classification (OFF) (ON)

Device Detection Pipeline DT-classifiers

33

Airshark: how it works

slide-34
SLIDE 34

Detection Accuracy

> 98% accuracy at signal strengths >= -80 dBm

34

slide-35
SLIDE 35

Many trials

35

Including integrating with a commercial off-the-shelf AP platform

slide-36
SLIDE 36

Works with

  • We built Linux drivers for spectral sampling for

the AR92xx and AR93xx NICs

  • Open sourced and publicly available
  • Have also released an Android version

– For Samsung Nexus devices

36

Check wisense.io

Available from Google Play

slide-37
SLIDE 37

37

Cloud-based RF Management

COAP: Coordination framework for Open Access Points

slide-38
SLIDE 38

38

Cloud-based RF Management

COAP: Coordination framework for Open Access Points

slide-39
SLIDE 39

39

Interference estimation and mitigation techniques Analytics

Cloud-based RF Management

COAP: Coordination framework for Open Access Points

slide-40
SLIDE 40

40

Analytics

Spectrum Management Commands

Cloud-based RF Management

Interference estimation and mitigation techniques

COAP: Coordination framework for Open Access Points

slide-41
SLIDE 41

41

Analytics

Cloud-based RF Management

Spectrum Management Commands

Interference estimation and mitigation techniques

COAP: Coordination framework for Open Access Points

slide-42
SLIDE 42

42

  • OpenWRT based APs

– ALIX 2d2 platform: (500 MHz AMD Geocode CPU, 256 DDR RAM, flash storage)

  • 30+ APs deployed in homes

& apartment complexes for 3+ years

  • Cloud controller hosted in
  • ff-the-shelf Linux server

Field Trials

MDU Deployment 1 14 APs Madison, WI MDU Deployment 2

slide-43
SLIDE 43

Microware cuts throughput

43

slide-44
SLIDE 44

Data is sorted

25.5 13 30.1 16 100% improvement most of the time

slide-45
SLIDE 45

45

Enabling third-party apps in home gateways for and

Suman Banerjee (suman@cs.wisc.edu)