![Page 1: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/1.jpg)
Lecture 14: Bringing Connectivity “Everywhere”
Some material adapted from Deepak Vasisht (MIT/MSR)
6.S062: Mobile and Sensor Computing
Aerial-based Connectivity & Agriculture IoT
![Page 2: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/2.jpg)
2
Goal: Bringing Connectivity to the Remote and Disconnected Areas of the Planet
Google X’s Project Loon Facebook’s Project Aquila
High Interest in Aerial-based Connectivity
Others including Microsoft, Boeing, etc.
![Page 3: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/3.jpg)
• Bring connectivity to rural areas
• Disaster Relief
3
Goal: Bringing Connectivity to the Remote and Disconnected Areas of the Planet
![Page 4: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/4.jpg)
Challenges • Power: Constrained
• Need to last for a long time
• Control: Flight paths • Minimal power consumption
• Communications: Long-range links
• Data Rates4
Solar Energy
![Page 5: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/5.jpg)
Challenges • Power: Constrained
• Need to last for a long time
• Control: Flight paths • Minimal power consumption
• Communications: Long-range links
• Data Rates5
Solar Energy
• Stratosphere • Drone paths
Low Frequencies
• 10s MHz bandwidth • Millimeter waves
![Page 6: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/6.jpg)
Common Opportunities: Atmospheric Conditions and Predictability • Leverage Stratosphere in Loon/
Aquila • No “problematic” weather
conditions (rain, winds, etc.)
• Different stratospheric layers have different predictable currents
• Thermodynamics for changing levels in stratosphere
6
![Page 7: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/7.jpg)
FarmBeats: An IoT System for Data-Driven Agriculture
NSDI 2017
![Page 8: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/8.jpg)
Why Agriculture?Agricultural output needs to double by 2050 to meet the demands
– United Nations1
Popu
lati
on (
Billi
ons)
0
2
5
7
9
1950 1975 2000 2025 20501: United Nations Second Committee (Economic & Financial), 2009
8
![Page 9: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/9.jpg)
Why Agriculture?Agricultural output needs to double by 2050 to meet the demands
– United Nations1
Popu
lati
on (
Billi
ons)
0
2
5
7
9
1950 1975 2000 2025 2050
But… • Water levels are receding • Arable land is shrinking • Environment is being degraded
91: United Nations Second Committee (Economic & Financial), 2009
![Page 10: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/10.jpg)
Why Agriculture?Agricultural output needs to double by 2050 to meet the demands
– United Nations
Popu
lati
on (
Billi
ons)
0
2
5
7
9
1950 1975 2000 2025 2050
Number of World’s Hungry People
10
![Page 11: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/11.jpg)
Solution: Data-Driven Agriculture
Ag researchers have shown that it: • Reduces waste • Increases productivity • Ensures sustainability
11
Traditional vs Data-driven approach
![Page 12: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/12.jpg)
But…
According to USDA, high cost of manual data collection prevents farmers from using data-driven agriculture
12
![Page 13: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/13.jpg)
IoT System for Agriculture
13
![Page 14: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/14.jpg)
Problem 1: No Internet Connectivity • Most farms don’t have any internet coverage
• Even if connectivity exists, weather related outages can disable networks for weeks
14
![Page 15: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/15.jpg)
Problem 2: No Power on the Farm• Farms do not have direct power sources
• Solar power is highly prone to weather variability
15
![Page 16: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/16.jpg)
Problem 3: Limited Resources • Need to work with sparse sensor deployments
• Physical constraints due to farming practices
• Too expensive to deploy and maintain
16
![Page 17: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/17.jpg)
Beyond Agriculture
How can one design an IoT system in challenging resource-constrained environments?
Mining Oil Fields
17
![Page 18: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/18.jpg)
Rest of this lecture• FarmBeats: An end-to-end IoT system that enables seamless
data collection for agriculture
18
FarmBeats Farm Services
![Page 19: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/19.jpg)
Rest of this lecture• FarmBeats: An end-to-end IoT system that enables seamless
data collection for agriculture
• Solves three key challenges: • Internet Connectivity • Power Availability • Limited Sensor Placement
• Deployed in two farms in NY and WA for over six months
19
![Page 20: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/20.jpg)
Challenge: Internet Connectivity
(Farmer’s home/office) Cloud
20
![Page 21: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/21.jpg)
Challenge: Internet Connectivity
(Farmer’s home/office) Cloud
Sensors • Few miles away • Obstructed by crops, canopies, etc
21
![Page 22: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/22.jpg)
Approach: Use TV White Spaces
22
• Can provide long-range connectivity
• Can travel through crops and canopies, because of low frequencies
• Large chunks are available in rural areas=> can support large bandwidth
![Page 23: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/23.jpg)
Idea: Use TV White Spaces
(Farmer’s home/office)
Base Station
TV White Spaces
Cloud
Few miles
Sensors
• Weak Connectivity • Prone to outages
23
Wi-Fi, BLE
![Page 24: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/24.jpg)
Approach: Compute Locally and Send Summaries• PC on the farm delivers time-sensitive services locally
• Combines all the sensor data into summaries
• 2-3 orders of magnitude smaller than raw data
• Cloud delivers long-term analytics and cross-farm analytics
24
![Page 25: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/25.jpg)
FarmBeats Design
Gateway PC (Farmer’s home/office)
Base Station
TV White Spaces
Cloud
Few miles
Sensors25
![Page 26: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/26.jpg)
In this lecture • FarmBeats: An end-to-end IoT system that enables seamless
data collection for agriculture
• Solves three key challenges: ✓Internet Connectivity • Limited Sensor Placement • Power Availability
• Deployed in two farms in NY and WA for over six months
26
![Page 27: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/27.jpg)
Challenge: Limited Resources• Need to work with sparse sensor deployments
• Physical constraints due to farming practices
• Too expensive to deploy and maintain
• How do we get coverage with a sparse sensor deployment?
27
![Page 28: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/28.jpg)
Approach: Use Drones to Enhance Spatial Coverage• Drones are cheap and automatic
• Can cover large areas quickly
• Can collect visual data
28
Combine visual data from the drones with the sensor data from the farm
![Page 29: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/29.jpg)
Idea: Use Drones to Enhance Spatial Coverage
Sparse Sensor Data
Precision MapPanoramic OverviewDrone Video
29
![Page 30: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/30.jpg)
Formulate as a Learning Problem
Training Data
Panoramic Overview
Prediction
30
![Page 31: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/31.jpg)
Model Insights• Spatial Smoothness: Areas close to each other
have similar sensor values
• Visual Smoothness: Areas that look similar have similar sensor values values
31
![Page 32: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/32.jpg)
Model: Gaussian Processes
Features (visual)
Kernel (Model visual similarity)
Output (say, moisture)
Spatial Smoothness
• Training Phase: Learn K and W
• Test Phase: Generate outputs for unknown areas
![Page 33: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/33.jpg)
Using Sparse Sensor Data
Sensor Data
Precision MapPanoramic OverviewDrone Video100 kB summary
FarmBeats can use drones to expand the sparse sensor data and create summaries for the farm
33
![Page 34: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/34.jpg)
In this talk• FarmBeats: An end-to-end IoT system that enables seamless
data collection for agriculture
• Solves three key challenges: ✓Internet Connectivity ✓Limited Sensor Placement • Power Availability
• Deployed in two farms in NY and WA for over six months
34
![Page 35: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/35.jpg)
Challenge: Power Availability is Variable
Gateway (Farmer’s home/office)
Farm
TV White Spaces
Cloud
Battery dies due to cloudy/rainy/snowy
weather
35
![Page 36: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/36.jpg)
Challenge: Power Availability is Variable• Solar powered battery saw up to 30% downtime in cloudy
months
• Miss important data like flood monitoring
36
How do we deal with weather-based power variability?
![Page 37: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/37.jpg)
Approach: Weather is Predictable
• Use weather forecasts to predict solar energy output
• Ration the load to fit within power budget
37
![Page 38: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/38.jpg)
Idea: Weather is Predictable•
38
![Page 39: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/39.jpg)
Solution: Weather is predictable
0
5
10
15
20
0 1.3 2.5 3.8 5
Optimal for minimum latency
FarmBeats can use weather forecasts to duty cycle the
base station, with minimum latency
39
![Page 40: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/40.jpg)
How would you design the sensors?• Low-power — backscatter
• problems: intermittent, or base station runs out of power • Limited range
• Semi-passive?
• Power decays with 1/d^2 (Sphere) => waste less energy by multiple harvesters
• Can even harness power from whitespace emissions
40
![Page 41: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/41.jpg)
In this lecture• FarmBeats: An end-to-end IoT system that enables seamless
data collection for agriculture
• Solves three key challenges: ✓Internet Connectivity ✓Limited Sensor Placement ✓Power Availability
• Deployed in two farms in NY and WA for over six months
41
![Page 42: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/42.jpg)
Deployment• Six months deployment in two farms: Upstate NY
(Essex), WA (Carnation) • The farm sizes were 100 acres and 5 acres
respectively • Sensors:
• DJI Drones • Particle Photons with Moisture, Temperature, pH
Sensors • IP Cameras to capture IR imagery as well as
monitoring
• Cloud Components: Azure Storage and IoT Suite42
![Page 43: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/43.jpg)
Deployment Statistics• Used 10 sensor types, 3 camera types and 3 drone versions
• Deployed >100 sensors and ~10 cameras
• Collected >10 million sensor measurements, >0.5 million images, 100 drone surveys
• Resilient to week long outage from a thunderstorm
43
![Page 44: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/44.jpg)
FarmBeats: Usage
Gateway (Farmer’s home/office)
Farm
TV White Spaces
Cloud
44
![Page 45: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/45.jpg)
Example: Panorama
Water puddle Cow excreta Cow Herd Stray cow
45
![Page 46: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/46.jpg)
Precision Map: Panorama Generation
46
![Page 47: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/47.jpg)
Precision Map : Moisture
47
![Page 48: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/48.jpg)
Precision Map : pH
48
![Page 49: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/49.jpg)
Precision Map: Accuracy
Mea
n Er
ror
0
0.25
0.5
0.75
1
Temp (F) pH (0-14) Moist (0-6)
FarmBeatsLeastCount
FarmBeats can accurately expand coverage by orders of magnitude using a sparse sensor deployment
49
![Page 50: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/50.jpg)
Weather-Aware Duty CyclingCl
oud
Cove
r (%
)
0
23
45
68
90
Day
0 0.75 1.5 2.25 3Ba
tter
y %
0
25
50
75
100
Day
0 0.75 1.5 2.25 3
No Duty Cycling
50
![Page 51: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/51.jpg)
Weather-Aware Duty CyclingCl
oud
Cove
r (%
)
0
23
45
68
90
Day
0 0.75 1.5 2.25 3
FarmBeats Duty Cycling
Batt
ery
%
0
25
50
75
100
Day
0 0.75 1.5 2.25 3
Reduced downtime from 30% to 0% for month long data (September)
51
![Page 52: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/52.jpg)
![Page 53: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/53.jpg)
Other Related Works• Wireless Sensor Networks: Sensor networks for agriculture
(Baggio `05, Sanchez et al `11, Lee et al `10,…), LPWAN technologies (LoRA, SIGFOX, …)
• Agriculture: Precision agriculture (Bratney et al `99, Mueller et al `12, Cassman et al `99,..), Nutrient measurement (Kim et al `09, Hanson et al `07)
• ICTD: Information access and user interfaces (Zhao et al `10, Doerflinger et al 2012)
53
![Page 54: 6.S062: Mobile and Sensor Computing · Google X’s Project Loon Facebook’s Project Aquila High Interest in Aerial-based Connectivity Others including Microsoft, Boeing, etc. •](https://reader034.vdocument.in/reader034/viewer/2022042418/5f3486921ff2c129f77a98ca/html5/thumbnails/54.jpg)
Summary• Aerial-based Connectivity (Loon, Aquila) & Agriculture IoT
• Challenges: Power, Control, Communication Range, Bandwidth, Weather
• Opportunities: Duty cycling, sparse sampling, weather prediction, thermodynamics, learning and sensor fusion, Drones
• Farmbeats: End-to-end IoT system for Farming
54