EN.600.450.01.FA11 Network Embedded Systems/Sensor...

EN.600.450.01.FA11

Network Embedded Systems/Sensor Networks

Week 1: Introduction and Applications

Marcus Chang @ CS JHU

with help from Andreas Terzis and Prabal Dutta

1

Administrivia

2

Tuesday, Thursday 1:30-2:45 Shaffer 202

Instructor: Marcus Chang (mchang AT cs.jhu.edu)

Teaching Assistant: Zainan Victor Zhou (zzn AT jhu.edu)

General course consulting and help with programming assignments

Class website: http://hinrg.cs.jhu.edu/joomla/classes.html

Class material will also be on: http://blackboard.jhu.edu/

Office Hours: By appointment (Shaffer 200)

Embedded System

“An embedded system is a computer system designed to do one or a few dedicated and/or specific functions“

“It is embedded as part of a complete device often including hardware and mechanical parts”

“By contrast, a general-purpose computer, such as a personal computer (PC), is designed to be flexible and to meet a wide range of end-user needs”

- Wikipedia

3

E.g. Environmental Monitoring: Sample sensors

Store measurements

Process data

Forward results

Wireless Sensor Networks

4

Course Outcomes

Learn about embedded systems and wireless networks Program microcontrollers

Program Android cell phones

Learn about low-power system design Cross-platform optimization

Learn how to (start to) do research Critically review the literature

Learn how to discover and formulate problems

Learn about experiment design, interpreting and presenting results

5

Prerequisites

CS344 or other networking course, Operating Systems, systems programming

Must feel comfortable programming in C

Grad student attitude (self-directed, deal with open ended problems/assignments, highly motivated, critical, dedicated)

6

What will this class be like?

This is partly a seminar course We will be reading and discussing papers (10%)

Class attendance and participation mandatory

Approximately 2-3 papers per week

Tuesdays, topic overview (lecture)

Thursdays, 1-2 students present a paper and lead discussion (10%)

All students hand in a review

2 programming assignments (25%) Introduction to programming embedded systems

Big project (30%)

Work in teams of 2 people

Midterm, but no final!

7

Readings and Reviews

You are responsible for completing assigned readings before lecture

Submit the review through BlackBoard

Review is due before beginning of lecture and should include

At least two thought-provoking questions on the assigned paper

Discussion of any strengths and weaknesses

Two possible directions for extensions on the ideas / topic presented in the paper.

Your questions should critically evaluate the paper (e.g., questioning the assumptions, questioning whether the experiments are lacking (and why), flaws in the analysis, etc).

You are allowed to miss two reviews but send email to let me know

8

Assignments and Project

Implement body sensor network Transmit data from wireless sensors to Android phone

Store measurements in the cloud

Visualize measurements on Android phone and the web

Assignments Android programming (JAVA)

Wireless sensor programming (TinyOS, nesC)

Project Cross-platform optimization

Implement your own device driver (memory mapped IO, C)

Progress reports, give presentation, write final report

9

Moore’s Law:

IC transistor count doubles every two years

10 Photo Credit: Intel

Flash memory cost

11

12

Hendy’s “Law”:

Pixels per dollar doubles annually

Credit: Barry Hendy/Wikipedia

13

Bell’s Law, revised:

Corollary to the Laws of Scale

UMich Phoenix Processor Introduced 2008

Initial clock speed

106 kHz @ 0.5V Vdd Number of transistors

92,499 Manufacturing technology

0.18 µ

Photo credits: Intel, U. Michigan

14

MEMS Accelerometers:

Rapidly falling price and power

[Analog Devices, 2009] ADXL345

10 µA @ 10 Hz @ 6 bits

25 µA @ 25 Hz

[ST Microelectronics, annc. 2009]

O(mA)

15

Energy harvesting and storage

Thermoelectric Ambient

Energy Harvester [PNNL]

Shock Energy Harvesting

CEDRAT Technologies

Electrostatic Energy

Harvester [ICL]

Thin-film batteries

Piezoelectric

[Holst/IMEC]

RF Energy harvesting [Powercast]

Current Generation Mote: Epic Core

Microcontroller CPU: 8 MHz RAM: 10K RAM Program: 48K ROM Energy consumption:

6 mW (active) 60 uW (sleep mode)

Radio

250 kbps Energy consumption:

60 mW

16

Why Wireless Sensor Networks?

Sensors, computation, and communication feasible at small packages and large numbers

Ability to monitor phenomena at fine spatial AND temporal granularities

Vision: Embedded Networked Sensing will reveal previously unobservable phenomena

A new type of scientific instrument: macroscope

17

Example

Arctic lake monitoring

Remote and harsh environment

Manual water sampling expensive and time consuming

18

Wireless sensor network

In-situ sensing

Wireless data collection

19

20

The System Challenge

applications

service

network

system

architecture

data mgmt

Monitoring & Managing Spaces and Things

technology

MEMS

sensing

Power

Comm. uRobots

actuate

Miniature, low-power connections to the physical world

Proc

Store

21

[Ganti06]

A decade of sensor network applications

Mobility High Low

Low

High

Power

[Liu04]

[Tolle05]

[Werner-Allen06]

[Szewczyk04]

[Jiang09]

[Hull06]

[Malinowski07]

[Wark07]

[Abdelzaher07]

[Thiele08]

[Lifton07]

[Lorincz08]

[Aoki09]

SOLVED

ACTIVE ACTIVE

EMERGING

22

Mobility makes energy and communication challenges

fundamentally harder in low-power systems

Energy Must carry it along Or harvest it from the ambient

environment And deal with inherent

uncertainty of harvesting Implies dynamic duty cycles How to measure usage?

Link Topological turmoil Link. What link? Never before seen link What radio channel? When, where to look? Can’t just probe during

deployment History is a poor/no guide

Network Stability of routing peers? Routers, hosts, both?

Transport Disruption-tolerant Store-and-forward

“Weather + mobility = uncertain energy budget”

- Jacob Sorber, Sensys 2007

J. Sorber et al., “Eon: A Language and Runtime for Perpetual Systems”, Sensys’07, Sydney, Australia

23

The fight goes on

Summary of Technical Challenges

24

Energy consumption

Low-power wireless

Resource constraints

Dealing with the physical world

Unattended operation

Coordination

Schedule

25

Week 1: Introduction and Applications

Week 2: Mote Hardware

Week 3: Embedded Programming

Week 4: Medium Access Control

Week 5: Link Estimation and Tree Routing

Week 6: IP Networking

Week 7: Time Synchronization

Week 8: Energy Management

Week 9: Review and Midterm

Week 10: Operating Systems and Programming Languages

Week 11: Advanced Networking Topics

Week 12: Localization

Week 13: Energy Harvesting

Week 14: TBD