1

This Century Challenges:

Embedding the Internet

Deborah Estrin

UCLA Computer Science Department

destrin@cs.ucla.edu

http://lecs.cs.ucla.edu/estrin

2

Enabling Technologies

Embedded Networked

Sensing

Control system w/Small form factorUntethered nodes

ExploitcollaborativeSensing, action

Tightly coupled to physical world

Embed numerous distributed devices to monitor and interact with physical world

Network devices to coordinate and perform higher-level tasks

Exploit spatially and temporally dense, in situ, sensing and actuation

3

Embedded Networked Sensing Potential

• Micro-sensors, on-board processing, and wireless interfaces all feasible at very small scale– can monitor

phenomena “up close”

• Will enable spatially and temporally dense environmental monitoring

• Embedded Networked Sensing will reveal previously unobservable phenomena

Seismic Structure response

Contaminant Transport

Marine Microorganisms

Ecosystems, Biocomplexity

4

“The network is the sensor” (Oakridge Natl Labs)

Requires robust distributed systems of thousands of physically-embedded, often untethered, devices.

5

From Embedded Sensing to Embedded Control

• Embedded in unattended “control systems”– Different from traditional Internet, PDA, Mobility applications that

interface primarily and directly with human users– More than control of the sensor network itself

• Critical applications extend beyond sensing to control and actuation– Transportation, Precision Agriculture, Medical monitoring and drug

delivery, Battlefied applications• Critical concerns extend beyond traditional networked systems

– Usability, Reliability, Safety– Robust interacting systems under dynamic operating conditions– Often mobile, uncontrolled environment,– Not amenable to real-time human monitoring

• Need systems architecture to manage interactions– Current system development: one-off, incrementally tuned, stove-

piped– Serious repercussions for piecemeal uncoordinated design:

insufficient longevity, interoperability, safety, robustness, scalability...

6

PhysicalDistributed

Micro(Embedded Networked Sensing)

Centralized (Traditional Sensor

Systems)

Macro (Shared Scientific

Instruments (telescopes))

Virtual(Internet)

7

New Design Themes

• Long-lived systems that can be untethered and unattended – Low-duty cycle operation with bounded latency– Exploit redundancy and heterogeneous tiered systems

• Leverage data processing inside the network– Thousands or millions of operations per second can be done using energy

of sending a bit over 10 or 100 meters (Pottie00)– Exploit computation near data to reduce communication

• Self configuring systems that can be deployed ad hoc– Un-modeled dynamics of physical world cause systems to operate in ad

hoc fashion – Measure and adapt to unpredictable environment– Exploit spatial diversity and density of sensor/actuator nodes

• Achieve desired global behavior with adaptive localized algorithms– Dynamic, messy (hard to model), environments preclude pre-configured

behavior– Cant afford to extract dynamic state information needed for centralized

control or even Internet-style distributed control

8

Why cant we simply adapt Internet protocols and “end to end” architecture?

• Internet routes data using IP Addresses in Packets and Lookup tables in routers– Humans get data by “naming data” to a search engine– Many levels of indirection between name and IP address– Works well for the Internet, and for support of Person-to-Person

communication• Energy-constrained (un-tethered, small-form-factor),

unattended systems cant tolerate communication overhead of indirection

• Embedded systems can’t rely on human intelligence, elasticity, to compensate for system ambiguities

9

ENS Research Focus

• Critical research needed in “systems”

– Component technology (sensors, low power devices, RF) is far ahead of our ability to exploit

• Must develop, distributed, in-network, autonomous event detection capabilities

– Adaptive Self-Organization to achieve reliable, long-lived, operation in dynamic, resource-limited, harsh environment.

– Collaborative, multi-modal, processing and active database techniques

– Primitives for programming aggregates to create an autonomous, adaptive, monitoring capability across 1000s of nodes

– Sensor coordinated actuation will enable truly self-configuring and reconfiguring systems by allowing for adaptation in physical space

– Safety, Predictability, Usability, particularly as we embed sophisticated behaviors in previously-”simple” objects.

• Strive toward an Architecture and associated principles by building working systems, studying them, iterating

– Analogous to TCP/IP stack, soft state, fate sharing, and eventually, self-similarity, congestion control…

– What is our stack, metrics, system taxonomy…

10

Sample Layered Architecture

Application processing, Distributedquery processing, QOT tradeoffs

Routing

Self-configuring network topology

MAC, Time, Location

Phy: comm, sensing, actuation, SP

User Queries, External Database

Data dissemination, aggregation,storage, caching

11

Metrics

• Efficiency– System lifetime/System resources

• Resolution/Fidelity– Detection/Identification

• Latency– Response time

• Robustness– To variable system and input state]– Security to malicious or buggy nodes

• Scalability– Over space and time

12

Systems Taxonomy: Dimensions

• Spatial and Temporal Scale– Sampling interval– Extent– Density (of sensors relative to stimulus)

• Variability– Ad hoc vs. engineered system structure– System task variability– Mobility (variability in space)

• Autonomy– Multiple sensor modalities– Computational model complexity

• Resource constrained– Energy, BW– Storage, Computation

13

Traffic/Load/Event Models: Dimensions

• Frequency (spatial, temporal)– Commonality of events in time and space

• Locality (spatial, temporal)– Dispersed vs. clustered/patterned

• Mobility– Rate and pattern

14

Constructs for in network processing

• Nodes pull, push, and store named data (using tuple space) to create efficient processing points in the network– e.g. duplicate suppression, aggregation, correlation

• Nested queries reduce overhead relative to “edge processing”• Complex queries support

collaborative signal processing– propagate function

describing desired locations/nodes/data (e.g. ellipse for tracking)

• Interesting analogs to emergingpeer-to-peer architectures

15

Directed Diffusion

• Basic idea– name data (not nodes) with externally relevant attributes

• Data type, time, location of node, SNR, etc– diffuse requests and responses across network using application driven

routing (e.g., geo sensitive or not)– optimize path with gradient-based feedback– support in-network aggregation and processing

• Data sources publish data, Data clients subscribe to data– However, all nodes may play both roles

• A node that aggregates/combines/processes incoming sensor node data becomes a source of new data

• A sensor node that only publishes when a combination of conditions arise, is a client for the triggering event data

– True peer to peer system

• Implemented defines namespace and simple matching rules in the form of filters– Linux (32 bit proc) and TinyOS (8 bit proc) implementations

16

Of more interest than simple Aggregation areNested Queries

(Source: Heidemann et. al.)

Use application-levelinformation to scopeand process data.

user

audio

lightsensors

flat

nested

17

Nested Query Evaluation(A real experiment w/sub-optimal hardware)

• Nested queries greatly improve event delivery rate

• Specific results depend on experiment– placement– limited quality MAC

• General result: app-level info needed in sensor nets; diffusion is good platform

even

ts s

ucce

ssfu

lly r

ecei

ved

(%)

number of light sensors

flat

nested

60

1 2 3 4

80

40

20

18

Sub-optimal aggregation tree constructionsSub-optimal aggregation tree constructions(From Krishnamachari et.al.)(From Krishnamachari et.al.)

• On a general graph if k nodes are sources and one is a sink, the aggregation tree that minimizes the number of transmissions is the minimum Steiner tree. NP-complete

• Center at Nearest Source (CNSDC): All sources send through source nearest to the sink.

• Shortest Path Tree (SPTDC): Merge paths.

• Greedy Incremental Tree (GITDC): Start with path from sink to nearest source. Successively add next nearest source to the existing tree.

• AC: Distinct paths from each source to sink.

19

Source placement: event-radius modelSource placement: event-radius model (From Krishnamachari et.al.)(From Krishnamachari et.al.)

20

Comparison of energy costsComparison of energy costs (From Krishnamachari et.al.)(From Krishnamachari et.al.)

Opportunism always pays;Greed pays only when things get very crowded

(From Intanagowiwat et.al. ns-2 more detailed simulations)

22

Self-Organization with Localized Algorithms

• Self-configuration and reconfiguration essential to lifetime of unattended systems in dynamic, constrained energy, environment– Too many devices for manual configuration

– Environmental conditions are unpredictable• Example applications:

– Efficient, multi-hop topology formation: node measures neighborhood to determine participation, duty cycle, and/or power level

– Beacon placement: candidate beacon measures potential reduction in localization error

• Requires large solution space; not seeking unique optimal • Investigating applicability, convergence, role of selective global

information

23

Adaptive Topology Schemes

• SPANBenjie Chen, Kyle Jamieson, Robert Morris, Hari Balakrishnan, MIT, http://www.pdos.lcs.mit.edu/papers/span:wireless01– Goal: preserve fairness and capacity while providing energy

savings (minimize number of coordinators while still preserving network capacity).

– Mechanism: elects coordinators to create backbone topology.– Limitation: Depends on ad-hoc routing protocol to get list of

neighbors and connectivity matrix between them.• ASCENT

Alberto Cerpa and Deborah Estrin, UCLA, http://lecs.cs.ucla.edu/~cerpa/ASCENT-final-infocom-pdf1.3.pdf– Goal: exploit the redundancy in the system (high density) to save

energy while providing a topology that adapts to the application needs

– Mechanism: empirical adaptation. Each node assesses its connectivity and adapts participation in multi-hop topology based on the measured operating region.

– Limitation

24

Performance Results(From Chen et. al. simulations)

25

Performance Results(From Cerpa, Simulations and Implementation)

Energy Savings (normalized to the Active case, all nodes turn on) as a function of density. ASCENT provides significant amount of energy savings, up to a factor of 5.5 for high density scenarios.

26

Programming Paradigm

• How do we task a 1000+ node dynamic sensor network to conduct complex, long-lived queries and tasks ??

• Map isotherms and other “contours”, gradients, regions– Record images wherever acoustic signatures

indicate significantly above-average species activity, and return with data on soil and air temperature and chemistry in vicinity of activity.

– Mobilize robotic sample collector to region where soil chemistry and air chemistry have followed a particular temporal pattern and where the region presents different data than neighboring regions.

• Pattern identification: how much can and should we do in a distributed manner?

27

Towards a Unified Framework for ENS

• General theory of massively distributed systems that interface with the physical world

– low power/untethered systems, scaling, heterogeneity, unattended operation, adaptation to varying environments

• Programming the Collective

– What local behaviors will result in global tasks

– Programming model for instantiating local behavior and adaptation

– Abstractions and interfaces that do not preclude efficiency

• Large-scale experiments to challenge assumptions behind heuristics

– Measurement tools

– Data sets

28

Pulling it all together

Collaborative Signal

Processing and Active

Databases

Adaptive Self-Configuration

Sensor Coordinated

Actuation

Environmental Microsensors

CENS Core Research Academic Disciplines

NetworkingCommunications

Signal ProcessingDatabases

Embedded SystemsControls

Optimization…

BiologyGeology

BiochemistryStructural Engineering

EducationEnvironmental Engineering

NetworkingCommunications

Signal ProcessingDatabases

Embedded SystemsControls

Optimization…

BiologyGeology

BiochemistryStructural Engineering

EducationEnvironmental Engineering

29

Follow up

• Embedded Everywhere: A Research Agenda for Networked Systems of Embedded Computers, Computer Science and Telecommunications Board, National Research Council - Washington, D.C., http://www.cstb.org/

• DARPA Programs• http://dtsn.darpa.mil/ixo/sensit.asp• http://www.darpa.mil/ito/research/nest/

• Related projects at UCLA and USC-ISI• http://cens.ucla.edu• http://lecs.cs.ucla.edu• http://www.isi.edu/scadds

• Many other emerging, active research programs• UCB: Culler, Hellersein, BWRC, Sensorwebs, CITRIS• MIT: Chandrakasan, Balakrishnan• Cornell: Gherke, Wicker• Univ Washington: Boriello• UCSD: Cal-IT2