Mixer Effjcient Many-to-All Broadcast in Dynamic Wireless Mesh - - PowerPoint PPT Presentation

▶
mixer
SMART_READER_LITE
LIVE PREVIEW

Mixer Effjcient Many-to-All Broadcast in Dynamic Wireless Mesh - - PowerPoint PPT Presentation

Nov 21, 2022 299 likes •754 views

Mixer Effjcient Many-to-All Broadcast in Dynamic Wireless Mesh Networks Carsten Herrmann, Fabian Mager, Marco Zimmerling Networked Embedded Systems Lab, TU Dresden, Germany Why Many-to-All Communication? Why Many-to-All universal: can

slide-1
SLIDE 1

Mixer

Effjcient Many-to-All Broadcast in Dynamic Wireless Mesh Networks

Carsten Herrmann, Fabian Mager, Marco Zimmerling Networked Embedded Systems Lab, TU Dresden, Germany

slide-2
SLIDE 2

Why Many-to-All Communication?

slide-3
SLIDE 3

3

universal:

  • can serve any possible traffjc patuern

(one-to-one, one-to-many, all-to-all, etc.) fundamental for a growing number of emerging applicatjons and services:

  • coordinatjon and distributed control

in Cyber-Physical Systems

– factory automatjon – collaboratjve agents, drone swarms – (I)IoT edge

  • programming abstractjons (based on data replicatjon)
  • fault tolerance mechanisms (based on state replicatjon)
  • over-the-air programming / updates, ...

Why Many-to-All

slide-4
SLIDE 4

4

Requirements for Many-to-All

  • fast (10…500 ms end-to-end)
  • reliable
  • support for dynamic mesh topologies
  • support for adequate message sizes (tens of bytes)
  • energy effjcient (weight, cost)

[1] Akerberg et al., Future research challenges in wireless sensor and actuator networks targetjng industrial automatjon, IEEE INDIN 2011

slide-5
SLIDE 5

5

Current Solutjons

  • Multj-Sink Routjng [1]:

– degenerates under high network dynamics

  • Sequentjal Flooding (S-Glossy) [2]:

– suboptjmal scaling O(M · T)

  • Pipelined Flooding [3]:

– not universal (only one-to-all) – infeasible under high network dynamics

[1] e.g. Motuola et al., MUSTER: Adaptjve Energy-Aware Multjsink Routjng in Wireless Sensor Networks, IEEE Transactjons on Mobile Computjng 2011 [2] e.g. Ferrari et al., Effjcient network fmooding and tjme synchronizatjon with Glossy, ACM/IEEE IPSN 2011 [3] e.g. Du et al., When Pipelines Meet Fountain: Fast Data Disseminatjon in Wireless Sensor Networks, ACM SenSys 2015

slide-6
SLIDE 6

6

Our Contributjon

Mixer, a new many-to-all broadcast primitjve for dynamic wireless mesh networks

  • signifjcantly outperforms prior many-to-all solutjons,

approaches order-optjmal scaling O(M + T)

  • provides nearly perfect reliability

despite signifjcant network dynamics

  • supports the full spectrum from one-to-all

to all-to-all communicatjon

slide-7
SLIDE 7

7

Our Contributjon

Mixer, a new many-to-all broadcast primitjve for dynamic wireless mesh networks

  • signifjcantly outperforms prior many-to-all solutjons,

approaches order-optjmal scaling O(M + T)

  • provides nearly perfect reliability

despite signifjcant network dynamics

  • supports the full spectrum from one-to-all

to all-to-all communicatjon Applicatjon Higher Layer Protocol Mixer Physical Layer

slide-8
SLIDE 8

8

init(1)

Our Contributjon

mixer_init(node_id) mixer_write(index, *msg, size) mixer_arm(mode) mixer_start() mixer_read(index) implementatjon available as open source

init(2) init(3)

slide-9
SLIDE 9

8

init(2) write(2,…) write(3,…) init(1) write(1,…) init(3)

Our Contributjon

mixer_init(node_id) mixer_write(index, *msg, size) mixer_arm(mode) mixer_start() mixer_read(index) implementatjon available as open source

slide-10
SLIDE 10

8

init(2) write(2,…) write(3,…) arm() start() init(1) write(1,…) arm() start() init(3) arm() start()

Our Contributjon

mixer_init(node_id) mixer_write(index, *msg, size) mixer_arm(mode) mixer_start() mixer_read(index) implementatjon available as open source

slide-11
SLIDE 11

8

Our Contributjon

mixer_init(node_id) mixer_write(index, *msg, size) mixer_arm(mode) mixer_start() mixer_read(index) implementatjon available as open source

init(2) write(2,…) write(3,…) arm() start() read(1) read(2) read(3) init(1) write(1,…) arm() start() read(1) read(2) read(3) init(3) arm() start() read(1) read(2) read(3)

slide-12
SLIDE 12

Ingredients of Mixer

slide-13
SLIDE 13

10

Ingredients of Mixer

Key Concepts

  • Random Linear Network Coding (RLNC)

→ overlay fmoods

  • Synchronous Transmissions

→ enable capture and spatjal reuse

slide-14
SLIDE 14

11

task: disseminate 3 messages Sequential Flooding (S-Glossy) Linear Network Coding (example) slot 1 1 0 0 | 21 1 0 0 | 21 ... 1 0 0 | 21 1 1 0 | 43 0 1 0 | 22 0 1 1 | 45 0 1 0 | 22 0 0 1 | 23 ... 0 0 1 | 23

  • slot 6

0 0 1 | 23

  • robustness

1 packet 1 packet cost 6 slots 4 slots + computations

Linear Network Coding

slide-15
SLIDE 15

12

Sequentjal Flooding

slide-16
SLIDE 16

12

Sequentjal Flooding

slide-17
SLIDE 17

12

Sequentjal Flooding

slide-18
SLIDE 18

12

Sequentjal Flooding

slide-19
SLIDE 19

12

Sequentjal Flooding

slide-20
SLIDE 20

12

Sequentjal Flooding

slide-21
SLIDE 21

12

Sequentjal Flooding

slide-22
SLIDE 22

12

Sequentjal Flooding

slide-23
SLIDE 23

12

Sequentjal Flooding

slide-24
SLIDE 24

12

Sequentjal Flooding

slide-25
SLIDE 25

13

Mixer

slide-26
SLIDE 26

13

Mixer

slide-27
SLIDE 27

13

Mixer

slide-28
SLIDE 28

13

Mixer

slide-29
SLIDE 29

13

Mixer

slide-30
SLIDE 30

14

Ingredients of Mixer

Key Concepts

  • Random Linear Network Coding (RLNC)

→ overlay fmoods

  • Synchronous Transmissions

→ enable capture and spatjal reuse Effjcient Architecture

  • transport layer with sideload feature
  • deliberate scheduling of (matrix)
  • peratjons

Challenges

  • When should a node send or listen?
  • What should a node send (combine)?
slide-31
SLIDE 31

14

Ingredients of Mixer

Key Concepts

  • Random Linear Network Coding (RLNC)

→ overlay fmoods

  • Synchronous Transmissions

→ enable capture and spatjal reuse Effjcient Architecture

  • transport layer with sideload feature
  • deliberate scheduling of (matrix)
  • peratjons

Smart Policies / Features ("Spices")

  • Semi-Coordinated Transmissions
  • Explicit Innovatjon Forwarding
  • Knowledge-based Startup Behavior
  • Actjve Requests
  • Smart Shutdown
slide-32
SLIDE 32

15

Mixer in Actjon

…

slide-33
SLIDE 33

Evaluation

slide-34
SLIDE 34

17

Setup

  • implementatjon on TelosB

– MSP430, 4 MHz, 10KB RAM, IEEE 802.15.4 – ARM port is in progress…

  • extensive tests on

– FlockLab (ETH Zürich, 27 nodes, 4 hops) – Indriya (NU Singapore, 94 nodes, 8 hops)

slide-35
SLIDE 35

18

Experiments

slide-36
SLIDE 36

19

Reliability

100% Mixer delivered all messages in all FlockLab and Indriya experiments

slide-37
SLIDE 37

20

Experiments

  • general performance: impact of

– message size

slide-38
SLIDE 38

21

Performance all-to-all

Mixer outperforms S-Glossy by up to 3.8x

slide-39
SLIDE 39

22

Experiments

  • general performance: impact of

– message size – number of messages – initjal message distributjon

slide-40
SLIDE 40

23

Performance many-to-all

Mixer is versatjle and scales well

slide-41
SLIDE 41

24

Experiments

  • general performance: impact of

– message size – number of messages – initjal message distributjon

  • impact of node failures
  • impact of node mobility
  • potentjal of faster CPUs and physical layers
slide-42
SLIDE 42

Conclusion

slide-43
SLIDE 43

26

Conclusion

Mixer, a many-to-all broadcast primitjve

  • designed for dynamic wireless mesh networks
  • embeds RLNC and synchronous transmissions into an effjcient architecture

and adds smart policies to make the combinatjon „spicy“

  • supports any initjal message distributjon, i.e., complete spectrum from one-

to-all to all-to-all communicatjon patuerns Visit htups://mixer.nes-lab.org

➔ source code, tutorial projects, documentatjon, … ➔ TelosB (MSP430) available right now, ARM coming soon