Offloading Floating Car Data through V2V Communications Razvan - - PowerPoint PPT Presentation

▶
offloading floating car data
SMART_READER_LITE
LIVE PREVIEW

Offloading Floating Car Data through V2V Communications Razvan - - PowerPoint PPT Presentation

Feb 04, 2024 583 likes •849 views

Offloading Floating Car Data through V2V Communications Razvan Stanica (INSA Lyon), Marco Fiore (IEIIT - CNR), Francesco Malandrino (Politecnico di Torino) 3mes Journes Nationales des Communications dans les Transports (JNCT) Nevers - 30 May

slide-1
SLIDE 1

Offloading Floating Car Data through V2V Communications

Razvan Stanica (INSA Lyon), Marco Fiore (IEIIT - CNR), Francesco Malandrino (Politecnico di Torino) 3èmes Journées Nationales des Communications dans les Transports (JNCT) Nevers - 30 May 2013

slide-2
SLIDE 2

 Floating Car Data  Mobility Trace  Optimal Gain  Degree-based Mechanism  Degree-based with Confirmation  Reservation-based Mechanism

1 Offloading Floating Car Data 30.05.2013 Razvan Stanica INSA Lyon JNCT 2013

slide-3
SLIDE 3

FCD

 Up to 100 Electronic Control Units in a car  Large amount of data produced, but not stored  Possible applications:

  • Fleet management
  • Remote diagnostic
  • Customized car insurance
  • Collaborative urban sensing
  • Road traffic monitoring
  • Navigation systems

2 Offloading Floating Car Data 30.05.2013 Trace Optimal Gain DB DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Floating Car Data

slide-4
SLIDE 4

FCD

 Up to 100 Electronic Control Units in a car  Large amount of data produced, but not stored  Possible applications:

  • Fleet management
  • Remote diagnostic
  • Customized car insurance
  • Collaborative urban sensing
  • Road traffic monitoring
  • Navigation systems

2 Offloading Floating Car Data 30.05.2013 Trace Optimal Gain DB DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Floating Car Data

To enable such applications, data needs to be centralized and mined

slide-5
SLIDE 5

FCD

 Tom Tom HD Traffic, Meihui TrafficCast, Peugeot Connect Apps …

3 Offloading Floating Car Data 30.05.2013 Trace Optimal Gain DB DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Current model

slide-6
SLIDE 6

FCD

 Tom Tom HD Traffic, Meihui TrafficCast, Peugeot Connect Apps …

3 Offloading Floating Car Data 30.05.2013 Trace Optimal Gain DB DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Current model

  • Very expensive: price limits

the frequency of FCD collection

  • The advantage of the low

penetration ratio vs. Market share and application success

slide-7
SLIDE 7

FCD

 Tom Tom HD Traffic, Meihui TrafficCast, Peugeot Connect Apps …

3 Offloading Floating Car Data 30.05.2013 Trace Optimal Gain DB DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Current model

  • Very expensive: price limits

the frequency of FCD collection

  • The advantage of the low

penetration ratio vs. Market share and application success FCD can represent a serious challenge for both cellular

  • perators (network capacity) and

service providers (collection price)

slide-8
SLIDE 8

FCD

 Use of Vehicle-to-Vehicle communication

4 Offloading Floating Car Data 30.05.2013 Trace Optimal Gain DB DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Offloading FCD

slide-9
SLIDE 9

FCD

 Use of Vehicle-to-Vehicle communication

4 Offloading Floating Car Data 30.05.2013 Trace Optimal Gain DB DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Offloading FCD

  • Local gathering and, perhaps,

fusion of FCD: reduced network load, reduced cost, more data

slide-10
SLIDE 10

FCD

 Use of Vehicle-to-Vehicle communication

4 Offloading Floating Car Data 30.05.2013 Trace Optimal Gain DB DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Offloading FCD

  • Local gathering and, perhaps,

fusion of FCD: reduced network load, reduced cost, more data Addressed questions: What is the possible gain brought by offloading FCD? Is there a simple, practical mechanism for FCD offload?

slide-11
SLIDE 11

FCD

 Largest freely available vehicular mobility dataset  24h on a 400km2 region in the Köln area: more than 700k trips  Synthetic trace respecting real macroscopic measures  OpenStreetMap, SUMO, TAPAS, Gawron’s relaxation  For more information, see:

  • http://kolntrace.project.citi-lab.fr/
  • S. Uppoor, O. Trullols-Cruces, M. Fiore, J.M. Barcelo-Ordinas, Generation and

Analysis of a Large-scale Urban Vehicular Mobility Dataset, IEEE TMC 2013 5 Offloading Floating Car Data 30.05.2013

Trace

Optimal Gain DB DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Vehicular mobility trace

slide-12
SLIDE 12

FCD

 Each second in the trace results in a connectivity graph  Two vehicles separated by less than R meters are considered connected (this does not imply a unit disk graph radio propagation model)  This allows the detection of important properties: assortative network

6 Offloading Floating Car Data 30.05.2013

Trace

Optimal Gain DB DB-C RB Razvan Stanica INSA Lyon JNCT 2013

V2V connectivity

slide-13
SLIDE 13

FCD

 Gain measured as percentage of vehicles that do not use the cellular uplink  The FCD offloading problem maps to a classical Minimum Dominating Set problem  MDS is NP-hard, but bounded approximation algorithms exist

7 Offloading Floating Car Data 30.05.2013 Trace

Optimal Gain

DB DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Optimal gain

More than 60% of the FCD can be

  • ffloaded through V2V

communication in more than 70 %

  • f the cases
slide-14
SLIDE 14

FCD 8 Offloading Floating Car Data 30.05.2013 Trace

Optimal Gain

DB DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Impact of daytime

 Up to 90% gain in the time intervals 6:30am-9am and 3:30pm-7pm  This second time interval maps to the daily peak in cellular traffic

slide-15
SLIDE 15

FCD 9 Offloading Floating Car Data 30.05.2013 Trace

Optimal Gain

DB DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Impact of geographical area

 The gain depends on the vehicular density, so it is not evenly distributed  Some areas present a limited gain throughout the entire day, while a 95% gain can be achieved at peak hours in the central region

slide-16
SLIDE 16

FCD 10 Offloading Floating Car Data 30.05.2013 Trace Optimal Gain DB DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Looking for practical solutions

 Centralized MDS already NP-hard  Many distributed algorithms proposed for backbone construction in wireless sensor networks  Trade-off between the quality of the MDS approximation and the amount of needed communication  Best algorithms find a dominating set after multiple communication rounds (a different message can be exchanged with each neighbor during a round)  Convergence time too long for a dynamic vehicular network

slide-17
SLIDE 17

FCD 10 Offloading Floating Car Data 30.05.2013 Trace Optimal Gain DB DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Looking for practical solutions

 Centralized MDS already NP-hard  Many distributed algorithms proposed for backbone construction in wireless sensor networks  Trade-off between the quality of the MDS approximation and the amount of needed communication  Best algorithms find a dominating set after multiple communication rounds (a different message can be exchanged with each neighbor during a round)  Convergence time too long for a dynamic vehicular network We relax the MDS condition, and focus on convergence time

slide-18
SLIDE 18

FCD Trace Optimal Gain

DB

DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Degree-based mechanism

11 Offloading Floating Car Data 30.05.2013

 Every vehicle recovers the useful FCD from its neighbors through regular V2V beaconing (already standardized, transmitted on the control channel)  A vehicle uses the cellular uplink with probability

parameter of the mechanism node degree

 DB does not recover FCD from all the vehicles  Ratio of uncovered nodes:

probability that a node has degree d

slide-19
SLIDE 19

FCD Trace Optimal Gain

DB

DB-C RB Razvan Stanica INSA Lyon JNCT 2013

Degree-based mechanism

12 Offloading Floating Car Data 30.05.2013

k=1 k=2

 With k=1, we obtain the same gain as for the MDS, but only 70% of vehicles are covered  With k=2, 90% of the vehicles are covered, but the offloading gain decreases

slide-20
SLIDE 20

FCD Trace Optimal Gain DB

DB-C

RB Razvan Stanica INSA Lyon JNCT 2013

Degree-based with confirmation

13 Offloading Floating Car Data 30.05.2013

 A simple confirmation mechanism for 100% coverage: a node that uses the cellular uplink announces this to its neighbors  Nodes uncovered at the end of a collection period transmit FCD on the cellular uplink  Maps to a cost optimization problem:

nodes with less than k neighbors always transmit nodes transmitting at the end of the collection period nodes relaying FCD for their neighbors

slide-21
SLIDE 21

FCD Trace Optimal Gain

DB

DB-C RB Razvan Stanica INSA Lyon JNCT 2013 14 Offloading Floating Car Data 30.05.2013

 The optimal k value varies, but the differences between the gain obtained with an optimal k and a fixed k=2 are minimal  A 20% difference compared with the optimal gain can be noticed at peak hours

Degree-based with confirmation

slide-22
SLIDE 22

FCD Trace Optimal Gain DB DB-C

RB

Razvan Stanica INSA Lyon JNCT 2013

Reservation-based mechanism

15 Offloading Floating Car Data 30.05.2013

 Try to improve the performance of DB-C, while keeping 100% coverage  Reservation period at the beginning of every collection period  Each vehicle randomly chooses a slot to transmit a reservation message  All the vehicles receiving a reservation message cancel their back-off  In the ideal scenario the RB mechanism reaches the same solution as a centralized MDS heuristic  In reality: limited number of slots, message collision, radio propagation problems

slide-23
SLIDE 23

FCD Trace Optimal Gain DB DB-C

RB

Razvan Stanica INSA Lyon JNCT 2013

Reservation-based mechanism

16 Offloading Floating Car Data 30.05.2013

 The performance of the mechanism quickly increases with the number of slots  The gain matches the one obtained with a centralized MDS algorithm even in realistic scenarios

slide-24
SLIDE 24

Razvan Stanica INSA Lyon JNCT 2013 17 Offloading Floating Car Data 30.05.2013

Conclusion

 Floating Car Data is becoming an important challenge for operators, providers and users  We show that using V2V communication to offload FCD can bring important benefits  Three simple mechanism are discussed, and their performance indicates that FCD offloading is feasible in practice  Still a lot of work to do: penetration ratio, actual gain on the cellular network, actual gain for service providers, different offloading mechanisms…

slide-25
SLIDE 25

Offloading Floating Car Data through V2V Communications

Razvan Stanica (INSA Lyon), Marco Fiore (IEIIT - CNR), Francesco Malandrino (Politecnico di Torino) 3èmes Journées Nationales des Communications dans les Transports (JNCT) Nevers - 30 May 2013