Bluetooth Scatternet Formation Algorithms

1
BluetoothScatternet Formation Algorithms

• Fazel Keshtkar
• http//www.site.uottawa.ca/akesh081

Prof Dr. Boukerche http//www.site.uottawa.ca/bo
ukerch/ March 31,2004
2
Outline

• Problem Statement
• Introduction Bluetooth Architecture
• Scatternet Formation Algorithm
• Performance Evaluation
• Conclusion
• References

3
Problem Description

• Increase the Communication Rang
• Increase number of BDs involve in a system.
• Localized maintenance
• How does a allocation of isolated devices from a
  Scatternet
• Dynamically either leaving or moving as well as
  join the Bluetooth Scatternet.

4
Introduction

• The Bluetooth Technology designed with the
  purpose of replacing cabling with neighbor
  devices to
• connect computers and mobile phones to external
  devices and
• accessories via wireless links.

5
Introduction

• Rapidly deploy wireless infrastructures using low
  cost
• Easily available devices such as PADs, Notebooks
  and cellular phones,,,,

6
Introduction

• Scatternet is a set of piconets connected
  through shared devices
• The base Bluetooth network infrastructure is
  represented by a Piconet
• that is formed by up to 8 Bluetooth Devices
  (BDs) actively participating in the communication

7
Introduction

• one of which has the role of master while the
  others act as salves.
• The communication range of a BD is around
  10-30mt.
• for some devices up to 100mt.
• this range is the maximum piconet radius.

8
Introduction

• The Bluetooth specification
• Includes the possibilty of building scatternets
• Allow to increase the communication
• Range and the number of BDs involved in a system
• Bluetooth is an emerging low-cost and low-power
  short-range radio technology

9
Introduction

• Bluetooth network is composed of piconets.
• Each piconet contain one master and up to seven
  slaves.
• Piconet can be connected into a large Scatternet
  by sharing Slaves.

10
Bluetooth Architecture
M
S
S
L2CAP
M
S
Host Controller Interface
S M
S S
LMP
Baseband layer
S
S
Bluetooth Radio
(b)
(a)
(a) Bluetooth Architecture, (b) Example of a
scatternet infrastructure.
11
Bluetooth Architecture

• L2CAP(Logical Link Control and Adaptation)
• The network layer services are provided by L2CAP
• The basband and LMP accessed through the Host
  Control Interface(HCI).
• LMP(Link Manager Protocol) Allows to perform the
  BDs, link and packets configuration.

12
Bluetooth Architecture

• the Piconet is built in two steps
• The Inquiry phase
• The Page phase
• In the Inquiry Phase, a BD may discover
  neighbours by sending broadcast messages(inquiry
  procedure).
• the Page phase, the master assigns to each
  active slave a 3-bit Active Member Address
  AM-ADDR

13
Bluetooth Scatternet Formation Algorithms

• 1-BlueTree Algorithm
• use for large-scale Bluetooth network
• Bluetree state connecting all its neighbours as
  slaves
• the Bluetree slaves act as masters for their
  neighbours and so on recursively.

14
Bluetooth Scatternet Formation Algorithms

• Bluetree introduces two protocol for scatternet
  formation
• First is based on a designated node
• The second seeds up by selecting more than one
  root for tree formation
• and then merging the trees generated by each
  root.

15
Bluetooth Scatternet Formation Algorithms

• The protocol is organized in two phase
• First phase
• a subnet of the nodes will be selected as init
  nodes
• Second Phase
• the protocol merges the generated sub-trees into
  one Scatternet

16
Bluetooth Scatternet Formation Algorithms

• 2-Bluenet Algorithm
• Bluenet buils scatterent by initially aggregating
  the BDs into piconets involving
• at most Nmax slaves.
• in Bluenet Nmax5.

17
Bluetooth Scatternet Formation Algorithms

• Bluenet scheme forms the scatternet based on the
  following rules
• Rule-1 Avoid forming through piconets inside a
  piconet
• Rule-2 For a bridge node, avoid setting up more
  than one connections to the same piconet.
• Rule-3 Inside a piconet, the master tries to
  aquire some number of slaves

18
Bluetooth Scatternet Formation Algorithms

• Bluenet algorithm is organized into 3-phase
• Phase-1 Initial piconet formed with some
  separate Bluenet nodes left.
• Phase-2 Separate Bluenet nodes get connected to
  initial piconets.
• Phase-3 Piconets get connected to form a
  sactternet.

19
Bluetooth Scatternet Formation Algorithms

• 3-BTCP Algorithm( Bluetooth Topology Construction
  Protocol)
• BTCP has more flexibility in constructing the
  scatternet.
• Not suitable for dynamic environments
• where devices can join and leave after the
  scatternet is formed.

20
Bluetooth Scatternet Formation Algorithms

• BTCP is a 3-phase algorithms
• First phase
• A coordinator is elected with a complete
  knowledge of all devices
• Seconde phase
• the coordinator choose other p-1 masters
• form p piconets, and p(p-1)/2 bridges

21
Bluetooth Scatternet Formation Algorithms

• Third Phase
• forms a completely connected the topology among
  the piconets in the third phase.
• This coordinator determines and takes other
  masters how a scatternet should be formed.
• If the coordinate fails, the formation protocol
  has to be restored.

22
Bluetooth Scatternet Formation Algorithms

• 4-LMS Algorithm
• a Distributed Randomized algorithm
• "New Blutooth Scatternet Formation Algorithm

23
Bluetooth Scatternet Formation Algorithms

• How does a collection. of isolated devices form a
  scatternet?
• each device is not aware of the other devices.
• Therefore, the scatternet formation protocol must
  be distributed.
• devices are in the communication range of each
  other.
• Thus, potentially, any two devicescan be
  connected directly.

24
Bluetooth Scatternet Formation Algorithms

• LMS algorithm has to two major performance
  measure
• Time Complexity Amount of time to form a
  Scatternet.
• LMS Scatternet has O(Logn) Time Complexity, n is
  number of BDs.
• Message Complexity Number of messages sent
  between the devices.
• Message complexity is O(n).

25
Bluetooth Scatternet Formation Algorithms

• 5-TSF Algorithm (Tree Scatternet Formation
  Algorithm)
• TSF characterized by a Tree Scatternet
• By building a forest of piconets
• allows nodes to start communicating while the
  sactternet is under construction.
• In TSF, nodes and BDs dynamically leaving and
  joining the scatternet.
• can not merge two sactternet if their roots are
  out of range.

26
Bluetooth Scatternet Formation Algorithms

• TSF, has the following properties
• Conectivity TSF rapidly converges toward a
  steady-sate in which all nodes can reach
• each other, at any time.
• Healing TSF avoiding loops and healing network
  partition.

27
Bluetooth Scatternet Formation Algorithms

• 6-ODBT Algorithm(On-Demand Bluetooth Scatternet
  Formation
• ODBT characterize a communication connecting a
  set of BDs in a tree topology
• scatternet formation is started by an initiator
  node,
• This node becomes the tree root

28
Bluetooth Scatternet Formation Algorithms

• the other BDs are progressively
• grafted to the structure
• ODBT can, support out-of-range
• allows the data forwarding during the tree
  construction
• minimize the number of piconets by maximizing the
  number of BDs involved in each piconet

29
Performance Evaluation
Table 2. Comparison among scatternet formation
algorithms.
30
Conclusion

• ODBT builds efficient tree structures
• minimizing the number of bridges connecting the
  piconets
• ODBT shows a stable behaviour for increasing
  number of involved BDs.
• Able to dynamically reconfgure the scatternet to
  deal with joining, leaving and moving BDs,
• support out-of-range devices.

31
References

• 1. Law C., Mehta A.K., Siu K.-Y. A New
  Bluetooth Scatternet Formation Protocol. Proc.
  IEEE Global Telecommunications Conference
  (GLOBECOM01), 2001, 28642869.
  http//perth.mit.edu/ ching/pubs/ScatternetProtoco
  l.pdf.
• 2. E. Pagano, G. Paolo Rossi, Stefano Tebaldi
  An On-Demand Bluetooth Scatternet Formation
  Algorithm, WONS 2004, LNCS 2928
  pp.130-143,2004,IFIP International Federation for
  Information Processing 2004

32
References

• 3. Tan G., Miu G., Guttag J., Balakrishnan H.
  An E.cient Scatternet Formation Algorithm for
  Dynamic Environments. IASTED International
  Conference on Communications and Computer
  Networks (Nov. 2002)
• 4. Wang Z., Thomas R.J., Haas Z. Bluenet - a
  new scatternet formation scheme. Proceedings of
  the 35th Annual Hawaii International Conference
  on System Sciences (Jan. 2002).
  http//wnl.ece.cornell.edu/Publications/hicss02.ps
  .
• 5. Zaruba G.V., Basagni S., Chlamtac I.
  Bluetrees-scatternet formation to enable
  Bluetooth-based ad hoc networks. Proc. IEEE
  International Conference on Communications (ICC
  2001), 2001, 273277.
• 6. Zhifang Wang, Robert J. Thomas, Zygmunt Haas
  Bluenet a New Scatternet Formation Scheme