Heterogeneous Services And Architectures for NextGeneration Wireless Networks

1
Heterogeneous Services And Architectures for
Next-Generation Wireless Networks
And
Integrating Heterogeneous Wireless Technologies
A Cellular Aided Mobile Ad hoc Network (CAMA)
2
Heterogeneous Services and Architecture for
Next-Generation Wireless Networks

• Simone Frattasi, Rasmus L. Olsen, Frank H.P.
  Fitzek, Ramjee Prasad (Aalborg University,
  Denmark)
• Mauro De Sanctis (University of Rome, Tor
  Vergata)
• http//ctif.aau.dk/

3
Next Generation Networks (NGNs)

• 4G
• MAGIC extension of 3G cellular (cellular
  structure with over 100 mbit/s rates) linear 4G
  vision (Japan)
• Seamless service across multitude of wireless
  systems (non cellular) concurrent 4G version
• Focus on personalized heterogeneous services

4
Wireless Personal Area Network

• Person to Person (P2P) vs. Person to Machine
  (P2M)

5
Services

• Resource Sharing
• Efficiently organize resource consuption (e.g.
  power, bandwidth)

6
Services (cont.)

• Synchronization
• Global (per cell) vs Local (among related
  terminals)
• Local is more flexible (choose local master on
  sign in to a common multicast)
• IP Datacasting
• Data delivery without user awareness ?
• Device to User (D2U) interaction
• User to Device (U2D) interaction
• Digital audio/video broadcasting/WPAN/WLAN/cellula
  r (in off-peak times)
• Video clips, sports highlights, movie trailers,
  partial data delivery as advertisement for data
  service
• Infrastructure Context Manager (CM) and Service
  Manger (SM)

7
Integrating Heterogeneous Wireless Technologies
A Cellular Aided Mobile Ad hoc Network (CAMA)

• Bharat Bhargava, Xiaoxin Wu, Yi Lu, Weichao Wang

8
CAMA

• Use cellular network to improve ad hoc networks
  (like Signaling System 7 (SS7))
• Offer QoS, and AAA (authentication,
  authorization, and accounting) why AAA ??
• Manage the connections of Mobile Ad Hoc Users
  (MTs)
• Only control data goes through cellular base
  stations, all user data is kept in ad hoc network
• Routing is not fixed managed by CAMA

9
CAMA example
10
CAMA can aid in

• Synchronization (offer clock sync)
• Authentication (use cellular infrastructure or
  some other mean like PKI)
• Power saving (estimate transmit power since
  distance between MTs is known)
• Radio resource allocation (manage multiple
  channels in an ad hoc network)
• Broadcasting and multicasting (broadcast through
  cellular base station ??? contradicts their
  claim about all data in contained within ad hoc
  network)
• Finding cluster head in all clustered ad hock
  routing (aid in clustered ad hoc routing that was
  proposed by other research)

11
Feasibility analysis

• Ad hoc cell is small (because the transmit radius
  is small so 1 cell can cover all of the ad hoc
  network)
• UMTS(3G) has sync and broadcast channels (some of
  which could be reserved for CAMA)
• Technology is available and cheap
• Cellular providers try create more data services

12
Previous work

• Heterogeneous integrated wireless networks (how
  ad hoc networks enhance cellular networks)
• Ad hoc routing with positioning
• GPS, GPSR route to peer closest to destination
• VHR position advertisement large overhead
• self positioning when GPS is not available
• ABR - association beacon protocol
• Security
• Open medium, dynamic topology changing,
  cooperative algorithms, lack of centralized
  monitoring
• Key distribution without a CA
• Routing digital signature to authenticate, hash
  chain to insure hop count information
• Architecture for intrusion detection

13
CAMA

• MT has GPS, and position is sent to CAMA through
  cellular base station (also cellular position
  service can be used) a position of each MT is
  well known
• Routing done by CAMA or MT (the study assumes
  that CAMA will coordinate routes between src. and
  dest. this also avoids periodic downlink
  position broadcasts, but slower since an MT might
  wait for CAMA to make its decision, also scale
  considerations

14
CAMA routing

• Position information table per MT
• d (delta) d threshold)

15
Routing cont
N different routes are found for S to T
S-B-H-G-T and S-C-E-I-T 1 (B,C) and (B, C) 2
(F,H,E,D) and (H,E) 3 (G, N, I, K) and (G, I) 4
(L, T, J) and (T) Packet end-to-end delay is used
for ranking (transmission back off
propagation)

16
Routing cont(2)

• Routing steps
• MT sends request to CAMA (resend in case there is
  no response)
• CAMA replies with a complete route including all
  intermediate hops estimated transmission powers
• Routing info is carried in the header of each
  packet
• Position update
• MT sends its position update in a GPS aided
  positioning (if a change is relatively small no
  update is needed)
• Security (relates only to ad hoc network portion)
• CAMA agent can be a central security point for
  key distribution
• Broadcast messages when intrusion is detected
• If the GPS info is false, and there is no means
  to verify it CAMA might fail
• DoS attacks

17
Security cont

• False positioning information
• Try to guess based on cellular position
• Intrusion detection
• MTs can send a failed route message when the
  route did not work and CAMA can try to identify
  the malicious nodes
• Credit system
• Byzantine Behavior
• Encrypt and sign header info
• Report misbehaving nodes
• Message hash verifications
• Anti Jamming
• CAMA can detect the jamming condition and via MT
  reports and make a decision to send client to
  different channel or maybe issue a jamming alert

18
Simulation results

• 100 MTs (1000m X 1000m)
• Each MT moves with random direction and random
  velocity between 0-3m/s (pedestrian speed)
• Ad hoc channel 1mbit/s
• MSGPR is used for routing
• AODV ad hoc on-demand distance vector routing
• DSR dynamic source routing protocol

19
Delivery Ratio
20
More Results
21
More Results (2) .
22
Result cont(3)
23
Results cont(4).
24
Conclusions

• Questions ??
• WiMax
• http//www.networkcomputing.com/showArticle.jhtml?
  articleID191801683
• http//www.engadget.com/2006/08/07/caltrain-commut
  er-rail-pulls-wimax-at-79-mph/
• http//moment.cs.ucsb.edu/AODV/
• http//www.cs.cmu.edu/dmaltz/dsr.html