Teletraffic Performance Analysis of Multiclass OFDMTDMA Systems with AMC

1
Teletraffic Performance Analysis of Multi-class
OFDM-TDMA Systems with AMC

• Hua Wang, Villy B. Iversen

2
Outline

• Introduction
• Previous work
• OFDM Transmission with AMC
• Traffic model analytical model
• Numerical examples
• Conclusion

3
Introduction

• OFDM technology is immune to inter-symbol
  interference and frequency selective fading
• Many standards have adopted OFDM as the air
  interface such as IEEE 802.16
• The economical usefulness of a system is usually
  measured by the Erlang capacity
• Erlang-B formula cant be used in wireless
  networks due to the time-varying nature of
  wireless links
• AMC is proposed to enhance the spectrum
  efficiency
• With AMC, the bandwidth allocation to each user
  is no longer deterministic, but rather dynamic

4
Previous work

• Qingwen L., Shengli Z., and Georgious B. Queuing
  With Adaptive Modulation and Coding Over Wireless
  Links Cross-Layer Analysis and Design
• A finite-length queue is coupled with AMC
• Concentrate on a single-user case
• Tarhini, C., and Chahed, T. System capacity in
  OFDMA-based WiMAX
• Erlang capacity calculation with fixed modulation
  scheme
• Two traffic streams are considered
• Hua Wang, Villy B. Iversen Erlang Capacity of
  Multi-class TDMA Systems with Adaptive Modulation
  and Coding
• Evaluate the Erlang capacity of multi-class TDMA
  systems with AMC by separating the calculation of
  blocking and outage probabilities

5
OFDM Transmission with AMC

• An OFDM-TDMA system with M subchannels
• A frame is divided into K time slots
• Users transmit in the assigned time slots over
  all subchannels
• Channel quality of each subchannel of each user
  is independent identically distributed
• Each subchannel follows a Rayleigh fading
• Received SNR on subchannel m is a random variable
  with pdf

6
OFDM Transmission with AMC (cont.)

• AMC scheme is applied to adjust transmission mode
  according to channel conditions
• N total number of transmission mode (N5)
• Partition the SNR range into N1 intervals
• Algorithm for determining the boundary of each
  interval with prescribed PER is in Ref. 1
• Transmission mode n on subchannel m is chosen
  when
• Transmission mode n on subchannel m will be
  chosen with probability

7
OFDM Transmission with AMC (cont.)

• Rm number of bits can be transmitted over
  subchannel m in one time slot
• R number of bits can be transmitted over all
  subchannels in one time slot

8
Definition of Unit Channel in Multi-class Systems

• L types of service classes
• Each service class has a bit rate requirement of
  r_i bits per frame
• Specify a common channel bandwidth which has
  constant bit rate of r_unit bits per frame
• Number of unit channels required by service class
  i
• Number of time slots occupied by a unit channel
  is a random variable

9
Calculation of Outage Probability

• In AMC, the number of time slots allocated to
  each user is varying frame by frame
• Outage is defined when the total number of
  required time slots exceed the available time
  slots
• Outage probability in state x (x unit channels
  are occupied)
• where

10
State Dependent Acceptance Probability

• A new connection attempt will be accepted with
  certain prob. under the condition that the
  acceptance of the new connection will not cause
  the system outage prob. Above the threshold
• Acceptance prob. of a single-channel call in
  state x
• Acceptance prob. of a d-channel call in state x

11
Traffic model

• BPP traffic model
• Binomial Zlt1, Engset case
• Poisson Z1, Poisson arrival process
• Pascal Zgt1, Pascal arrival process
• Traffic parameters
• Offered traffic A (the average number of call
  attempts per mean service time)
• Peakedness Z (the burstyness of the arrival
  process)
• Number of unit channels needed d

12
Algorithms for Calculating Global State
Probabilities

• The connection-level characteristics of
  multi-class OFDM-TDMA systems with AMC can be
  modelled by a multi-rate loss system with
  state-dependent blocking

13
Algorithms for Calculating Global State
Probabilities (cont.)

• Delbrouck 7 developed a general algorithm for
  calculating the global state probabilities for
  multi-rate loss system with BPP traffic
• Reference 5 extends the Delbroucks algorithm
  to include state-dependent blocking

14
Performance Measures

• Time Congestion
• Traffic Congestion
• Call Congestion

15
Numerical Results

• Consider an OFDM-TDMA system with 5 subchannels
• Total number of time slots for data transmission
  within a frame is 100, each of which contains 4
  OFDM symbols
• Number of transmission mode is set to be 5 with
  target PER at 10e-4
• Target outage probability is 2
• Three types of service classes are considered

16
Numerical Results (cont.)
17
Conclusion

• Developed an analytical model of multi-rate loss
  system with state-dependent blocking to evaluate
  the performance of multi-class OFDM-TDMA with AMC
• The blocking prob. of a call attempt depends both
  on the state of the system and the bandwidth
  requirement
• Due to reversibility, we have local balance and
  may calculate the global state probabilities by
  an effective algorithm

18

• Thank you
• Any questions?