Radio Resource Management and QoS based on Power Control

1

Radio Resource Management and QoS based on Power
Control
TASSL Presentation 08/10/01
2
Background

• Radio Resources
• Base stations, channels,transmit power,
  bandwidth, data rate
• Providing QoS for real-time data users require
  delay constraints and minimum through-put.
• Management of radio resources is crucial to the
  efficiency of system operation and congestion
  prevention.
• Radio Resource Management components
• Resource estimation
• Power control
• Rate adaptation
• Link adaptation
• Uplink and Downlink Admission Control

3
RRA
Resource Estimator
Interference conditions
Capacity model
Radio channel characteristics
Allocation figures
Load on base station
Quality requirements
Link Adaptation Instantaneous capacity of
wireless channel is time varying Making the
system capable of adapting itself to the
capacity Improves spectral efficiency
4
Call Admission Control
Admit or block new connections depending on the
current status of the system in terms of loading
and interference. a. Number-based CAC randomly
chooses the new calls based on the current and
maximum number of users that the system could
tolerate per service. b. Interference-based CAC
new call blocked if observed interference level
exceeds a predefined threshold c.
Transmission-power-based CAC accept new calls
based on the terminal power of existing calls
5
Power control and Rate adaptations

• Advantages of power control -
• Optimum power to each signal so that adequate
  quality is achieved
• Reduce battery drain of portables
• Unnecessary interference to other signals
  avoided
• Increase in number of simultaneous users
• Gain in capacity during bad radio channel
  conditions

Case 1 Transmitting power fixed Transmission
rate reduced to meet QoS requirement ( Voice
) Case 2 Transmission suspended during bad
conditions ( Data)
6
Extension of SIM application
Framework to multicast image and text data based
on Imageviewer application(by P.Meer et.al) used
for the setup. Base station connecting to
wireline network (multicast- enabled) and one or
more wireless hosts Base Station part of the
multicast wireline network and unicasts message
to the wireless hosts Wireless hosts capable of
both receiving as well as transmitting data.
Terminal 2
Wireless Terminals
Wired network
Base Station
Terminal 1
7
Network Setup
Joins the multicast conference and waits for any
image to be broadcasted
Base Station
Wired network running text and image sharing
application
Waits for wireless clients to set up connectivity
1.Opens connection with base
Base Station
Wireless Client
2.Sends parameters (Distance, transmitting power
and Transfer rate)
Stores the client parameters.Checks to see if
distance above threshold
Base Station
Wireless Client
If distance above threshold warns client that it
will receive only partial information(image with
less resolution)
8
Utility formulae

• (Based on work by Dr. David Goodman and Prof.
  Mandayam at WINLAB)
• Utility (Definition) -Measure of satisfaction
  experienced by a person
  using a product or service.
• Considered to be zero if signal-to-interference
  ratio (?) is below a certain target level (?o)
• Assumed constant if above the threshold (? gt ?o)

b/J

• Ui

E Total energy expended

• Efficiency function (1 2BERi)M
• M Total bits in each packet including
  information and channel coding bits

9
Calculation
Step 1 Based on the client parameters distance
and power , calculate ? for every client
connected to the base
Step 2 When data is sent to the each client
from base, calculate utility based on total bits
sent over information bits. Step 3 - Calculate
same when it is the client sending the data to
base. (Utility is calculated keeping
retransmission into consideration)
10
Implementation benefit

• We now have the benefit measure from the
  achieved signal strength at the receiver
• Minimum transmission power needed to achieve
  required utility, can be determined and relayed
  to the sender
• Reducing signal power at each station will
  introduce overall reduction of noise factor
• Hence overall power consumption optimized

11
Maximum Utility solution
?Ui
Max utility occurs at -
0
?pi
Applying this to every client will result -gt ?
All terminals transmitting at power
corresponding to this signal-to-interference will
achieve maximum utility. Also reducing power by a
infinitesimal fraction will increase overall
utility.
Prof .Goodman and Prof .Mandayam, "Power
Control for Wireless Data,"IEEE Personal
Communications Magazine, April 2000.
12
Future work

• To show that the optimum SIR value from
  calculations is nearly same with the
  experimental value
• Data correction schemes to improve information
  content at receiver
• Implement other resource management techniques