Spectrum trading in cognitive radio networks: A marketequilibriumbased approach

1
Spectrum trading in cognitive radio networks A
market-equilibrium-based approach

• Advisor Wei-Yeh Chen
• Student ? ? ?
• Reference
• D. Niyato , E. Hossain, Spectrum trading in
  cognitive radio networks A market-equilibrium-bas
  ed approach, IEEE Wireless Commun., vol. 15, no.
  6, pp. 71 - 80 , Dec. 2008 .

2
Outline

• Introduction
• Spectrum sharing and spectrum trading
• Spectrum trading Structure
• Spectrum trading Related issues
• Spectrum trading Solution approaches
• System model
• Equilibrium in spectrum sharing and pricing
• Expectation and learning
• Conclusion

3
Introduction(1/2)

• Frequency spectrum is the scarcest(???) radio
  resource in wireless communication networks.
• The concept of cognitive radio was introduced to
  improve the frequency spectrum utilization in
  wireless networks

4
Introduction(2/2)

• We introduce a market-equilibrium-based spectrum
  trading mechanism that uses spectrum demand and
  supply of the primary and secondary users,
  respectively(???).
• Since spectrum supply is stochastic(???) in
  nature, a distributed and adaptive learning
  algorithm is used for the secondary users to
  estimate(??) spectrum price and adjust the
  spectrum demand accordingly so that the market
  equilibrium can be reached.

5
Spectrum sharing and spectrum trading(1/2)

• Two major steps in spectrum sharing are spectrum
  exploration(??) and spectrum exploitation(??).
• The objectives of spectrum exploration are to
  discover and maintain(??) the statistics(??) of
  spectrum usage, and identify the spectrum
  opportunities.

6
Spectrum sharing and spectrum trading(2/2)

• Spectrum trading is the process of exchanging
  spectrum, which can be performed based on the
  exchange of different resources or money.
• Spectrum trading determines the structure of
  radio resource selling and buying.
• Pricing is a major issue in spectrum trading that
  determines the value of the spectrum to the
  spectrum seller and buyer.

7
Spectrum trading Structure

• Single Seller (Monopoly) The simplest structure
  of spectrum trading arises when there is only a
  single seller in the system.
• Multiple Sellers (Oligopoly) This market
  structure consists of multiple sellers offering
  radio spectrum to the market.
• No Permanent Seller (Exchange Market) In this
  market structure there is no permanent(???)
  spectrum seller, and all users have the right to
  access the spectrum.

8
Spectrum trading Related issues

• Spectrum Pricing
• Spectrum Supply and Cost of Spectrum Sharing
• Utility Function and Spectrum Demand
• Competition and Cooperation in Spectrum Sharing

9
Spectrum pricing

• Price plays an important role in spectrum trading
  since it indicates the value of spectrum to both
  the seller and buyer.
• For the buyer, the price paid to the spectrum
  seller would depend on the satisfaction achieved
  through the usage of that spectrum.
• For the spectrum seller, the price determines its
  revenue(??).

10
Spectrum supply and cost of spectrum sharing

• In a cognitive wireless system this spectrum
  supply can be in terms of the number of frequency
  channels, the number of time slots , or transmit
  power given the price charged to the buyer.
• There are two types of cost Fixed cost is
  incurred due to the investment(??) in
  infrastructure, variable cost is incurred due to
  performance degradation(??) resulting fro
  sharing/selling the spectrum.

11
Utility function and spectrum demand

• In spectrum trading, spectrum demand determines
  the amount of spectrum the buyer wants to access
  for a given price so that its satisfaction is
  maximized.
• The spectrum demand function can be derived(??)
  based on maximization of utility of secondary
  users for a given price.

12
Competition and cooperation in spectrum
sharing(1/2)

• A competition occurs when each of the cognitive
  radio entities has its self-interest and is
  rational(???) about maximizing its own benefit.
• Competition can be among multiple spectrum
  sellers to attract(??) more buyers or among
  spectrum buyers to obtain the best
  quality/quantity(?) of spectrum at the lowest
  possible price.

13
Competition and cooperation in spectrum
sharing(2/2)

• Entities involved(??) in spectrum trading may
  have a choice to cooperate so that a better
  solution can be achieved.
• The sellers can cooperate to choose higher prices
  so that they earn a profit higher than that in
  case of competition

14
Spectrum trading Solution approaches

• Microeconomic Approach
• Classical Optimization Approach
• Noncooperative Game
• Bargaining(??) Game
• Auction(??)

15
Microeconomic approach

• The solution of this approach is based on market
  equilibrium, which denotes(??) a price for which
  spectrum demand equals spectrum supply.
• At a market equilibrium, the sellers profit and
  buyers satisfaction are maximized.

16
Classical optimization approach

• A classical optimization formulation(??) consists
  of an objective to be maximized/minimized and a
  set of constraints(??).
• A classical optimization problem can be
  formulated by the controller entities for
  spectrum trading to maximize the profit of the
  spectrum owner by adjusting(??) the specrum
  price.

17
Noncooperative game

• In spectrum trading, multiple primary users offer
  prices to sell spectrum to secondary users
  intending to maximize their profits.
• Noncooperative game formulations are widely used
  to analyze and obtain an equilibrium solution
  that satisfies all of the entities.

18
Bargaining game

• A bargaining game formulation can be used in
  situations where players can cooperate, and a
  player can influence(??) the action of other
  players in trading the radio spectrum.
• In this game the players can negotiate(??) and
  bargain(??) with each other.

19
Auction

• Auction is performed by buyers who submit(??)
  their bids(??) to a seller.
• The seller decides how much of and to whom to
  sell the spectrum.
• This auction is suitable for a situation where
  the price of the resource is undetermined(???)
  and is variable with the buyers requirements.

20
System model

• The primary service controller broadcasts price
  information to all secondary service
  controllers.
• Then spectrum demands from the secondary service
  controllers are fed back to the primary service
  controller to update the price based on spectrum
  supply function.

21
(No Transcript)
22
Equilibrium in spectrum sharing and pricing

• The objective of the primary service is to
  maximize profit through selling/sharing the radio
  spectrum with the secondary service, the aim of
  the secondary service is to maximize the
  satisfaction of the connections.

23
Expectation and learning(1/2)

• We consider a learning algorithm, namely, GFM.
• This algorithm uses recursive(??) updating to
  obtain the actual information under
  uncertainty(????), which is the price offered by
  the primary service.
• pe t 1 pe t at(pt-pet)

24
Expectation and learning(2/2)

• pe t is the estimated(??) price by the
  secondary service at time t, and pt is the
  current price from the primary service.
• In this learning algorithm the estimated price in
  the previous(???) iteration(??) is corrected in
  the direction of error weighted by the learning
  rate, which is a function of the observed(??)
  price.

25
Spectrum demand and supply functions
26
convergence to the equilibrium price and size of
allocated spectrum
27
Conclusion

• ????????????????????????????????,?????????????????
  ??????,???????????????????,?????????????????,?????
  ???