Generally, license for spectrum access is provided to a primary user or service provider through an auction process in a primary market as indicated below;
When the allocated spectrum is under-utilized, the licensed user can lease the spectrum in a secondary market to an unlicensed user which temporarily demands the spectrum for a particular service. In the primary market, the process of spectrum allocation is lengthy and inflexible due to the regulatory requirements. Therefore, spectrum leasing in the secondary market, which is not controlled by the government, appears to be an attractive tool to promote efficient use of the radio spectrum. Spectrum leasing (or trading) between the licensed and unlicensed users will be a significant issue in cognitive radio networks. In economics, trading is defined as a process of exchanging commodity or service in a market. This process can be performed through direct exchange of commodity or service (i.e. bartering) or through a medium of exchange which is generally the money. The concept of trading can be applied to spectrum leasing in the secondary market, which constitutes the concept of spectrum trading. Specifically, spectrum trading is defined as a process of selling and buying of spectrum resources in different dimensions (e.g. frequency band, time slot). Pricing is an important issue here for both the licensed user (or primary service provider) selling the spectrum and the unlicensed users (or secondary service provider or secondary user) buying the spectrum.
While dynamic spectrum access encompasses technical functionalities including spectrum sensing at the physical and MAC layers, channel access, routing, and higher layer protocols, spectrum trading can be regarded as its other component which deals with the economic aspects of dynamic spectrum access as indicated below;
Spectrum trading can be considered as a component of spectrum management and, therefore, it is required to be integrated with other components in a cognitive radio network. Recall that two major steps in spectrum sharing are spectrum exploration and spectrum exploitation. Spectrum trading is a process between spectrum exploration and exploitation. A spectrum seller has to perform spectrum exploration to identify spectrum opportunities. Then, these spectrum opportunities can be sold to the spectrum buyer. After obtaining the right to access, the spectrum buyer performs the spectrum exploitation step to utilize the spectrum to achieve its objectives under the constraints defined by the spectrum seller.
Two major structures of spectrum trading are as follows:
• Single seller – monopoly: In this case, there is only a single seller in the system and this is referred to as a monopoly market. In this market, a seller can maximize its revenue given the demand from the buyers. In this case, the buyers can either compete or cooperate to buy spectrum opportunities from the seller. This behavior (i.e. competition, cooperation) will affect the revenue of a seller and depending on the behavior of the buyers, a seller may adapt the trading parameters accordingly.
• Multiple sellers – oligopoly: In this case, there are multiple sellers offering radio spectrum to the market and this is referred to as an oligopoly market. The buyer has a choice to choose the best offer to maximize its satisfaction in both performance and price. Since there is a competition among the sellers, the profit of a spectrum seller is always less than that in a single-seller case.
Alternatively, a spectrum trading market may not have a permanent seller (e.g. as in a commons-use spectrum access model). In this spectrum trading structure, all users have the right to access the spectrum. However, if a particular user (e.g. a renting user) requires to have a spectrum share more than another user (e.g. a leasing user), the other user will need to be compensated. For example, a renting user may provide an incentive to a leasing user and access the spectrum with special right for a certain period of time. When a renting user accesses the spectrum with special right, credit is given to the leasing user. Afterwards, when necessary, the leasing user may use this credit to access the spectrum as a renting user with special rights while another user becomes a leasing user.
In spectrum trading, the objective of a seller is to maximize the revenue/profit, while that of a buyer is to maximize the utility of spectrum usage. However, these objectives generally conflict with each other As the seller increases the price to achieve higher revenue, the utility of a buyer decreases due to the higher cost. A similar effect on the QoS performance is observed when the amount of spectrum allocated to an unlicensed user is varied. In particular, as the spectrum size allocated to an unlicensed user increases, the utility of an unlicensed user increases, but the performance of a licensed user degrades. Therefore, an optimal and stable solution for spectrum trading in terms of price and allocated spectrum would be required so that the revenue and utility are maximized while both the seller and the buyer are satisfied and do not want to deviate from the solution.
When developing a spectrum trading model, the following three aspects need to be considered.
Mode: The mode refers to (1) the change of spectrum ownership and (2) the change of use (e.g. different wireless access technologies) due to spectrum trading. After spectrum trading is done, the spectrum ownership will be transferred from the spectrum seller to the appropriate buyer (i.e. change of spectrum ownership). Also, the type of wireless service operating on the traded spectrum can be changed (i.e. change of use).
Extent: It defines the degree of a spectrum owner’s rights and obligations which will be transferred to a buyer, i.e. (1) complete transfer and (2) shared. In a complete transfer case, all rights and obligations of spectrum access are completely transferred to the buyer. In contrast, in the shared case, both spectrum owner and buyer share the rights and obligations.
Duration: The duration determines the length of time the traded spectrum can be accessed by a buyer. Different scales of duration can be defined as follows: (1) short term lease, (2) long-term lease, (3) sale-and-buy-back, and (4) permanent. While the duration of short-term lease may be a few hours, that of long-term lease could be a few months. For the sale-and-buy-back case, the spectrum is sold and can be bought back by a spectrum owner if it is demanded. In the permanent case, the spectrum is permanently sold to a buyer. This buyer can access the spectrum until the license expires.
