The contradiction between limited spectrum resources and the increasing bandwidth demand facilitates the evolution of the next generation of a mobile communication paradigm. While 5G is being put into widespread commercial use, studies on 6G have been carried out. As we all know, licensed spectrum resources account for a considerable proportion of mobile communication service. Licensed spectrum access (LSA) can guarantee the licensed users’ quality of service (QoS) at a high level. Different from 4G and 5G licensed spectrum distribution, 6G licensed spectrum distribution faces more challenges, including more connections, more decentralized locations and more security risks. The striking two distinguishing features from 6G to 5G are the introduction of a terahertz band [1] and Space–Ground Integrated Network (SGIN) architecture [2]. Although terahertz communication technology can significantly improve data transmission rates, it also brings greater path transmission damage and smaller cellular coverage. That is to say more micro base stations are needed to realize ubiquitous and wide-area wireless communication coverage. The wider spatial distribution is exactly one of the important characteristics of SGIN. Hence, it is inevitable for Mobile Network Operators (MNOs) to change their current centralized business model to a more flexible and decentralized one. This irreversible evolution is driven by emerging technologies, such as network virtualization, dynamic spectrum sharing, blockchain and so on. To address the unfair problem in the current licensed spectrum accessing mechanism, utilizing the blockchain technology is the study's aim of this paper.

Usually, in 4G and 5G mobile networks, MNOs distribute licensed spectrum resources according to a user’s service protocols agreed upon in advance. A licensed user’s periodic demand will be satisfied in a certain coverage region according to current geographic location. These service protocols are regulated through binding Service-Level Agreements (SLAs). Therefore, the present LSA spectrum access framework is called the distribution on demand model. Under this model, MNOs distribute the spectrum resources to different Primary Users (PUs) or a Primary Base Station (PBS) according to their demand. Some dishonest users would exaggerate their spectrum demand or violate the spectrum using regulations, obtaining extra interest. The common misconducts include transmitting with a bigger power than permitted, using a different carrier frequency than allocated and using the spectrum for more time than permitted [3]. However, there lacks an effective supervision and punishment measures for the violations. As a result, the dishonest users can obtain extra illegal interest compared to the honest users. Obviously, this is unfair for the honest users. On the other hand, the existing research results usually assume that operators and MNOs are honest participants in the spectrum distribution process. This means that users believe the obtained bandwidth resources are the same as the nominal value. Nevertheless, MNOs are actually rational participants, and the provided services may be discounted in order to obtain more benefits. For occasional and negligible service downgrades, users may not perceive without professional detection tools’ help. But if it is the other way around, the MNO will be complained about, or the users will even switch to another telecom service provider. Furthermore, for the above two kinds of bad behaviors of users and MNOs, although the detection means have been rather available, the supervision and audit means are still not rich.

To sum up the application status and related research results on 5G licensed spectrum distribution, the shortcomings of the present distribution model are mainly reflected in the following three aspects:

(1)

Unfairness between honest and dishonest users. For some dishonest PBS and PUs, violations of spectrum access regulations would not bring serious consequences, but acquire extra incomings. These violations may hurt honest users’ LSA authorities, leading to the unfairness in the spectrum distribution process.

(2)

Lack of supervision and audit mechanism. It is difficult for users to defend their rights when the spectrum accessing service provided with MNOs is degraded. To guarantee the fairness between MNOs and spectrum users, there is an urgent need to introduce a transparent supervision and auditing mechanism to help users defend their rights.

(3)

Existing incentives are inefficient for the operators. Under the present LSA mechanism, users belonging to a specific operator can only passively accept the LSA services provided with the MNOs. And MNOs obtain revenue from the upper tier operators. For them, there is no incentive to provide better service to users. For the PBS and PUs, misbehaviors in spectrum usage would not lead to a disadvantage in subsequent spectrum access. Thus, for the users, there lacks the incentive to maintain good credit.

PBS and PUs play key roles in future 6G ultra-dense mobile networks; sufficient spectrum resources are of vital importance for them to serve for the subordinate user nodes. The present licensed spectrum distribution faces the challenges of an unfair status and lack of a supervision and audit mechanism. Therefore, towards 6G-envisioned communications, how to effectively and fairly distribute the licensed spectrum from telecom operators to PBS and PUs is a problem that needs to be solved in the future. Moreover, to protect honest users’ interest and encourage MNOs to provide better LSA services, a supervision and auditing mechanism is an urgent need. To summarize, a more fair licensed spectrum distribution or primary-level allocation method is the scientific question we are interested in.

Since Nakamoto proposed Bitcoin [4] in 2008, the concept of blockchain has attracted worldwide attention. As an open decentralized ledger system, blockchain effectively combines cryptography and distributed consensus mechanisms to ensure data transparency and tamper resistance. Moreover, blockchain technology is also widely applied to many fields such as the Internet of Things (IoT) ^[5][6][5,6], secure storage ^[7][8][7,8] and supply chain management ^[9][10][9,10]. In recent years, researchers in academia and industries are beginning to explore the use of blockchain technology for spectrum allocation ^{[11][12][13][14]}[11,12,13,14].

2. Spectrum Distribution

Spectrum distribution is a main wireless channel access mechanism, where bandwidth is shared from MNOs to PUs and PBS. This mechanism is also called the primary-level spectrum distribution. In the literature [15], a novel LSA spectrum distribution algorithm is proposed, which can penalize users violating the LSA spectrum using rules by introducing a penalty mechanism. At the same time, it provides extra spectrum as incentive to the users complying with the regulations. Li M. proposes a spectrum distribution algorithm based on the idea of a proportional fairness algorithm, which uses the dynamic calculation of the user distribution weight values and the interference value of the current available spectrum resources. Through the dynamic adjustment of the device allocation weight value during the distribution process, a more fair spectrum distribution is achieved [16].

3. Spectrum Using Behavior Detection

The detection of the abnormal usage of a spectrum is the premise for spectrum management. For 6G spectrum distribution, spectrum usage behavior detection is the key component to build the trust value assessment mechanism and to further realize fair spectrum distribution. Liu et al. propose an algorithm for detecting abnormal behaviors based on electromagnetic data mining. The method is of a good accuracy and real-time performance [17]. In the literature [18], blockchain technology and machine learning are applied to detect malicious users in the IoT network. The proposed method can store the data including the spectrum access moment, occupied frequency and transmitting power, and separate the normal users from malicious ones with machine learning. A multi-attribute-based fairness-driven algorithm is proposed for the determination and interruption of SUs’ services to ensure fairness among services in the network’s resource utilization in [19].

4. Auditing Mechanism Based on Blockchain

Blockchain can be regarded as a time-stamped transaction recording system, which can record all transactions that have occurred on the blockchain. The transactions recorded on the blockchain are open, transparent, decentralized and hard to tamper with. To better evaluate the spectrum accessing service provided with the MNOs, it is important to supervise and audit the MNOs’ behaviors. Wang et al. propose a novel auditing mechanism supporting public auditing on shared data stored in the cloud. To improve the efficiency of auditing multiple tasks, the mechanism is further extended to support batch auditing [20]. Shang et al. design an identity-based dynamic data auditing scheme that is capable of performing dynamic auditing for big data storage service. To guarantee the correctness of the data update each time, a data structure, namely a Merkle hash tree, is used. The scheme can authenticate block tags and support dynamic operation with integrity assurance [21]. For the illegal authorization and key disclosure risks, Hei et al. design a blockchain-based auditing scheme; the auditor in the scheme can detect the malicious behaviors. Two smart contracts on Ethereum are respectively adopted to trace the two misbehaviors [22].