As an inevitable trend of future fifth generation (5G) networks, software-defined architecture has many advantages in providing centralised control and flexible resource management. However, it is also confronted with various security challenges and potential threats with emerging services and technologies. As the focus of network security, intrusion detection systems (IDSs) are usually deployed separately without collaboration. They are also unable to detect novel attacks with limited intelligent abilities, which are hard to meet the needs of software-defined 5G. In this study, the authors propose an intelligent IDS taking the advances in software-defined technology and artificial intelligence based on software-defined 5G architecture. It flexibly integrates security function modules which are adaptively invoked under centralised management and control with a global view. It can also deal with unknown intrusions by using machine learning algorithms. Evaluation results prove that the intelligent IDS achieves better performance with lower overhead. It is also verified that the selected machine learning algorithms show better accuracy and reduced false alarm rate in flow-based classification.

With the standardisation of the fifth generation (5G) wireless network interface was established in 3GPP, the features of 5G arouse great interest both in academics and industry, even government. The significant advancements of 5G wireless network technology will make a great increase in the demand for content-rich streaming media services. What's more, many mobile devices with heterogeneous network interfaces (4G, WiFi, and even 5G) connect multiple paths for data delivery simultaneously. Multiple Transmission Control Protocol (MPTCP) is recommended by Internet Engineering Task Force (IETF) to multi-path parallel transmission, which is an important transport layer protocol. Although there are many attractive benefits by using MPTCP for data delivery, it becomes sensitive for MPTCP to network attacks. When a subflow with the MPTCP association suffers from some network attacks (e.g. denial of service attacks), there is undoubted that it will cause large numbers of retransmission data. Unfortunately, the regular MPTCP retransmission mechanisms retransmit data over the same path, which may cause receiver buffer blocking. In this study, the authors propose effective retransmission mechanisms in MPTCP for retransmitting data on a non-congested path. The simulation results show that the proposed retransmission mechanism has better performance than the regular’ in the network attacks.

In this study, the authors consider the physical-layer security for transmission in underlay cognitive radio networks, where consisting of a primary source–destination pair, a secondary transmitter–receiver pair and a common eavesdropper (E). To transmit data in a secure and reliable manner, they propose a transmission protocol for an underlay cognitive radio network. In particular, the secondary user is allowed to operate on the primary spectrum bandwidth when the interference threshold is satisfied. To protect the transmission confidentiality of primary system, PS node transmits a signal, which is linearly combined with the jamming signal (as artificial noise) and expectation signal, to interfere E. Furthermore, they propose a superimposed coding to combine with the artificial noise and expectation signal at PS node. We also analyse the performance of the proposed transmission protocol. We obtain the outage probability of secondary and primary systems. Simultaneously, they derive the closed-form expression of intercept probability for primary system. In addition, they verify the performance of their proposed transmission protocol by numerical experiment. The numerical results show that their proposed transmission protocol not only has the better outage performance for secondary system but also has the better intercept performance for primary system.

Social networks have become part of today's life; however, they also create numerous privacy problems for their users as they reveal sensitive information. Thus, the privacy preservation has become a big problem in social networks and there are several researches related to this topic. However, the current research achievements for this issue are only ad hoc solutions and the broad picture of the problem is not clear. This paper surveys a wide range of related researches and realises that the topic includes many sub-areas; for example, the privacy in publishing social network data for the use of third-party consumers or the privacy of users in the leakage of individuals’ information to unexpected people in their social circle. The study also analyses the advantages as well as limitations of proposed solutions. On the basis of that the research outlines key outstanding issues and recommends directions for the future research. Moreover, the research also investigates common privacy metrics and popular privacy-preserving techniques to provide readers some basic tools to resolve open problems in the topic. The ultimate purpose of this research is to pave a solid background for people, who are interested in the topic, to do research further.

With the development of cloud computing in the mobile communications industry, the credibility between users and the cloud has become an important obstacle to the development of mobile cloud services. Therefore, the user's abnormal behaviour detection is particularly important. The authors propose a selective clustering ensemble algorithm based on fractal dimension. In their proposed algorithm, they firstly use the sliding window to dynamically obtain data to improve the accuracy of user behaviour acquisition. Secondly, they use the initial clustering stage and incremental clustering stage to produce clustered members. Thirdly, they also use the Duun_index to select the clustered members, then the selection of high-quality clustered members with the voting algorithm to get the final result. Finally, they use the average difference of change to determine whether the current behaviour is abnormal based on the clustering results. The experimental results show that the proposed scheme has a better performance than the traditional clustering algorithm in clustering accuracy. Moreover, their algorithm can improve the detection rate and accuracy rate, respectively.