Accountable and Transparent Resource Sharing and Provisioning in Future Networks

Student thesis: Doctoral ThesisDoctor of Philosophy

Abstract

The rapid development of network technologies has escalated the demand for network connections; methods and strategies of network sharing and resource management are introduced every day. With the upcoming 6G and beyond networks, technologies such as Industry 4.0 and autonomous surgery will become mature and prevalent. These applications vary with each other in several ways, such as data requirement, computational power, and location. Hence, the monolithic service provisioning architecture is no longer viable to cope with the diversity in service demand. The key network players, such as network planners and service providers will need to redesign and redefine the ways network resources are managed, distributed, and shared. Indeed, technologies such as Software Defined Networking (SDN), and Network Function Virtualisation (NFV) provide viable resource sharing mechanisms. Yet, resource sharing needs to broaden its spectrum and reach every granularity of the network architecture; that is, it must include more variables in the equations that govern viable and scalable network resource sharing. While an immense amount of research has already been conducted on network resource sharing techniques, little attention has been paid to an accountable and transparent contracting mechanism for network resource management and sharing.

The main objective of this thesis is to explore the potential of distributed ledger technologies for accountable and transparent networks resource management and sharing, with the focus on a tamper-proof contracting mechanism. This thesis explores smart contracts as the contracting mechanism and evaluates their viability as Service Level Agreements (SLAs). As a comprehensive suitability assessment is required, we explore, firstly, the network applications where smart contracts are a potential contracting solution. A comprehensive study is done on the potential of smart contracts as SLAs in future network applications.

Next, we propose an end-to-end novel architecture JITRA – Just-in-Time Resource Allocation with distributed ledgers for 5G and beyond. This architecture proposes short-term and flexible service contracts instead of traditional long-term and monolithic ones. To explore the resource reservation, we have studied network slicing as a use case for resource provisioning and developed a Mininet-based simulation for network slicing on similar, but simpler, real-world network topologies. The network slice was created in the evaluation in minutes and the contract was executed in the order of just seconds.

To enable data integrity in JITRA, we next propose cryptographic accumulators and a side-channel based (Quality of Service) QoS monitoring mechanism. A comprehensive evaluation of three different types of cryptographic accumulators was conducted: RSA accumulators, secure bloom filters and sketches. The evaluation results show that secure bloom filters outperform other accumulation techniques in terms of latency, size and setting up time.

Finally, we propose BEAT – Blockchain-Enabled Accountable and Transparent for infrastructure sharing in 6G and beyond. BEAT is an open architecture in which several network users own network devices; the usage is recorded to a distributed ledger. An Interrogation Protocol is further proposed to handle situations of a dispute between the network users. Our evaluations show that BEAT adds only seconds of overhead to the network flows on a limited resource device.




Date of Award1 Mar 2023
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorOsvaldo Simeone (Supervisor) & Mohammed Shikh-Bahaei (Supervisor)

Cite this

'