1 - 4 of 4
ItemUser Plane Hardware Acceleration in Access Networks: Experiences in Offloading Network Functions in Real 5G Deployments( 2022-01-04)Fulfilling the ambitious Quality of Service demands of today’s wireless networks, especially low latency, high bandwidths and availability, is a big challenge for researchers, network architects, and operators. Each networking component on the data path between the user equipment and the destination data network, e.g., the Internet, must provide the highest performance to meet these requirements. This work demonstrates how different network elements of the user plane, describing the whole path of user traffic, can be sped up with different hardware acceleration technologies. For that, we demonstrate how to build up a 5G standalone campus network for evaluation, working end-to-end with real user equipment and open-source software components. Further, we analyze the user-plane network functions of 5G networks from the radio access network to the core. Based on our real 5G setup, the practical evaluation of the analysis results shows up how the 5G user-plane hardware can be accelerated best.
ItemRobustness in Nonorthogonal Multiple Access 5G Networks( 2022-01-04)The diversity of fifth generation (5G) network use cases, multiple access technologies, and network deployments requires measures of network robustness that complement throughput-centric error rates. In this paper, we investigate robustness in nonorthogonal multiple access (NOMA) 5G networks through temporal network theory. We develop a graph model and analytical framework to characterize time-varying network connectedness as a function of NOMA overloading. We extend our analysis to derive lower bounds and probability expressions for the number of medium access control frames required to achieve pairwise connectivity between all network devices. We support our analytical results through simulation.
ItemGroup Key Management in Wireless Sensor Networks: Introducing Context for Managing the Re-keying Process( 2022-01-04)This paper proposes an algorithmic solution to Group Key Management (GKM) in Wireless Sensor Networks (WSN), which could address a single point of failure in cybersecurity. The paper moves away from the traditional (de)centralized and distributed solution in GKM and focuses on GKM decision making based on a) the context in which WSN and their nodes communicate, and b) the semantic which describe the environment where WSN and their nodes reside. The proposed algorithm defines which node, within the WSN, could start a re-keying process by generating a group key, and why/how this decision on the re-keying has been made. The algorithm is computable and thus it would be feasible to implement it in software applications built upon a set of WSN nodes in constantly changeable and dynamic mobile computing environments.