Multi-access edge computing (MEC) has recently been proposed to aid mobile end devices in providing compute-and data-intensive services with low latency. Growing service demands by the end devices may overwhelm MEC installations, while cost constraints limit the increases of the installed MEC computing and data storage capacities. At the same time, the ever increasing computation capabilities and storage capacities of mobile end devices are valuable resources that can be utilized to enhance the MEC. This article comprehensively surveys the topic area of device-enhanced MEC, i.e., mechanisms that jointly utilize the resources of the community of end devices and the installed MEC to provide services to end devices. We classify the device-enhanced MEC mechanisms into mechanisms for computation offloading and mechanisms for caching. We further subclassify the offloading and caching mechanisms according to the targeted performance goals, which include latency minimization, energy conservation, and enhanced security. We identify the main limitations of the existing device-enhanced MEC mechanisms and outline future research directions.

Cooperative edge offloading to nearby end devices via Device-to-Device (D2D) links in edge networks with sliced computing resources has mainly been studied for end devices (helper nodes) that are stationary (or follow predetermined mobility paths) and for independent computation tasks. However, end devices are often mobile, and a given application request commonly requires a set of dependent computation tasks. We formulate a novel model for the cooperative edge offloading of dependent computation tasks to mobile helper nodes. We model the task dependencies with a general task dependency graph. Our model employs the state-of-the-art deep-learning-based PECNet mobility model and offloads a task only when the sojourn time in the coverage area of a helper node or Multi-access Edge Computing (MEC) server is sufficiently long. We formulate the minimization problem for the consumed battery energy for task execution, task data transmission, and waiting for offloaded task results on end devices. We convert the resulting non-convex mixed integer nonlinear programming problem into an equivalent quadratically constrained quadratic programming (QCQP) problem, which we solve via a novel Energy-Efficient Task Offloading (EETO) algorithm. The numerical evaluations indicate that the EETO approach consistently reduces the battery energy consumption across a wide range of task complexities and task completion deadlines and can thus extend the battery lifetimes of mobile devices operating with sliced edge computing resources.

Traffic light control falls into two main categories: Agnostic systems that do not exploit knowledge of the current traffic state, e.g., the positions and velocities of vehicles approaching intersections, and holistic systems that exploit knowledge of the current traffic state. Emerging fifth generation (5G) wireless networks enable Vehicle-to-Infrastructure (V2I) communication to reliably and quickly collect the current traffic state. However, to the best of our knowledge, the optimized traffic light management without and with current traffic state information has not been compared in detail. This study fills this gap in the literature by designing representative Deep Reinforcement Learning (DRL) agents that learn the control of multiple traffic lights without and with current traffic state information. Our agnostic agent considers mainly the current phase of all traffic lights and the expired times since the last change. In addition, our holistic agent considers the positions and velocities of the vehicles approaching the intersections. We compare the agnostic and holistic agents for simulated traffic scenarios, including a road network from Barcelona, Spain. We find that the holistic system substantially increases average vehicle velocities and flow rates, while reducing CO2 emissions, average wait and trip times, as well as a driver stress metric.

Congestion Control (CC) limits the transmission rate of data packets in the transport layer protocol, e.g., in TCP and QUIC, of a sending endpoint so as to mitigate the congestion in packet-switched communication networks. Numerous studies have developed and evaluated Congestion Control Algorithms (CCAs) for a wide range of network types. Prior surveys have covered the functioning of the existing CCAs and their performance characteristics. In contrast, this survey focuses on the quantitative evaluation of CCAs; specifically, we comprehensively survey the CCA evaluation methodologies, i.e., the why and how of CCA evaluations.We define a CCA evaluation as a combination of a prescribed evaluation scenario with prescribed evaluation metrics. In turn, we define an evaluation scenario as a combination of a prescribed network (e.g., path) model and a particular traffic model. We present comprehensive sets of network and traffic model characteristics (i.e., network scenario characteristics) that can influence the performance of CCAs, as well as a comprehensive set of evaluation metrics for measuring the performance of CCAs. We taxonomize CCA evaluations into the main categories: Steady-state, transient-state, network variation, dynamic (e.g., wireless) network, and competition evaluations. We place and survey the existing CCA evaluation studies in this taxonomy to uncover the main trends and shortcomings of the CCA evaluation field to date. To facilitate the comprehensive rigorous evaluation of CCAs, we have developed the publicly available CCA evaluation framework ccperf, which incorporates the insights from our survey. We also outline future research directions to further advance the comprehensive rigorous evaluation of CCAs.

System-level simulations play a crucial role in evaluating the behaviors of cellular networks, yet most studies rely on synthetic simulation scenarios rather than reproducing realistic network deployments. This paper presents vBerlinV2N, a system-level simulation model based on the network simulator 3 (ns-3) that replicates the BerlinV2N measurement campaign conducted in a live Long Term Evolution (LTE) network. We describe the methodology for setting up and calibrating the simulation to match real-world observations, including network topology, radio conditions, and user mobility. We discuss the challenges of accurately recreating real deployments in simulations by identifying missing information in the public domain and limitations of ns-3. Furthermore, we generate a simulation dataset with vBerlinV2N that closely resembles the BerlinV2N measurement dataset. A statistical evaluation of the measured and simulated datasets validates that the simulated vBerlinV2N dataset captures the essential statistical patterns of the measured BerlinV2N dataset. We analyze the distributions as well as the temporal and spatial variations of received signal quality metrics of the radio environments and compare the end-to-end throughput, delay, and handovers.