Resource Dependency and Multi-tenancy in Mobile Edge Clouds (MEC) present major resource orchestration challenges owing to limited and interdependent resources and dynamic deployment environments. Traditional cloud computing solutions are resource-dependency-agnostic, failing to maintain fair resource allocation and achieve high Service Level Objectives (SLOs) rates. In this paper, we present EdgeCore, a Resource Dependency (RD)-aware multi-tenant orchestrator for MEC. EdgeCore includes three integral components: 1) RD Resolver, which resolves the RD problem ignored by prior orchestrators; 2) EdgeDRF establishes multi-resource fairness among multiple tenants in MEC where prior solutions fall short; 3) EdgePilot utilizes the Resource Monitor, Resource Recommender, and Screening Algorithm, to continuously maintain the fair resource allocation across various deployments. EdgeCore is the first resource dependency-aware orchestrator outperforming the state-of-the-art AutoPilot and H-DRF with 20% higher SLO satisfaction rates evaluated in disaster response scenarios. Its lightweight implementation can be generalized to a multitude of MEC environments. EdgeCore code is available at https://github.com/LENSS/EdgeCore-SEC2024.

Mobile Edge Computing (MEC) has been gaining a major interest for use in Cyber-Physical Systems (CPS) for Disaster Response and Tactical applications. These CPS generate a very large amount of mission-critical and personal data that require resilient and secure storage and sharing. In this article, we present the design, implementation, and evaluation of a framework for resilient data storage and sharing for MEC in CPS targeting the aforementioned applications. Our framework is built on the resiliency of three main components: EdgeKeeper, which ensures resilient coordination of the framework’s components; RSock, which provides resilient communication among CPS’s nodes; and R-Drive/R-Share which, leveraging EdgeKeeper and RSock, provides resilient data storage and sharing. EdgeKeeper employs a set of replicas and a consensus protocol for storing critical meta-data and ensuring fast reorganization of the CPS; RSock decides an optimal degree for replicating data that is communicated over lossy links. R-Drive employs an adaptive erasure-coded and encrypted resilient data storage; R-Share, leveraging RSock provides resilient peer-to-peer data sharing. We implemented our proposed framework on rapidly deployable systems (e.g., manpacks, testMobile Edge Clouds) and on Android devices, and integrated it with existing MEC applications. Performance evaluation results from three real-world deployments show that our framework provides resilient data storage and sharing in MEC for CPS.

This poster presents a multi-access network testbed designed to serve as an essential tool for research on Access Traffic Steering, Switching, and Splitting (ATSSS) in 4G and 5G networks. By integrating Non-3GPP Inter-Working Function (N3IWF) and 5G core from different open-source projects, our testbed enables multiaccess sessions in 4G-LTE, 5G NSA, and 5G SA. Furthermore, our testbed incorporates a transport converter into the 5G core network, allowing User Equipment (UE) to utilize multi-access networks, even in the absence of multi-path TCP (MPTCP) support from external servers. We demonstrated the effectiveness of our testbed by utilizing ATSSS for detecting cybersecurity attacks, marking the first instance of employing ATSSS capabilities to improve cybersecurity in 4G/5G networks.

Accurate and prompt delivery of Emergency Medical Services (EMS) is critical in emergency incidents, e.g., man-made or natural disaster areas. However, quickly selecting the correct EMS protocol(s) (which dictate the medical procedures to be administered to patients) in complex medical scenarios, remains a key, demanding task for Emergency Medical Technicians (EMT). In this paper, we present EMSAssist, the first end-to-end mobile voice assistant at the edge for EMS. EMSAssist consists of three major components that address technical challenges present in state-of-the-art solutions: 1) For the first time, EMSAssist proposes and applies a few-sample fine-tuning technique in medical speech recognition task, that achieves a faster and more accurate speech transcription on our EMS audio dataset, when compared to Google Cloud Speech-to-Text; 2) A WordPiece tokenizer helps boosting the end-to-end EMS protocol selection accuracy by retrieving useful information from incorrect transcriptions; 3) A novel data customization framework that enables our data-driven EMSMobileBERT model to become the new state-of-the-art for EMS protocol selection. Extensive end-to-end evaluation results at the edge show EMSAssist can more accurately select EMS protocols (Top-5 accuracy above 96%) for EMTs, with end-to-end latencies of around 4.2 seconds.

We present EMSAssist, the first end-to-end mobile voice assistant for emergency medical services (EMS). EMSAssist allows Emergency Medical Technicians (EMT) to verbally describe patients' signs and symptoms and uses EMTs' voice input to recommend top-5 EMS protocols. Through this demo, we allow the attendees to evaluate EMSAssist through a pair of Google Glass and a mobile phone. Both mobile devices can collect users' voices as input and output top-5 recommended protocols. A companion youtube video of using EMSAssist on the Google Glass is provided: https://www.youtube.com/watch?v=bj7aQJKf4aE

The popularity of video cameras has spawned a new type of application called multitask video processing, which uses multiple CNNs to obtain different information of interests from a raw video stream. Unfortunately, the huge resource requirements of CNNs make the concurrent execution of multiple CNNs on a single resource-constrained mobile device challenging. Existing solutions solve this challenge by offloading CNN models to the cloud or edge server, compressing CNN models to fit the mobile device, or sharing some common parts of multiple CNN models. Most of these solutions, however, use the above offloading, compression or sharing strategies in a separate manner, which fail to adapt to the complex edge computing scenario well. In this paper, to solve the above limitation, we propose AMVP, an adaptive execution framework for CNN-based multitask video processing, which elegantly integrates the strategies of CNN layer sharing, feature compression, and model offloading. First, AMVP reduces the total computation workload of multiple CNN inference by sharing some common frozen CNN layers. Second, AMVP supports distributed CNN inference by splitting big CNNs into smaller components running on different devices. Third, AMVP leverages a quantization-based feature compression mechanism to reduce the feature transmission traffic size between two separate CNN components. We conduct extensive experiments on AMVP and the experimental results show that our AMVP framework can adapt to different performance goals and execution environments. Compared to two baseline approaches that only share or offload CNN layers, AMVP achieves up to 61% lower latency and 10% higher throughput with comparative accuracy.