Browsing by Author "Hong, Choong Seon"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Dynamic Resource Allocation for Optimized Latency and Reliability in Vehicular Networks
    Ashraf, Muhammad Ikram; Liu, Chen-Feng; Bennis, Mehdi; Saad, Walid; Hong, Choong Seon (IEEE, 2018)
    Supporting ultra-reliable and low-latency communications (URLLC) is crucial for vehicular traffic safety and other mission-critical applications. In this paper, a novel proximity and quality-ofservice-aware resource allocation framework for vehicle-to-vehicle (V2V) communication is proposed. The proposed scheme incorporates the physical proximity and traffic demands of vehicles to minimize the total transmission power over the allocated resource blocks (RBs) under reliability and queuing latency constraints. A Lyapunov framework is used to decompose the power minimization problem into two interrelated sub-problems: RB allocation and power optimization. To minimize the overhead introduced by frequent information exchange between the vehicles and the roadside unit (RSU), the resource allocation problem is solved in a semi-distributed fashion. First, a novel RSU-assisted virtual clustering mechanism is proposed to group vehicles into disjoint zones based on mutual interference. Second, a per-zone matching game is proposed to allocate RBs to each vehicle user equipment (VUE) based on vehicles' traffic demands and their latency and reliability requirements. In the formulated one-to-many matching game, VUE pairs and RBs rank one another using preference relations that capture both the queue dynamics and interference. To solve this game, a semi-decentralized algorithm is proposed using which the VUEs and RBs can reach a stable matching. Finally, a latency-and reliability-aware power allocation solution is proposed for each VUE pair over the assigned subset of RBs. Simulation results for a Manhattan model show that the proposed scheme outperforms the state-of-art baseline and reaches up to 45% reduction in the queuing latency and 94% improvement in reliability.
  • Thumbnail Image
    Proactive edge computing in fog networks with latency and reliability guarantees
    Elbamby, Mohammed S.; Bennis, Mehdi; Saad, Walid; Latva-aho, Matti; Hong, Choong Seon (2018-08-20)
    This paper studies the problem of task distribution and proactive edge caching in fog networks with latency and reliability constraints. In the proposed approach, user nodes (UNs) offload their computing tasks to edge computing servers (cloudlets). Cloudlets leverage their computing and storage capabilities to proactively compute and store cacheable computing results. In this regard, a task popularity estimation and caching policy schemes are proposed. Furthermore, the problem of UNs’ tasks distribution to cloudlets is modeled as a one-to-one matching game. In this game, UNs whose requests exceed a delay threshold use the notion of hedged-requests to enqueue their request in another cloudlet, and offload the task data to whichever is available first. A matching algorithm based on the deferred-acceptance matching is used to solve this game. Simulation results show that the proposed approach guarantees reliable service and minimal latency, reaching up to 50 and 65% reduction in the average delay and the 99th percentile delay, as compared to reactive baseline schemes.
  • Thumbnail Image
    Self-Organizing Democratized Learning: Toward Large-Scale Distributed Learning Systems
    Nguyen, Minh N. H.; Pandey, Shashi Raj; Tri Nguyen Dang; Eui-Nam Huh; Tran, Nguyen H.; Saad, Walid; Hong, Choong Seon (IEEE, 2022-05-10)
    Emerging cross-device artificial intelligence (AI) applications require a transition from conventional centralized learning systems toward large-scale distributed AI systems that can collaboratively perform complex learning tasks. In this regard, democratized learning (Dem-AI) lays out a holistic philosophy with underlying principles for building large-scale distributed and democratized machine learning systems. The outlined principles are meant to study a generalization in distributed learning systems that go beyond existing mechanisms such as federated learning (FL). Moreover, such learning systems rely on hierarchical self-organization of well-connected distributed learning agents who have limited and highly personalized data and can evolve and regulate themselves based on the underlying duality of specialized and generalized processes. Inspired by Dem-AI philosophy, a novel distributed learning approach is proposed in this article. The approach consists of a self-organizing hierarchical structuring mechanism based on agglomerative clustering, hierarchical generalization, and corresponding learning mechanism. Subsequently, hierarchical generalized learning problems in recursive forms are formulated and shown to be approximately solved using the solutions of distributed personalized learning problems and hierarchical update mechanisms. To that end, a distributed learning algorithm, namely DemLearn, is proposed. Extensive experiments on benchmark MNIST, Fashion-MNIST, FE-MNIST, and CIFAR-10 datasets show that the proposed algorithm demonstrates better results in the generalization performance of learning models in agents compared to the conventional FL algorithms. The detailed analysis provides useful observations to further handle both the generalization and specialization performance of the learning models in Dem-AI systems.