اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدA Joint Energy and Latency Framework for Transfer Learning over 5G Industrial Edge Networks
141
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
In this paper, we propose a transfer learning (TL)-enabled edge-CNN framework for 5G industrial edge networks with privacy-preserving characteristic. In particular, the edge server can use the existing image dataset to train the CNN in advance, which is further fine-tuned based on the limited datasets uploaded from the devices. With the aid of TL, the devices that are not participating in the training only need to fine-tune the trained edge-CNN model without training from scratch. Due to the energy budget of the devices and the limited communication bandwidth, a joint energy and latency problem is formulated, which is solved by decomposing the original problem into an uploading decision subproblem and a wireless bandwidth allocation subproblem. Experiments using ImageNet demonstrate that the proposed TL-enabled edge-CNN framework can achieve almost 85% prediction accuracy of the baseline by uploading only about 1% model parameters, for a compression ratio of 32 of the autoencoder.
قيم البحث
اقرأ أيضاً
Unlike theoretical distributed learning (DL), DL over wireless edge networks faces the inherent dynamics/uncertainty of wireless connections and edge nodes, making DL less efficient or even inapplicable under the highly dynamic wireless edge networks
(e.g., using mmW interfaces). This article addresses these problems by leveraging recent advances in coded computing and the deep dueling neural network architecture. By introducing coded structures/redundancy, a distributed learning task can be completed without waiting for straggling nodes. Unlike conventional coded computing that only optimizes the code structure, coded distributed learning over the wireless edge also requires to optimize the selection/scheduling of wireless edge nodes with heterogeneous connections, computing capability, and straggling effects. However, even neglecting the aforementioned dynamics/uncertainty, the resulting joint optimization of coding and scheduling to minimize the distributed learning time turns out to be NP-hard. To tackle this and to account for the dynamics and uncertainty of wireless connections and edge nodes, we reformulate the problem as a Markov Decision Process and then design a novel deep reinforcement learning algorithm that employs the deep dueling neural network architecture to find the jointly optimal coding scheme and the best set of edge nodes for different learning tasks without explicit information about the wireless environment and edge nodes straggling parameters. Simulations show that the proposed framework reduces the average learning delay in wireless edge computing up to 66% compared with other DL approaches. The jointly optimal framework in this article is also applicable to any distributed learning scheme with heterogeneous and uncertain computing nodes.
In this paper, we aim at interference mitigation in 5G millimeter-Wave (mm-Wave) communications by employing beamforming and Non-Orthogonal Multiple Access (NOMA) techniques with the aim of improving networks aggregate rate. Despite the potential cap
acity gains of mm-Wave and NOMA, many technical challenges might hinder that performance gain. In particular, the performance of Successive Interference Cancellation (SIC) diminishes rapidly as the number of users increases per beam, which leads to higher intra-beam interference. Furthermore, intersection regions between adjacent cells give rise to inter-beam inter-cell interference. To mitigate both interference levels, optimal selection of the number of beams in addition to best allocation of users to those beams is essential. In this paper, we address the problem of joint user-cell association and selection of number of beams for the purpose of maximizing the aggregate network capacity. We propose three machine learning-based algorithms; transfer Q-learning (TQL), Q-learning, and Best SINR association with Density-based Spatial Clustering of Applications with Noise (BSDC) algorithms and compare their performance under different scenarios. Under mobility, TQL and Q-learning demonstrate 12% rate improvement over BSDC at the highest offered traffic load. For stationary scenarios, Q-learning and BSDC outperform TQL, however TQL achieves about 29% convergence speedup compared to Q-learning.
Common Public Radio Interface (CPRI) is a successful industry cooperation defining the publicly available specification for the key internal interface of radio base stations between the radio equipment control (REC) and the radio equipment (RE) in th
e fronthaul of mobile networks. However, CPRI is expensive to deploy, consumes large bandwidth, and currently is statically configured. On the other hand, an Ethernet-based mobile fronthaul will be cost-efficient and more easily reconfigurable. Encapsulating CPRI over Ethernet (CoE) is an attractive solution, but stringent CPRI requirements such as delay and jitter are major challenges that need to be met to make CoE a reality. This study investigates whether CoE can meet delay and jitter requirements by performing FPGA-based Verilog experiments and simulations. Verilog experiments show that CoE encapsulation with fixed Ethernet frame size requires about tens of microseconds. Numerical experiments show that the proposed scheduling policy of CoE flows on Ethernet can reduce jitter when redundant Ethernet capacity is provided. The reduction in jitter can be as large as 1 {mu}s, hence making Ethernet-based mobile fronthaul a credible technology.
Fog radio access network (F-RAN) and virtualisation are promising technologies for 5G networks. In F-RAN, the fog and cloud computing are integrated where the conventional C-RAN functions are diverged to the edge devices of radio access networks. F-R
AN is adopted to mitigate the burden of front-haul and improve the end to end (E2E) latency. On other hand, virtualization and network function virtualization (NFV) are IT techniques that aim to convert the functions from hardware to software based functions. Many merits could be brought by the employment of NFV in mobile networks including a high degree of reliability, flexibility and energy efficiency. In this paper, a virtualization framework is introduced for F-RAN to improve the energy efficiency in 5G networks. In this framework, a gigabit passive optical network (GPON) is leveraged as a backbone network for the proposed F-RAN architecture where it connects several evolved nodes B (eNodeBs) via fibre cables. The energy-efficiency of the proposed F-RAN architecture has been investigated and compared with the conventional C-RAN architecture in two different scenarios using mixed integer linear programming (MILP) models. The MILP results indicate that on average a 30% power saving can be achieved by the F-RAN architecture compared with the C-RAN architecture.
This letter proposes two novel proactive cooperative caching approaches using deep learning (DL) to predict users content demand in a mobile edge caching network. In the first approach, a (central) content server takes responsibilities to collect inf
ormation from all mobile edge nodes (MENs) in the network and then performs our proposed deep learning (DL) algorithm to predict the content demand for the whole network. However, such a centralized approach may disclose the private information because MENs have to share their local users data with the content server. Thus, in the second approach, we propose a novel distributed deep learning (DDL) based framework. The DDL allows MENs in the network to collaborate and exchange information to reduce the error of content demand prediction without revealing the private information of mobile users. Through simulation results, we show that our proposed approaches can enhance the accuracy by reducing the root mean squared error (RMSE) up to 33.7% and reduce the service delay by 36.1% compared with other machine learning algorithms.
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
