Automatic Pipeline Parallelism: A Parallel Inference Framework for Deep Learning Applications in 6G Mobile Communication Systems

Shi, Hailong; Zheng, Wenbin; Liu, Zifei; Ma, Ruhui; Guan, Haibing

doi:10.1109/jsac.2023.3280970

Cited by 5 publications

(1 citation statement)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent advancements in state-of-the-art applications, including intelligent manufacturing, autonomous driving, and remote operations, have necessitated that mobile networks support communication with an exceptionally low latency [ 1 , 2 , 3 ]. This demand has prompted the emergence of ultra-reliable low-latency communication (URLLC) as one of the three communication scenarios within fifth generation mobile networks (5G).…”

Section: Introductionmentioning

confidence: 99%

RIS-Aided Proactive Mobile Network Downlink Interference Suppression: A Deep Reinforcement Learning Approach

Wang

Sun

Cui

et al. 2023

Sensors

View full text Add to dashboard Cite

A proactive mobile network (PMN) is a novel architecture enabling extremely low-latency communication. This architecture employs an open-loop transmission mode that prohibits all real-time control feedback processes and employs virtual cell technology to allocate resources non-exclusively to users. However, such a design also results in significant potential user interference and worsens the communication’s reliability. In this paper, we propose introducing multi-reconfigurable intelligent surface (RIS) technology into the downlink process of the PMN to increase the network’s capacity against interference. Since the PMN environment is complex and time varying and accurate channel state information cannot be acquired in real time, it is challenging to manage RISs to service the PMN effectively. We begin by formulating an optimization problem for RIS phase shifts and reflection coefficients. Furthermore, motivated by recent developments in deep reinforcement learning (DRL), we propose an asynchronous advantage actor–critic (A3C)-based method for solving the problem by appropriately designing the action space, state space, and reward function. Simulation results indicate that deploying RISs within a region can significantly facilitate interference suppression. The proposed A3C-based scheme can achieve a higher capacity than baseline schemes and approach the upper limit as the number of RISs increases.

show abstract

Section: Introductionmentioning

confidence: 99%

RIS-Aided Proactive Mobile Network Downlink Interference Suppression: A Deep Reinforcement Learning Approach

Wang

Sun

Cui

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

Guest Editorial xURLLC in 6G: Next Generation Ultra-Reliable and Low-Latency Communications

She

Pan

Duong

et al. 2023

IEEE J. Select. Areas Commun.

View full text Add to dashboard Cite

A S ONE of the new communication scenarios in 5th-generation (5G) mobile communication systems, ultra-reliable and low-latency communications (URLLC) have stringent requirements on latency (around 1 ms) and reliability (up to 99.99999%). Nevertheless, existing 5G URLLC alone cannot fulfill all the Key Performance Indicators (KPIs) in emerging mission-critical applications like industrial automation, intelligent transportation, telemedicine, Tactile Internet, and Virtual/Augmented Reality (VR/AR). The 6th generation (6G) communication systems need to meet additional requirements on some of the following KPIs in combination with URLLC: high spectrum efficiency (SE)/throughput/energy efficiency (EE)/network availability/security as well as low Age of Information (AoI)/jitter/round-trip delay. These new requirements pose unprecedented challenges in terms of design methodologies and enabling technologies in 6G. To fill the gap between 5G URLLC and the diverse KPI requirements of the neXt generation URLLC (xURLLC), novel methodologies and innovative technologies are much needed.The Special Issue received 76 submissions, and 20 of them were selected after a careful review process. These papers cover a wide range of topics in xURLLC, including fundamental theories, network architecture, diverse KPIs, xURLLC for Metaverse, resource allocation, channel access, beam management, signal processing, and quantum telecomputation.

show abstract

SLoB: Suboptimal Load Balancing Scheduling in Local Heterogeneous GPU Clusters for Large Language Model Inference

Jiang,

Wang,

Cai

et al. 2024

IEEE Trans. Comput. Soc. Syst.

View full text Add to dashboard Cite

Automatic Pipeline Parallelism: A Parallel Inference Framework for Deep Learning Applications in 6G Mobile Communication Systems

Cited by 5 publications

References 67 publications

RIS-Aided Proactive Mobile Network Downlink Interference Suppression: A Deep Reinforcement Learning Approach

RIS-Aided Proactive Mobile Network Downlink Interference Suppression: A Deep Reinforcement Learning Approach

Guest Editorial xURLLC in 6G: Next Generation Ultra-Reliable and Low-Latency Communications

SLoB: Suboptimal Load Balancing Scheduling in Local Heterogeneous GPU Clusters for Large Language Model Inference

Contact Info

Product

Resources

About