Miao Guo

Miao Guo

Auto-generated profile, based on publicly available information

Miao Guo

Zhejiang University, Hangzhou, China

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 September 2022 in IEEE Transactions on Industrial Informatics
Verified
Journal Article
Closed access
MGMiao Guo
Miao Guo
Chaojie GuChaojie Gu
Chaojie Gu
Shibo HeShibo He
Shibo He
Zhiguo ShiZhiguo Shi
Zhiguo Shi
JCJiming Chen
Jiming Chen
Time-sensitive networking (TSN), an emerging network technology, requires high-performance scheduling mechanisms to deliver deterministic service in Industry 5.0. Cyclic queuing and forwarding (CQF) is launched to simplify the configuration complexity of the early stage mechanism time-aware shaper in TSN flow scheduling. Previous CQF studies adopt an inflexible incremental flow scheduling scheme, which consists of flow sorting, offset search, and resource judgment. However, we observe that flow sorting and offset search are mutually interdependent. The offset of a flow helps determine the resource status on the flow path, which can guide flow sorting. By utilizing the interaction between flow and offset, we design a novel scheduling approach that achieves high scheduling performance and time efficiency. Specifically, the proposed approach combines flow sorting and offset search together to select flow and its offset (i.e., ( flow, offset )) simultaneously. To effectively determine the selecting priority and select the potential optimal flow-offset combination, we define a unified metric, $mapping\,score$ , to quantify the schedulability of different flow and offset combinations. The extensive experiments demonstrate that the scheduling success rate of our proposed approach is on average 31.69% higher than the baseline and 4.57% higher than the state-of-the-art flow judgement approach (FLJ) method. Moreover, it outperforms the state-of-art FLJ method by 7.62% in large-scale linear topologies, indicating its great scalability in different network scales and complex topologies.

17 January 2024 in IEEE Network
Verified
Journal Article
Closed access
WTWenbin Tian
Wenbin Tian
Chaojie GuChaojie Gu
Chaojie Gu
MGMiao Guo
Miao Guo
Shibo HeShibo He
Shibo He
JKJiawen Kang
Jiawen Kang
JCJiming Chen
Jiming Chen
Driven by the continuous growth of network demands, Large-scale Deterministic Network (LDN) has been proposed to deliver deterministic services with improved latency and jitter, zero packet loss, configurable bandwidth, and high reliability across a broad geographical area. LDN serves as a bridge for businesses to connect and share information across different sectors with high-quality and customized network services. However, the conveyance of massive flows with stringent demands in a large topology poses technical challenges to LDN architecture, scheduling mechanisms, and operation methodologies. To shed light on the possible avenues for these issues, this article provides a comprehensive survey of LDN architecture and enabling technologies. The study begins with an overview of typical LDN architectures, including definitions, standards, and mechanisms, followed by an examination of the key technologies that support the LDN’s performance characteristics and implementation processes. Additionally, a case study of LDN scheduling is presented illustrating its application in heavy-traffic scenarios. Finally, this article outlines the open challenges and future directions in this field, highlighting areas of potential growth and development.

13 July 2023 in IEEE Transactions on Green Communications and Networking
Verified
Journal Article
Closed access
RHRenlang Huang
Renlang Huang
MGMiao Guo
Miao Guo
Chaojie GuChaojie Gu
Chaojie Gu
Shibo HeShibo He
Shibo He
JCJiming Chen
Jiming Chen
MSMingyang Sun
Mingyang Sun
As an emerging and promising network paradigm, network slicing creates multiple logical networks on shared infrastructure to provide services with customized Quality-of-Service (QoS) for heterogeneous devices and applications. However, network complexity and service heterogeneity pose a huge challenge in achieving optimal performance and ensuring stringent QoS requirements. In this paper, we design a hierarchical deep reinforcement learning based 5G radio access network slicing framework to achieve scalable and efficient resource allocation. By decomposing the resource allocation problem into a slice-level task and several user-level tasks, the proposed framework tackles each task with an agent, thus conquering insufficient exploration and achieving scalable management. Knowledge transfer and progressive learning are employed to improve training efficiency and stability, respectively. We apply collaborative training to eliminate distribution mismatch by refining value approximators and policies of agents alternately. Extensive experiments show that the proposed framework can learn effective resource allocation policies stably and efficiently and outperform other methods in network utility maximization and QoS assurance, which improves the network utility by 25% and 8% compared with the random strategy and the ADMM strategy, respectively. Furthermore, we validate that our framework is more robust to changes in network traffic conditions including network congestion.

23 July 2024 in ACM Transactions on Sensor Networks
Verified
Journal Article
Open access
MGMiao Guo
Miao Guo
Shibo HeShibo He
Shibo He
Chaojie GuChaojie Gu
Chaojie Gu
XGXiuzhen Guo
Xiuzhen Guo
JCJiming Chen
Jiming Chen
TWTongtong Wang
Tongtong Wang
Flow scheduling plays a pivotal role in enabling Time-Sensitive Networking (TSN) applications. Current flow scheduling mainly adopts a centralized scheme, posing challenges in adapting to dynamic network conditions and scaling up for larger networks. To address these challenges, we first thoroughly analyze the flow scheduling problem and find the inherent locality nature of time scheduling tasks. Leveraging this insight, we introduce the first distributed framework for IEEE 802.1Qbv TSN flow scheduling. In this framework, we further propose a multi-agent flow scheduling method by designing Deep Reinforcement Learning (DRL)-based route and time agents for route and time planning tasks. The time agents are deployed on field devices to schedule flows in a distributed way. Evaluations in dynamic scenarios validate the effectiveness and scalability of our proposed method. It enhances the scheduling success rate by 20.31% compared to state-of-the-art methods and achieves substantial cost savings, reducing transmission costs by 410× in large-scale networks. Additionally, we validate our approach on edge devices and a TSN testbed, highlighting its lightweight nature and ease of deployment.


1 January 2026 in IEEE Network
Verified
Journal Article
Closed access
YSYifei Sun
Yifei Sun
QCQianhan Chen
Qianhan Chen
MGMiao Guo
Miao Guo
Chaojie GuChaojie Gu
Chaojie Gu