Keliang Du

Dr. Keliang Du

Beijing University of Posts and Telecommunications, Beijing, China
+1
orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
25 September 2023 in IEEE Transactions on Network and Service Management
Verified
Journal Article
Closed access
YLYu Liu
Yu Liu
LWLuhan Wang
Luhan Wang
Zhaoming LuZhaoming Lu
Zhaoming Lu
Keliang DuKeliang Du
Keliang Du
GSGuochu Shou
Guochu Shou
Integrating terrestrial cellular network with Low Earth Orbit (LEO) satellite constellation has been a popular trend in beyond 5G and 6G era, called Satellite-Terrestrial Integrated Core Network (STICN). Core Network (CN) is an essential component responsible for authentication, security, mobility, data routing, etc. However, the terrestrial CN is designed for infrastructure-fixed and user-moving scenarios, which would cause the STICN to experience signaling storms and service interruptions when users access by rapid satellites. In this paper, we propose a distributed lightweight stateless satellite CN architecture, which could fit with the dynamic and limited-resource instincts of LEO satellites. And it can cooperate with terrestrial CN to provide seamless services. Firstly, the contexts of Network Functions (NF) are decoupled from itself and are managed in a common repository. Moreover, we design a cooperation mechanism between NFs to avoid frequent transmission of context and service interruption. Finally, extensive experiments are carried out on semi-physical simulation environments. The STICN performance could be improved by selecting the optimal number and location of each NF. Our evaluation shows that the proposed scheme could reduce the delay of the handover procedure by 37% and is more resilient compared with terrestrial CN.

1 June 2023 in Digital Communications and Networks
Verified
Journal Article
Open access
Keliang DuKeliang Du
Keliang Du
LWLuhan Wang
Luhan Wang
XWXiangming Wen
Xiangming Wen
YLYu Liu
Yu Liu
HNhaiwen niu
haiwen niu
SHShaoxin Huang
Shaoxin Huang

11 December 2020
Verified
Conference Paper
Closed access
Keliang DuKeliang Du
Keliang Du
XWXiangming Wen
Xiangming Wen
LWLuhan Wang
Luhan Wang
TNTien-Thinh Nguyen
Tien-Thinh Nguyen
The fifth generation of mobile communication network core network adopts micro-service architecture, overturning the original core network architecture, making it easier for 5G networks to realize network slicing, supporting large planning and high-speed terminal access. Many studies have been constructed on 5G technologies, but it is difficult to validate these theories and algorithms in a real-world environment. Therefore, we aim to build the open-source end-to-end 5G mobile communication system platform. In this paper, we develop a cloud-native based implementation of Access and Mobility Management Function (AMF) under the OAI 5G Core project. We firstly review the design principles of cloud-native applications and its relationship with serviced 5G core network functions. And then, the designed application architecture for the cloud-native based AMF and detailed implementation of system procedures, e.g. initial registration procedure, are described. In the end, the AMF is tested in a real E2E 5G system for its functionalities. Also, it's deployed in a cloud environment, tested with simulated gNB/UEs., for its stability and concurrency capacity.

1 December 2022 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Keliang DuKeliang Du
Keliang Du
LWLuhan Wang
Luhan Wang
YLYu Liu
Yu Liu
HNhaiwen niu
haiwen niu
SHShaoxin Huang
Shaoxin Huang
XWXiangming Wen
Xiangming Wen
The future mobile communication networks (beyond 5th generation (5G)) are evolving toward the service-based architecture where network functions are fine-grained, thereby meeting the dynamic requirements of diverse and differentiated vertical applications. Consequently, the complexity of network management becomes higher, and artificial intelligence (AI) technologies can improve AI-native network automation with their ability to solve complex problems. Specifically, deep reinforcement learning (DRL) technologies are considered the key to intelligent network automation with a feedback mechanism similar to that of online closed-loop architecture. However, the 0-delay assumptions of the standard Markov decision process (MDP) of traditional DRL algorithms cannot directly be adopted into real-world networks because there exist random delays between the agent and the environment that will affect the performance significantly. To address this problem, this paper proposes a random-delay-corrected framework. We first abstract the scenario and model it as a partial history-dependent MDP (PH-MDP), and prove that it can be transformed to be the standard MDP solved by the traditional DRL algorithms. Then, we propose a random-delay-corrected DRL framework with a forward model and a delay-corrected trajectory sampling to obtain samples by continuous interactions to train the agent. Finally, we propose a delayed-deep-Q-network (delayed-DQN) algorithm based on the framework. For the evaluation, we develop a real-world cloud-native 5G core network prototype whose management architecture follows an online closed-loop mechanism. A use case on the top of the prototype namely delayed-DQN-enabled access and mobility management function (AMF) scaling is implemented for specific evaluations. Several experiments are designed and the results show that our proposed methodologies perform better in the random-delayed networks than other methods (e.g., the standard DQN algorithm).

2 November 2024 in Electronics
Verified
Journal Article
Open access
PDF file
MHMing He
Ming He
Keliang DuKeliang Du
Keliang Du
HHHaoran Huang
Haoran Huang
QSQi Song
Qi Song
XLXinyu Liu
Xinyu Liu
Drones integrated with 5G New Radio (NR) base stations have emerged as a promising solution for efficient victim search and localization in emergency zones where cellular networks are disrupted by natural disasters. Traditional approaches relying solely on uplink Sounding Reference Signal (SRS) for localization face limitations due to User Equipment (UE) power constraints. To overcome this, our paper introduces BWSAR, a novel three-stage Search-and-Rescue (SAR) methodology leveraging 5G-NR beam sweeping technologies. BWSAR utilizes downlink Synchronization Signal Block (SSB) for coarse-grained direction estimation, guiding the drone towards potential victim locations. Subsequently, finer-grained beam sweeping with Positioning Reference Signal (PRS) is employed within the identified direction, enabling precise three-dimensional UE coordinate estimation. Furthermore, we propose a trajectory optimization algorithm to expedite the drone’s navigation to emergency areas. Simulation results underscore BWSAR’s efficacy in reducing positioning errors and completing SAR missions swiftly, within minutes.

16 May 2022
Verified
Conference Paper
Closed access
YYYunan Yan
Yunan Yan
Keliang DuKeliang Du
Keliang Du
LWLuhan Wang
Luhan Wang
HNHaiwen Niu
Haiwen Niu
XWXiangming Wen
Xiangming Wen
Cloud network integration (CNI) has been a new paradigm to better support diverse vertical applications. The virtualized mobile network deployed in private and public clouds is regarded as the trend of future network evolution. However, it is challenging for radio access network (RAN) protocols to be deployed in public clouds because of the strict requirements for stable cloud resources. In a CNI environment, there coexist a large number of services (e.g. network services and cloud services) and frequent task scheduling will result in a great deal of resources fluctuation, thus degrading RAN performance. To the best of our knowledge, current researches in CNI interests ignore the high processing requirements of RAN. Therefore in this paper, we propose a multi-pass deep Q network (MP-DQN) based short term task scheduling strategy to minimize the quality of service (QoS) fluctuation of RAN deployed in public clouds. First, taking into account the differences in the relationships between resources and QoS among various services, we formulated a continuous decision problem of task scheduling. Then, We employ MP-DQN to solve the decision problem, jointly optimizing the services QoS and the task scheduling success rate. We conduct a real-world experiment to obtain the cloud RAN CPU-QoS model. The experimental results reveal that our proposed MP-DQN based task scheduling strategy performs significantly better in minimizing RAN QoS fluctuation than the conventional task scheduling strategy.