Q-Learning-based Edge Node Resource Allocation Algorithm in the Environment of Power Distribution Internet of Things

2021 ◽

Vol 16 (2) ◽

Author(s):

Huashuai Zhang ◽

Tingmei Wang ◽

Haiwei Shen

Keyword(s):

Resource Allocation ◽

Reinforcement Learning ◽

Data Traffic ◽

Wireless Data ◽

Resource Allocation Algorithm ◽

Allocation Algorithm ◽

Q Learning ◽

Dense Networks ◽

Target Network ◽

Wireless Resource Allocation

The resource optimization of ultra-dense networks (UDNs) is critical to meet the huge demand of users for wireless data traffic. But the mainstream optimization algorithms have many problems, such as the poor optimization effect, and high computing load. This paper puts forward a wireless resource allocation algorithm based on deep reinforcement learning (DRL), which aims to maximize the total throughput of the entire network and transform the resource allocation problem into a deep Q-learning process. To effectively allocate resources in UDNs, the DRL algorithm was introduced to improve the allocation efficiency of wireless resources; the authors adopted the resource allocation strategy of the deep Q-network (DQN), and employed empirical repetition and target network to overcome the instability and divergence of the results caused by the previous network state, and to solve the overestimation of the Q value. Simulation results show that the proposed algorithm can maximize the total throughput of the network, while making the network more energy-efficient and stable. Thus, it is very meaningful to introduce the DRL to the research of UDN resource allocation.

Download Full-text

Deep Grid Scheduler for 5G NB-IoT Uplink Transmission

Security and Communication Networks ◽

10.1155/2021/5263726 ◽

2021 ◽

Vol 2021 ◽

pp. 1-10

Author(s):

Han Zhong ◽

Ruize Sun ◽

Fengcheng Mei ◽

Yong Chen ◽

Fan Jin ◽

...

Keyword(s):

Resource Allocation ◽

Internet Of Things ◽

Resource Utilization ◽

Smart Cities ◽

Dynamic Resource Allocation ◽

Smart Manufacturing ◽

Resource Allocation Algorithm ◽

Allocation Algorithm ◽

Environment Model ◽

Dynamic Resource

Since the birth of narrowband Internet of Things (NB-IoT), the Internet of Things (IoT) industry has made a considerable progress in the application for smart cities, smart manufacturing, and healthcare. Therefore, the number of UEs is increasing exponentially, which brings considerable pressure to the efficient resource allocation for the bandwidth and power constrained NB-IoT networks. In view of the conventional algorithms that cannot dynamically adjust resource allocation, resulting in a low resource utilization and prone to resource fragmentation, this paper proposes a double deep Q-network (DDQN)-based NB-IoT dynamic resource allocation algorithm. It first builds an NB-IoT environment model based on the real environment. Then, the DDQN algorithm interacts with the NB-IoT environment model to learn and optimize resource allocation strategies until it converges to the optimum. Finally, the simulation results show that the DDQN-based NB-IoT dynamic resource allocation algorithm is better than the traditional algorithm in the resource utilization, average transmission rate, and UE average queuing time.

Download Full-text