English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In this paper, we develop a knowledge-assisted deep reinforcement learning(DRL) algorithm to design wireless schedulers in the fifth-generation (5G)cellular networks with time-sensitive traffic. Since the scheduling policy is adeterministic mapping from channel and queue states to scheduling actions, itcan be optimized by using deep deterministic policy gradient (DDPG). We showthat a straightforward implementation of DDPG converges slowly, has a poorquality-of-service (QoS) performance, and cannot be implemented in real-world5G systems, which are non-stationary in general. To address these issues, wepropose a theoretical DRL framework, where theoretical models from wirelesscommunications are used to formulate a Markov decision process in DRL. Toreduce the convergence time and improve the QoS of each user, we design aknowledge-assisted DDPG (K-DDPG) that exploits expert knowledge of thescheduler design problem, such as the knowledge of the QoS, the targetscheduling policy, and the importance of each training sample, determined bythe approximation error of the value function and the number of packet losses.Furthermore, we develop an architecture for online training and inference,where K-DDPG initializes the scheduler off-line and then fine-tunes thescheduler online to handle the mismatch between off-line simulations andnon-stationary real-world systems. Simulation results show that our approachreduces the convergence time of DDPG significantly and achieves better QoS thanexisting schedulers (reducing 30% ~ 50% packet losses). Experimental resultsshow that with off-line initialization, our approach achieves better initialQoS than random initialization and the online fine-tuning converges in fewminutes.

Knowledge-Assisted Deep Reinforcement Learning in 5G Scheduler Design: From Theoretical Framework to Implementation