BiLSTM Based Reinforcement Learning for Resource Allocation and User Association in LTE-U Networks
- PDF / 1,510,254 Bytes
- 13 Pages / 439.37 x 666.142 pts Page_size
- 26 Downloads / 222 Views
BiLSTM Based Reinforcement Learning for Resource Allocation and User Association in LTE‑U Networks Zhikun Luo1 · Guanding Yu1
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract LTE-unlicensed (LTE-U) technology is a promising innovation to extend the capacity of cellular networks. The primary challenge for LTE-U is the fair coexistence between LTE systems and the incumbent WiFi systems. In this paper, we aim to maximize the long-term average per-user LTE throughput with long-term fairness guarantee by jointly considering resource allocation and user association on the unlicensed spectrum within a prediction window. We first formulate the problem as an NP-hard combinatorial optimization problem, then reformulate it as a non-cooperative game by applying the penalty function method. To solve the game, a novel reinforcement learning approach based on Bi-directional LSTM neural network is proposed, which enables small base stations (SBSs) to predict a sequence of future actions over the next prediction window based on the historical network information. It is shown that the proposed approach can converge to a mixedstrategy Nash equilibrium of the studied game and ensure the long-term fair coexistence between different access technologies. Finally, the effectiveness of the proposed algorithm is demonstrated by numerical simulation. Keywords LTE-U · Unlicensed resource allocation · Bi-directional LSTM (BiLSTM) · Reinforcement learning
1 Introduction With the prominent proliferation of mobile devices and multimedia applications, the mobile data traffic has been dramatically increased in the past few years [1]. The capacities of cellular networks have to be improved to meet the growing demands for communication resources. However, many existing capacity-excavating techniques on the licensed spectrum are limited due to the scarcity of licensed spectrum and the sophisticated intercell interference. In light of this, a promising technology named LTE-unlicensed (LTE-U) has been initiated as part of * Zhikun Luo [email protected] Guanding Yu [email protected] 1
College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310027, China
13
Vol.:(0123456789)
Z. Luo, G. Yu
LTE Release 13 and beyond by the 3rd Generation Partnership Project (3GPP) to allow users to access both licensed and unlicensed spectra under a unified LTE network infrastructure [2]. While having the potential to provide larger capacity and better coverage, LTE-U technology will inevitably cause performance degradation on incumbent WiFi systems which is severe if left without management. Thus, a fair and harmonious coexistence mechanism on unlicensed spectrum between different networks should be carefully designed. There are plenty of existing studies on LTE/WiFi coexistence under different scenarios. The authors in [3] have proposed a fair listen-before-talk (LBT) algorithm for the coexistence of LTE and WiFi. In [4], another mechanism called duty-cycle has been developed to guarantee th
Data Loading...
