Online Learning based Uplink Scheduling in HetNets with Limited Backhaul Capacity

Han，Zhenhua1; Tan，Haisheng1; Wang，Rui2; Tang，Shaojie3; Lau，Francis C.M.4

10.1109/INFOCOM.2018.8486336

2018-10-08

0743-166X

Proceedings - IEEE INFOCOM

2018-April

2348-2356

Honolulu, HI, United states

345 E 47TH ST, NEW YORK, NY 10017 USA

Institute of Electrical and Electronics Engineers Inc.

Heterogeneous cellular networks (HetNets) can significantly improve the spectrum efficiency, where low-power low-complexity base stations (Pico-BSs) are deployed inside the coverage of macro base stations (Macro-BSs). Due to cross-tier interference, joint detection of the uplink signals is widely adopted so that a Pico-BS can either detect the uplink signals locally or forward them to the Macro-BS for processing. The latter can achieve increased throughput at the cost of additional backhaul transmission. However, in existing literature the delay of the backhaul links was often neglected. In this paper, we study the delay-optimal uplink scheduling problem in HetNets with limited backhaul capacity. Local signal detection or joint signal detection is scheduled in a unified delay-optimal framework. Specifically, we first prove that the problem is NP-hard and then formulate it as a Markov Decision Process problem. We propose an efficient and effective algorithm, called OLIUS, that can deal with the exponentially growing state and action spaces. Furthermore, OLIUS is online learning based which does not require any prior statistical knowledge on user behavior or channel characteristics. We prove the convergence of OLIUS and derive an upper bound on its approximation error. Extensive experiments in various scenarios show that our algorithm outperforms existing methods in reducing delay and power consumption.

其他

英语

EI ; CPCI-S

[CNS 1526638] ; Chinese Academy of Sciences[QYZDY-SSW-JSC002] ; [61502201] ; [2017YFB1003000] ; [ECCS-1247944] ; [C7036-15G]

Computer Science ; Engineering ; Telecommunications

Computer Science, Hardware & Architecture ; Engineering, Electrical & Electronic ; Telecommunications

WOS:000509768900262

20184606060748

Base stations ; Behavioral research ; E-learning ; Heterogeneous networks ; Low power electronics ; Macros ; Markov processes ; Optimization ; Scheduling

Information Theory and Signal Processing:716.1 ; Radio Systems and Equipment:716.3 ; Computer Programming:723.1 ; Management:912.2 ; Optimization Techniques:921.5 ; Probability Theory:922.1 ; Social Sciences:971

2-s2.0-85056196533

Scopus

1.University of Science and Technology of China, ,China
2.South University of Science and Technology of China, ,China
3.University of Texas at Dallas, ,United States
4.University of Hong Kong, ,Hong Kong

Han，Zhenhua,Tan，Haisheng,Wang，Rui,et al. Online Learning based Uplink Scheduling in HetNets with Limited Backhaul Capacity[C]. 345 E 47TH ST, NEW YORK, NY 10017 USA:Institute of Electrical and Electronics Engineers Inc.,2018:2348-2356.