异构密集网络下基于POMDP负载感知的负载均衡算法研究

唐伦; 梁荣; 张亚; 陈前斌

doi:10.11999/JEIT161347

异构密集网络下基于POMDP负载感知的负载均衡算法研究

doi: 10.11999/JEIT161347

基金项目:

国家863计划项目(2014AA01A701)，国家自然科学基金(61571073)

计量
- 文章访问数: 1201
- HTML全文浏览量: 120
- PDF下载量: 220
- 被引次数: 0
出版历程
- 收稿日期: 2016-12-08
- 修回日期: 2017-03-24
- 刊出日期: 2017-09-19

Load Balance Algorithm Based on POMDP Load-aware in Heterogeneous Dense Cellular Networks

Funds:

The National 863 Program of China (2014AA　01A701), The National Natural Science Foundation of China (61571073)

摘要

摘要: 针对异构密集网络场景中业务不确定性而引起的网络负载不均衡，该文提出一种基于部分可测马尔科夫决策过程(POMDP)负载感知的负载均衡方法。该方法首先通过对系统用户在感知周期内数据包传输数量进行观察，预测出下一周期系统可能出现的负载状态。其次根据负载感知结果动态调整小区范围扩展偏置值(DCRE)，以达到优化系统整体负载均衡性的目的。最后采用启发式算法近似求解，能够快速得到次优解。仿真结果表明，该方案能在异构密集网络下提高系统负载均衡性,同时提升了系统吞吐量与系统资源利用率。
- 异构密集网络 /
- 负载均衡 /
- 小区范围扩展 /
- 部分可测马尔科夫决策过程
Abstract: In order to solve the load imbalance problem caused by uncertainty of traffic in heterogeneous dense cellular networks, this paper proposes a load balance algorithm through small cell range expansion. The proposed algorithm is based on Partially Observable Markov Decision Process (POMDP). By observing the packets of system user during the perceptual cycle, the next cycle system possible load state can be dopted. Then, the Dynamic Cell Range Expansion (DCRE) offset value is dynamically adjusted to take action in advance, reaching the purpose of optimizing the system load balance. To solve the problem efficiently, a heuristic algorithm is used to approximate and quickly get the suboptimal solution. Simulation results show that the proposed method can achieve load balance optimization in dense hetrogeneous network, and improve the system user throughput and resource utilization rate.
- Heterogeneous dense network /
- Load balance /
- Cell range expansion /
- Partially observable Markov decision process

HTML全文

参考文献(12)

YUNAS S F, VALKAMA M, and NIEMELA M. Spectral and energy efficiency of ultra-dense networks under different deployment strategies[J]. IEEE Communications Magazine, 2015, 53(1): 90-100. doi: 10.1109/MCOM.2015.7010521

HAO Peng, YAN Xiao, Y N Ruyue, et al. Ultra dense network: Challenges, enabling technologies and new trends[J]. China Communications, 2016, 13(2): 30-40.

JIANG H L, TONG E, LI Zhihang, et al. A power adjustment based eICIC algorithm for hyper-dense HetNets considering the alteration of user association[J]. Science China Information Sciences, 2015, 58(8): 1-15.

CHEN S, XING C, and FEI Z. Distributed resource allocation in ultra-dense networks via belief propagation[J]. China Communications, 2015, 12(11): 79-91. doi: 10.1109/ CC.2015.7365891.

QIU J, WU Q, XU Y, et al. Demand-aware resource allocation for ultra-dense small cell networks: An interference-separation clustering-based solution[J]. Transactions on Emerging Telecommunications Technologies, 2016, 27(8): 1071-1086.

JIN Y, QIU L, and LIANG X. Small cells on/off control and load balancing for green dense heterogeneous networks[C]. IEEE Wireless Communications and Networking Conference (WCNC), New Orleans, LA, USA, 2015 IEEE. 2015: 1530-1535. doi: 10.1109/WCNC.2015.7127695.

YASIR B A, SU G, and BACHACHE N. Range expansion for pico cell in heterogeneous LTE-A cellular networks[C]. International Conference on Computer Science and Network Technology, Changchun, China, 2012: 1235-1240. doi: 10. 1109/ICCSNT.2012.6526147

SASIKUMAR R, ANANTHANARAYANAN V, and RAJESWARI A. An intelligent pico cell range expansion technique for heterogeneous wireless networks[J]. Indian Journal of Science and Technology, 2016, 9(9).

MOON S, KIM B, MALIK S, et al. Interference management with cell selection using cell range expansion and ABS in the heterogeneous network based on LTE-advanced[J]. Wireless Personal Communications, 2015, 81(1): 151-160.

JIANG Haiqi. System utility optimization of cell range expansion in heterogeneous cellular networks[C]. 2016 8th IEEE International Conference on Communication Software and Networks (ICCSN), Beijing, China, 2016: 412-417. doi: 10.1109/ICCSN.2016.7586694.

NAKAZAWA S, MATSUOKA S, and OTSUKA H. Adaptive control CRE and its throughput performance in HetNet[J]. IEICE Communications Express, 2015, 4(4): 117-122.

YU T C and SILVESTER J A. A four-state DMMPP for characterizing multimedia traffic with short-term and long-term correlations[C]. IEEE International Conference on Communications, IEEE, Ottawa, Canada, 1999: 880-885 vol.2. doi: 10.1109/ICC.1999.765400.

施引文献

资源附件(0)

访问统计