Simulated Bifurcation Algorithm for MIMO Detection Wen Zhang1 and Yu-Lin Zheng1 1Hisilicon research Huawei Technologies Co. Ltd. Shenzhen China

2025-05-03 0 0 363.03KB 4 页 10玖币
侵权投诉
Simulated Bifurcation Algorithm for MIMO Detection
Wen Zhang1
, and Yu-Lin Zheng1
1Hisilicon research, Huawei Technologies Co., Ltd., Shenzhen, China
Abstract
We study the performance of the simulated bifurcation (SB) algorithm for signal detection in multiple-
input multiple-output (MIMO) system, a problem of key interest in modern wireless communication systems.
Our results show that SB algorithm can achieve significant performance improvement over the widely used
linear minimum-mean square error decoder in terms of the bit error rate versus the signal-to-noise ratio, as
well as performance improvement over the coherent Ising machine based MIMO detection method.
1. Introduction
Optimal maximum likelihood MIMO (ML-MIMO) detection is an NP-hard problem. So ML-MIMO detector
like sphere decoder is impractical to be implemented for large number of antennas because of exponential
computational complexity. Suboptimal linear decoders with polynomial complexity, such as minimum-mean
square error (MMSE) decoder, enable near-optimal bit error rate (BER) performance in Massive MIMO regime,
in which the number of antennas is much larger than the number of users. While the BER performance is poor
in large MIMO system where the number of antennas is equal to that of users.
The ML-MIMO detection problem can be formulated into a quadratic unconstrained binary optimization
(QUBO) problem or Ising model problem [1]. Some heuristic optimization methods based on Ising model solver,
such as quantum annealer [1], simulated annealing [2], oscillator Ising machine [3] and coherent Ising machine
(CIM) [4], has been proposed to solve the ML-MIMO detection problem recently and achieve near-optimal
solutions with low computational cost. We propose a near-optimal ML-MIMO detector based on simulated
bifurcation algorithm and achieve better BER performance than above methods and also linear decoder MMSE.
Ising model and simulated bifurcation
2. Ising model and simulated bifurcation
The Hamiltonian or the cost function of Ising model is defined by 𝐻𝐼𝑠𝑖𝑛𝑔 =𝑁
𝑖,𝑗 𝐽𝑖𝑗 𝑠𝑖𝑠𝑗+𝑁
𝑖𝑖𝑠𝑖, where
spins 𝑠𝑖∈ −1,1.Simulated bifurcation algorithm simulates adiabatic evolution of a classical nonlinear Hamilto-
nian system which contains Hamiltonian of an Ising model and approximates the optimal solution of correspond-
ing Ising model [?]. The nonlinear system Hamiltonian and the evolution described by equation of motions of
Corresponding author. E-mail:zhangwen20@huawei.com
1
arXiv:2210.14660v1 [cs.IT] 26 Oct 2022
摘要:

SimulatedBifurcationAlgorithmforMIMODetectionWenZhang1*,andYu-LinZheng11Hisiliconresearch,HuaweiTechnologiesCo.,Ltd.,Shenzhen,ChinaAbstractWestudytheperformanceofthesimulatedbifurcation(SB)algorithmforsignaldetectioninmultiple-inputmultiple-output(MIMO)system,aproblemofkeyinterestinmodernwirelesscom...

展开>> 收起<<
Simulated Bifurcation Algorithm for MIMO Detection Wen Zhang1 and Yu-Lin Zheng1 1Hisilicon research Huawei Technologies Co. Ltd. Shenzhen China.pdf

共4页,预览1页

还剩页未读, 继续阅读

声明:本站为文档C2C交易模式,即用户上传的文档直接被用户下载,本站只是中间服务平台,本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私,请立即通知玖贝云文库,我们立即给予删除!
分类:图书资源 价格:10玖币 属性:4 页 大小:363.03KB 格式:PDF 时间:2025-05-03

开通VIP享超值会员特权

  • 多端同步记录
  • 高速下载文档
  • 免费文档工具
  • 分享文档赚钱
  • 每日登录抽奖
  • 优质衍生服务
/ 4
客服
关注