arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2611.06271cs.LGeess.SP
收藏

用于单步MIMO-OFDM信道估计的一致性学习与蒸馏

Consistency Learning and Distillation for One-Step MIMO-OFDM Channel Estimation

KTH Royal Institute of Technology(KTH皇家理工学院)

Fredrik Nguyen, Qiwen Wang

AI总结:

针对MIMO-OFDM系统单步信道估计的高推理成本问题,提出CMT等方法,在降低模型规模和计算量的同时,显著提升了符号错误率性能。

摘要:

迭代神经信道估计器可在导频污染下提升恢复性能,但重复的网络评估会增加推理成本。本文针对多输入多输出正交频分复用(MIMO-OFDM)系统的单步信道估计,提出了观测锚定一致性和一致性中期训练(CMT)估计器。两种估计器均在延迟-波束域中优化导频衍生的最小二乘信道估计。一致性学习将沿噪声到干净路径的端点预测耦合,CMT则优化经一致性训练的紧凑学生模型,使其从中间轨迹状态预测流匹配教师的端点。在评估的信噪比范围内,与单步流匹配基线相比,紧凑的CMT学生将未编码符号错误率(SER)降低9.6%-20.2%(对数SER降低0.44-0.98 dB),同时参数减少74.5%,网络乘积累加操作减少73.0%。

英文摘要:

Iterative neural channel estimators can improve recovery under pilot contamination, but repeated network evaluations increase inference cost. This paper develops observation-anchored consistency and Consistency Mid-Training (CMT) estimators for one-step channel estimation in multiple-input multiple-output orthogonal frequency-division multiplexing (MIMO-OFDM) systems. Both estimators refine pilot-derived least-squares channel estimates in the delay-beam domain. Consistency learning couples endpoint predictions along noisy-to-clean paths. CMT then refines a compact consistency-trained student to predict a flow-matching teacher's endpoint from intermediate trajectory states. Across the evaluated SNR range, the compact CMT student reduces uncoded symbol error rate (SER) by 9.6-20.2% relative to the one-step flow-matching baseline (0.44-0.98 dB in logarithmic SER reduction), while using 74.5% fewer parameters and 73.0% fewer network multiply-accumulates.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑