In this paper, we propose a feedback-efficient hybrid precoding framework for wideband millimeter-wave (mmWave) multiple-input multiple-output orthogonal frequency-division multiplexing (MIMO-OFDM) systems. To mitigate the high cost of radio frequency (RF) chains and channel state information (CSI) feedback in large-scale antenna arrays, we first construct frequency-flat analog precoders by extracting dominant angle-of-arrival (AoA) and angle-of-departure (AoD) directions from sparse frequency-domain channels. For digital precoding, we design a quantized codebook using the Lloyd algorithm and develop a binary-search-based hierarchical interpolation algorithm that adaptively assigns codewords according to subcarrier correlation. The proposed method achieves sub-linear feedback scaling by reducing the feedback overhead from O(K) to O(K/M + log M), where K is the number of subcarriers and M is the pilot spacing. Simulation results demonstrate that the proposed method achieves comparable or superior spectral efficiency and bit error rate (BER) performance to existing clustering and interpolation schemes, while significantly reducing computational complexity and exhibiting robustness under imperfect CSI.

翻译：本文针对宽带毫米波多输入多输出正交频分复用（MIMO-OFDM）系统，提出一种反馈高效的混合预编码框架。为缓解大规模天线阵列中射频链路与信道状态信息反馈的高昂成本，我们首先从稀疏频域信道中提取主导到达角与离开角方向，构建频率平坦的模拟预编码器。针对数字预编码，采用Lloyd算法设计量化码本，并开发基于二分搜索的层级插值算法，该算法根据子载波相关性自适应分配码字。所提方法将反馈开销从O(K)降至O(K/M + log M)（其中K为子载波数，M为导频间距），实现次线性反馈扩展。仿真结果表明，在显著降低计算复杂度的同时，该方法在频谱效率和误码率性能上与现有聚类及插值方案相当或更优，并在非理想信道状态信息下具有鲁棒性。