超宽带采样技术是太赫兹通信和雷达、高端测量仪器、毫米波脉冲成像等前沿应用的基础。频率交织技术能够突破时间交织架构在模拟带宽和时钟稳健性方面的关键技术瓶颈，在未来超宽带采样领域拥有广阔的发展前景。目前频率交织采样大多基于混合滤波器组，由于结构复杂、集成难度大，相关研究仍面临诸多挑战。为实现频率交织系统电路规模的大范围缩减和高密度集成，本项目拟提出一种基于改进松弛迭代的全数字实时样点重构方法，并针对该方法在实现过程中存在的关键技术问题开展深入研究：首先研究简化的多通道频率交织系统中本振泄露与相位噪声在各交叠子带间的传播规律和作用机理；据此建立完备的广义时变非线性系统非一致性误差混合行为模型；最后结合自适应参数辨识研究实时在线的样点重构方法，最大程度实现电路非理想性修正的数字化。预期成果将推动新一代采集系统向小型化、轻量化、超大带宽（DC到亚太赫兹全频段实时精确获取）方向的发展。

Ultrawideband sampling are key enabling blocks for emerging applications in Terahertz communication and radar, high-end instrumentation, pulsed mm-wave imaging. Frequency-interleaved analog-to-digital conversion (FI-ADC) offers the opportunity to relax the bandwidth requirement on samplers and reduce the clock jitter susceptibility in time-interleaved ADCs. For higher linearity and integration density, a novel sampling framework based on fully digital real-time signal recovery, namely LPF-less (free of lowpass filters) FI-ADC, is proposed for the first time. It extensively curtails the RF circuit scale in hybrid filterbank based FI-ADCs without a loss of system transient performance. The main contributions of this project are threefold: (a) A sound analysis and modeling of LO leakage and phase noise in multichannel LPF-less FI-ADCs will be studied with the corresponding harmonic rejection and phase noise desensitization methods; (b) This project will also investigate the hybrid system behavioral model which includes multiple time-varying linear and nonlinear mismatch characteristics among channels; (c) Novel digital iterative structures for adaptive online identification, estimation and correction will be studied, and a high density RF/digital hybrid integration scheme will therefore be realized. The overall project objective is to promote the development of the next generation acquisition systems in the direction of miniaturization, lightweight and very large bandwidth (from DC to sub-THz). The simple implementation of circuit structure makes the LPF-less FI-ADC architecture promising for integration into a chip.