一种新的模拟到信息转换系统

Currently, analog-to-information conversion system (AIC) is mainly achieved by random matrix. However, the hardware implementation of the random matrix is complicated and difficult if not impossible. Inspired by the recent theory of compressive sensing (CS) and integrate-and-fire (IF) sampler, we propose a new sampling system which does not contain any random matrix and encodes in the time domain. The output signal sampled by this system can be accurately restored to the original signal. The sampling frequency of the system can be freely controlled if the signal is sparse and the sampling period is sufficiently long. Theoretically, the sampling frequency can be infinitesimal under those two conditions specified. As a result, the amount of data and the power consumption of the system can be significantly reduced. In addition, it is easy to be implemented in practice and has the advantage of no quantization error and so on. The proposed sampling system should have a great potential in applications in the field of broadband signals such as radar signal detection and biological signal measurement.