| Laser Detection and Ranging system(Li DAR)has excellent 3D imaging capability,it will be widely used in the fields of unmanned aerial vehicles,robots and car-assisted driving,pulsed Li DAR has more advantages in long distance ranging.In order to meet the application requirements,the pulse Li DAR is developing towards integration,miniaturization,solid state,high precision and low power consumption.By studying the characteristics of the classical signal processing circuit architecture of pulsed Li DAR,this paper proposes a new hybrid circuit architecture based on Time to Digital Converter(TDC)and Sample circuits(SH)and an adaptive noise reduction optimization method.The design verification of low-power signal processing circuit with high-speed sampling function and the research of new adaptive denoising algorithm are carried out to solve the problems of large area of pulse full waveform architecture,high power consumption and large influence of noise on output signal.The main work is as follows:To solve the problem of power redundancy caused by unbalanced sampling and output frequency in full-wave architecture,an adjustable sampling circuit of 0.8 ~3GHz is designed.The threshold trigger mechanism is adopted to control the high speed sampling and TDC circuit only working within the effective time of about 4% of the whole cycle.Then the signal is read out by the low-speed Analog to Digital Converter(ADC).It can effectively reduce the overall power consumption of the circuit.The sampling power can be reduced by 33.7 % when the time period is 0.2 ?s.In order to solve the problem of insufficient dynamic range of Li DAR system,the threshold discriminating circuit is used to make up for the shortage of short-range detection capability of high-gain fullwave architecture and effectively expand the dynamic range of the system.The optimization method is divided into two steps.The first step is adaptive noise estimation.After changing the sampling point from one-dimensional signal to twodimensional matrix,high-precision noise estimation is completed based on the boundedness and asymptotic property of noise eigenvalue spectrum.Then,key parameters of the algorithm are optimized according to sampling frequency,signal gradient change and noise level,so as to achieve stable noise reduction at different noise levels and sampling frequencies.In order to verify the effectiveness of the optimization method,two adaptive noise reduction algorithms are designed.An adaptive gaussian filtering algorithm is proposed for fast noise reduction.At the same time,an adaptive guided noise reduction algorithm is proposed for the applications that requires high accuracy but is insensitive to processing speed.Compared with the original algorithm,the optimized algorithm can significantly improve the fitting degree under various noise levels and waveform gradients.With chrt 0.18 ?m CMOS process,the composite architecture is divided into two parts,analog front-end circuit module and digital sampling module,to tape out.The analog front-end module includes TIA and high-speed comparator,and the digital sampling module includes TDC and high-speed sampling circuit.Through simulation,the circuit has 150 MHz bandwidth and 83.6 d B gain,the noise of the circuit is 8.37 pA/2/1Hz,and the sampling frequency can achieve 0.8 ~3 GHz adjustable,meeting the requirements of miniaturization and low power consumption of Li DAR system.The error of the noise estimation algorithm is less than 0.1% through simulation and indoor echo test of single and double targets.The proposed two adaptive denoising methods have a fitting degree of more than 99.02% within the sampling frequency of 0.5 ~15 GHz and the signal-to-noise ratio of 10 ~35 d B,which can realize the self-matching of pulse Li DAR parameters of different sampling frequencies in the variable noise environment. |
PDF Full Text Request