| Single-photon detector(SPD)is one of the key devices in the fields of quantum information,laser radar and bio-medical science.Single-photon detectors based on InGaAs(P)/InP avalanche photodiodes(APDs)are suitable for use in the near-infrared band,and are fast in response,low in cost,small in size.They require less cooling,and are easy-to-use in practice.However,their detection efficiency is comparatively low and their afterpulse probability is high,compared to superconductive nanowire single-photon detectors,and PMTs and Si APDs in visible wavelengths.Single-photon detectors are often used in applications such as quantum communication,laser radar,and fluorescence lifetime analysis.Different applications have different requirements for the performance and working conditions of the detectors,and their performance indicators are greatly affected by external parameters.The relationship between single-photon detector performance and its operating mode and parameters,especially the relationship between afterpulsing effects and various parameters for different applications using different single-photon detection technology,has important research significance and application value.In this paper,near-infrared free-running SPDs and gated SPDs based on InGaAs(P)/InP APDs were developed.The test methods and influencing factors of the performance,especially afterpulse effect of SPDs were studied in detail.The performance of the detectors and their effect on system performance were compared in laser ranging radar applications.The main contents are as follows:1.High-performance free-running detectors were developed,with the advantages of existing quenching and recovery circuits.A variety of different AQAR circuits for avalanche extraction and quenching at the anode or cathode of APDs were designed,and different circuit combinations had different quenching delays and excess bias voltages.Avalanche quenching performance of different circuits was compared in order to improve the circuit structure.Ultra-low-latency AQAR circuits were designed using commercially available SiGe HBT IC comparators,high-speed E-pHEMT RF transistors,and capacitance-balanced noise cancelling circuits,in which the latch function of the comparator was used for latching the quenched status of the APD,and the delay of the negative feedback loop was greatly lowered.Capacitance-balanced noise cancelling technique was introduced to the circuit to suppress the transient noise.The avalanche quenching time of the improved AQAR circuit was approximately 1 ns,which significantly improved the performance of designed free-running detectors.And it is compatible with gating function for applications and performance studies.A free-running detector was developed with the AQAC and PGA-284 InGaAsP/InP APD.The afterpulse probability and dark count rate were only approximately 4%and 482 Hz,respectively,at-40?,10%detection efficiency,and 1 ?s dead time.The shortest dead time of the detector was 35 ns,and the maximum detection efficiency was 25%.The overall performance of the detector has reached or even surpassed other detectors based on active quenching technology and free-running detectors based on negative-feedback avalanche diodes(NFADs),and has reached best-in-class performance in the world.2.In order to facilitate the study of the performance of high-speed gated SPDs,which has a lower afterpulse probability compared to free-running detectors,gated single-photon detectors with adjustable gate frequency were developed.Short-pulse-shaping circuit and double-stage RF amplifiers were designed for large-amplitude short gating pulse generation.A novel noise-cancelling technology was proposed,which is based on a gated-passive-quenching circuit(GPQC)with balanced APD-PIN junction capacitance,and Bessel low-pass filters,combining the advantages of various noise-cancelling techniques including balancing and filtering.The proposed GPQC had a noise suppression ratio of 34 dB when using a gating frequency of 200 MHz and below,and the insertion loss of it was low.When using a gating frequency of 200 MHz and below,the high-speed gated SPD using this technique had a dark count probability of 4×10-6/gate at 10%detection efficiency,and its afterpulse probability did not exceed 2%,and its maximum available detection efficiency was 25%.When using a gating frequency of 1.25 GHz,the dark count probability and afterpulse probability were 1.84×10-6/gate and 1%,respectively,at the detection efficiency of 10%,and the maximum available detection efficiency was 29%.The detector has good overall performance,and is low in cost,easy to use,and is convenient for parameter tuning.In general,it is suitable for performance study of the SPD.3.An automated SPD performance test device was designed,and the evaluation and characterization methods for the performance parameters of SPDs were systematically summarized and developed.This part mainly contains the following contents:(1)A single-photon detector performance test system was set up,and a FPGA-based performance test module embedded in the single-photon detector was designed,including functions such as temperature control,bias control,multi-function counter,and time-to-digital converter(TDC).Automated variable parameter measurement was achieved.Among them,the resolution of the TDC based on FPGA was 99 ps,the resolution of the TDC based on TDC-GPX was 25 ps.(2)A method for avalanche pulse waveform measurement based on APD self-luminescence and variable discrimination thresholds was developed.The APD's self-luminescence during avalanche process was measured using the time-correlated single-photon counting technique to obtain the avalanche pulse waveform,and the actual amplitude of the avalanche pulse was calculated by changing the discrimination threshold.The characteristics of the avalanche pulse were characterized by this method and its relationship with the bias voltage was obtained.Similarly,the avalanche pulse amplitude distribution of the gated single-photon detector and its relationship with the bias voltage were obtained by measuring the avalanche count rate at different discrimination thresholds.These two methods provided strong supports for the subsequent detection efficiency and afterpulse probability study.(3)An afterpulse probability measurement method using in-laser-period count statistics was proposed.The proposed method collects the statistics of the count in each laser period to evaluate detection efficiency and afterpulse probability,in order to reduce the influence of dark count rate fluctuations on the measurement results.One can use this method to evaluate the performance of the single-photon detector with lower cost and faster speed.(4)The variation of dark count rate and afterpulse probability with temperature and bias voltage were experimented,and main influencing factors were obtained.Results of dark count study showed that the main source of dark counts for non-preferred devices was the afterpulses generated by dark counts,avalanche pulses induced by defects and impurities,and intrinsic carriers in InGaAs;dark counts induced by intrinsic carriers in InGaAsP were relatively low in number.The afterpulse probability was studied in detail.The afterpulse probability in the wide time range from 20 ns to 12?s was studied in several range segments.Analysis was done by changing bias voltage and temperature and integrating the data.Results showed that the carrier release rate increased with the increase of temperature,the carrier capture rate decreased with the increase of temperature or excess bias,and the activation energy of the trapping energy level was calculated.Verification and development of existing carrier release rate model and activation energy model were also performed.(5)The performance of the detectors with multiple incident photons were also studied.The results show that single-photon detection efficiency and afterpulse probability of high-speed gated detectors increased in such condition,and the reason of the change in performance was analyzed.4.For the rapid development of laser radar applications,the performance of free-running detectors and high-speed gated single-photon detectors in non-coaxial laser radar system was compared.The influences of dark count,afterpulsing,and the masking effect of free-running detectors on the performance parameters such as peak signal-to-noise ratio(PSNR)of laser ranging radar system were studied.Results showed that for the applications where timing of the incident photons is not predictable,the actual detection efficiency of the high-speed gated single-photon detector was as low as 20%of its peak detection efficiency.Nevertheless,the PSNR was comparable with free-running detectors with a moderate hold-off time.The performance of detecting a target with free-running detectors was affected by both dark counts and previous targets,which was called"masking effects".When hold-off time was shorter than 500 ns,the detection efficiency and PSNR began to drop quickly.Thus,the high-speed gated single-photon detector worked well in "quasi-continuous" mode,but the actual detection efficiency was low,and thus it was only suitable for situations where sufficient photon counts can be accumulated in a short time.Results also showed that the overall performance of the free-running SPD was good when the hold-off time was about 1 ?s,but it should be noted that the number of received photons should be properly reduced so that the detection probability of the target can be kept at a lower order to reduce the masking effect.The main innovations in the paper are summarized as follows:1.An ultra-low-delay active quenching active recovery circuit were designed to significantly improve the performance of free-running detectors and was compatible with gated mode.The performance of the free-running detector developed based on PGA-284 InGaAsP/InP APD reached or even partially surpassed that of other active-quenching-based detectors and NFAD-based free-running detectors.2.A transient gate noise suppression circuit was developed based on balanced-capacitance method.An avalanche pulse extraction circuit with low insertion loss and high rejection ratio was designed based on the combination of balanced APD-PIN junction capacitance and low-pass filtering.With this circuit,the designed high-speed detectors had good overall performance,and the gating frequency and other parameters in them were easy to adjust and easy to use.3.A quantitative indirect measurement method for avalanche pulse waveform was developed.The method of measuring the avalanche pulse waveform based on APD self-luminescence is improved.The time-correlated single photon counts of APD self-luminescence under different discrimination thresholds were used to reconstruct the waveform of the avalanche pulse,and the actual amplitude was then calculated.This method provides strong support for the research of detection efficiency and afterpulse probability.4.A statistical measurement method was proposed to estimate the detection efficiency and the afterpulse probability using the statistics of the avalanche pulse counts in each laser pulse period,which reduced the influence of the dark count rate fluctuation on the afterpulse probability measurement result,and reduced the complexity of hardware implementation and calculation.The performance of the detector can be evaluated quickly in real time at low cost,while maintaining similar accuracy with other methods.5.Based on the measurement results of the variation of the afterpulse probability with temperature and bias voltage in a wide hold-off time range,afterpulsing characteristics of APD and its influencing factors were studied in depth.The characteristics of the release rate and the trapping rate of the carriers were obtained,and the calculation of trap activation energy was performed.6.The performance of high-speed gated detectors and free-running detectors were compared in a laser ranging system,in which applicable conditions of different detectors were obtained. |
PDF Full Text Request