We have designed a system which has an 8X8 array of photo diodes, we sense 10ns pulses of laser falling on those photo diodes with a repetetion rate of maximum 50Khz i.e. 20us, we were successfully able to sense them and detect them,please note that our system only senses those laser pulses and gives out the frequency of those pulses and amplitude/energy of those pulses, but doesn't do any kind of imaging

but now the problem statement has become a bit complex, for increased requirements we are forced to switch to a 64X64 array model, now we have to have 4096 stages of processing which will make the design very bulky

now i am thinking of solving this problem with a different approach, so in this need i need some suggestions, what alternatives can solve the problem by reducing the electronics stages

i thought of solving it by using a line scanning mechanism where 64 stages of processing will be given to 64 rows and 64 columns will be switched to use the same stages, but with this it will be very difficult to sense the 10ns pulses, because my sampling rate for each row should be at least 2GHz to acquire at least 10samples of the pulse the total sampling rate for scanning all columns is 2GHz X 64 = 128GHz sampling rate, working at this rate is very costly affair and over ambitious to go at a frame rate of 2GHz

please throw some light by suggesting any alternate techniques that these to solve the problem of increased replica of electronic stages

  • $\begingroup$ yes ofcourse our pulse repetetion rate is maximum 50Khz which is 20uSec $\endgroup$ – kakeh Apr 17 '16 at 6:22
  • 1
    $\begingroup$ Have you looked into commercial 2-d imaging sensors (CCD)? Those have storage elements that enable slow readouts. $\endgroup$ – Olli Niemitalo Apr 17 '16 at 8:27
  • $\begingroup$ but CCDs have a different wavelength band, i need 0.5-1.8um for which i am using individual photo diodes, i am thinking of how can we construct a CCD similar array from individual photo diodes , is it possible $\endgroup$ – kakeh Apr 19 '16 at 7:20
  • $\begingroup$ You could complement each photodiode with a switchable analog integrator. See datasheet for Texas Instruments / Burr Brown IVC102: ti.com.cn/cn/lit/ds/symlink/ivc102.pdf You could do a readout of the integrating capacitors in a multiplexed fashion. $\endgroup$ – Olli Niemitalo Apr 19 '16 at 9:07
  • $\begingroup$ That's beginning to sound like imaging though. $\endgroup$ – Olli Niemitalo Apr 19 '16 at 9:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.