Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!utgpu!water!watnot!watmath!clyde!rutgers!seismo!cmcl2!phri!roy From: roy@phri.UUCP Newsgroups: sci.electronics,sci.physics Subject: Re: Laser eavesdropping Message-ID: <2628@phri.UUCP> Date: Wed, 1-Apr-87 11:30:12 EST Article-I.D.: phri.2628 Posted: Wed Apr 1 11:30:12 1987 Date-Received: Sat, 4-Apr-87 09:35:18 EST References: <499@sw1e.UUCP> <15260@onfcanim.UUCP> Reply-To: roy@phri.UUCP (Roy Smith) Organization: Public Health Research Inst. (NY, NY) Lines: 39 Keywords: modulation Xref: utgpu sci.electronics:447 sci.physics:1012 In article <15260@onfcanim.UUCP> dave@onfcanim.UUCP (Dave Martindale) writes: > if the mirror is mounted near the edge of the pane of glass, it will > rotate as well as translate as the window flexes. The rotation will > change the angle of the reflected beam, and this angular change may be > measurable. An idea for a detector based on this idea: Assuming that the movement of the glass is small, the angle of deflection of the beam will also be small, so you can assume that the distance the beam is deflected is in direct proportion to the angle of deflection. Make a mask with a narrow slit in it which is wide at one and and narrow at the other. The wide end should be just smaller than the beam width, the narrow end should come to a point (i.e. zero width). The amount of light coming through the slit should now be a linear function of where along the slit the beam hits. Just measure the total amount of laser light striking a photorecepter behind the slit. If you want to get fancy, you can modulate the beam as it leaves the laser and put the signal output of the photoreceptor through a bandpass filter centered on this frequency (this will mask out incident light); now the voice signal will appear as AM modulation on top of the carrier you originally used to modulate the beam with. One problem might be that I've assumed a lot of linear steps; air pressure in the room to window flexing; window flexing to beam deflection, etc. If you really want to get fancy, calculate the real transfer function for voice -> room air pressure -> window flexing -> beam deflection. Now, instead of making the slit a linear taper, cut the slit to be the inverse of that transfer function and you compensate for all those nonlinearities in one step! All the various random signals like wind and doors opening and closing just get lost as out-of-band signals when you demodulate. -- Roy Smith, {allegra,cmcl2,philabs}!phri!roy System Administrator, Public Health Research Institute 455 First Avenue, New York, NY 10016 "you can't spell deoxyribonucleic without unix!"