[Home/Nieuws]  [Magazines]  [Meetings]  [Downloads]  [Redaktie]  [Geschiedenis]

LIDAR - A new threat
by BillSF

There has been much talk in recent years of 'police lasers' to catch speeders. A Gallium Arsenide (GaAs) or similar matterial that emits between 820 and 904nm can produce very fast pulses of (infrared) light very ecconomically. Must faster pulses are possible, but in one nano-second, light gets only 30cm
which is plenty accurate to catch a speeder. Basicly the devices are simple: Send out a series of pulses and time how long each one takes to return. This is what most of us think of when we think of real radar.

The lasers are quite directional, but are far more dispersed than the common pointer toy. This is needed so atleast some part of the automobile reflects strongly enough. It may be the shiny finish, some chrome, or even the windscreen. It would be mighty difficult to build a 'stealth car', if you could afford it! Readings may be possible at greater than 100 meters, but shorter pulses and shorter ranges are more likely. Even if the pulses are not really short, modern analogue techniques allow accurate averaging of sloppy pulses.

As a real example, fire a pulse at 100M. It will take 667nS to make the 100 meter round trip. Keep firing the pulses. (Rumour is 10k/S is typical) Time the returns, with easy nano-second pulses, the accuracy is 5cm! An auto at 100km/h is only going to move 2.78mm in 100uS, but keep a running average and you quickly determine the speed after the auto has moved less than five meters or so.

In the next part we will look at ways to beat this new method. You can pay between zero and perhaps a thousand Guilders. An absolutely detectable lamer device to a first class spoof are covered. It is hoped there will be enough interest in highway ECM, the laws covering it and possibly special projects that may lead to low cost marketable devices. This document will help assure there are no patents filed.

Countering the LIDAR threat.

You, the mad hacker and speed freak, have alot on your side to beat this method at reasonable cost. This is the general ECM (electronic counter measures) approach:

• Get a signature of the LIDAR.
  - What is the pulse rate?
  - What is the pulse width?
• Determine your distance.
  - The aparent intensity is the square of the distance
  - Know your exact speed 
  - Do the math and use the 'running average' (You know how far you've gone and how much brighter the light has become)
• Fire back 
  - Time the return pulse. (It will be the brightest 'object', and will need to have its delay adjusted as you approach.)
  - Note: You will still be 'some greater distance' from the LIDAR and then you vanish. Too late!

While the above is the 'proper way' and how state-of-the-art radar jammers work, there is a more traditional way, also developed for military use: Confusion.

• Best to get a signature, atleast detect it!
• Fire back, 'randomly' this time at each pulse and/or just make noise that worked fine with enemy radars throughout the Cold War.

If you are a radar person, the above should make alot of sence. This section will discuss some technology.

Measuring the laser

You will need a fast photodiode and what is called a transimpedance amplifier which simply converts the feable current of the photodiode to a usable video level. (video here in *not* TV duh!) Low cost fibre technology has made these high speed units readily available. (use the 1Gbit/s receivers) You will need
a daylight filter (unexposed Kodachrome or movie film is great) and some simple optics to give a reasonable field of view and *greatly* increase the sensitivity.

Any well designed optical receiver will have some sort of AGC loop and a means to tap the analogue signal and take a peak reading. In the 'quick and dirty' design, you probably don't care how strong the pulse is and you may not even care how wide they are!

For pulses less than 10nS, analogue techniques are definitely preferred. The gated ramps can be converted easily to the digital domain. The peak power of the pulse is an analogue measurement, which is quickly converted to the digital domain. Repitition rates much greater than 10k are probably not used,
so even a modest PIC controller (with built-in A/D) will do nicely for the front end. While it may be possible to do it all with a PIC, something more conventional (ARM610?) that will run C, would seem logical to do all calculations. Analogue techniques are best used to generate a pulse width and create the delay. Ofcourse the main processor knows how long to delay, how wide the pulse is to be and perhaps even control its power.

Confusion: An old military tacktic

Allright, lets do it easy and then *real easy*. Lasers from scrapped CD players or similar low-power units will be used. Optics (perhaps the CD player system will do?) will greatly spread the light. A bare laser-diode will emit about a 20x40 degree radiation pattern which may be just fine. This laser is *very*
low power and all you *must not do* is put it up to your eye. It may look  very faint red, but don't be fooled! (Please note that at any distance at all it is 100% safe as it is *not* a beam. The policeman's claim they he may be harmed is bogus on three counts. 1) The beam is widely dispersed after it is
past a few centimeters. 2) It is pulsed and the average power is again *very* low. 3) If Joe Cop fears for his eyesight, what must he be thinking when aiming at all those drivers?

Method one.

Use a photodiode, set a random delay (say between 10nS and 1uS) and fire back. You probably wish to have an indicator that tells you got hit. You may also wish to vary the pulse width of the laser, say between 1nS and 20nS or so. A simple noise generator and some other analogue circuitry is probably all that is needed.

Method two. (Real stupid, but it probably works - for now)

Just generate 'random' pulses between 10 and 1000nS and vary their width 'randomly'. Allways on and easy to detect. If into electronics, you can probably make this for 'free'. The box and board probably cost the most!

Conclusion:

This is by no means an exhaustive analysis of speed enforcement techniques. In addition to radar and lidar, there are several more methods. We all know the 'speedometer and car' method, making the CHP tops in their speed control worldwide. Speed can be determined from the air by timing cars passing marks. It can be measured by placing pairs of 'tripwires/hoses' on the road. It can be done 100% visually using a CHP developed technique that uses a little trig.

Very worrying are the cameras on the gantries. In addition to getting your velosity, it can be easily abused by governments. Since fonts on number plates vary little, OCR techniques are very simple. Beating this facist system may be easier and more 'stealthy' than most think. If few people care to 
beat 'highway monitoring', many techniques are going to remain legally hackable for years. But watch those 'bears in the air' and sharp eyed police! As an experienced driver at about 1Million km, this subject interests me. I also lost my license due to speeding tickets! (Later i get it back, move to Holland and find it isn't a very good idea to have an auto in Amsterdam.)

De informatie in 't Klaphek dient slechts een educatief doel. Gebruik van deze informatie zou strafbaar kunnen zijn. De redaktie wijst iedere verantwoordelijkheid voor gebruik door lezers van de in 't Klaphek opgenomen informatie af. De mening van een auteur weerspiegelt niet noodzakelijkerwijs de mening van de redaktie of uitgever.