I made a new kind of lock in amplifier

Is that not just FHSS, DSSS, CDMA or the like? Or do you mean you've patented it in a "novel" application (medical imaging)?

Tim

-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website:

Seems like it to me. He makes a big distinction between encoding "bipolar" signals vs. "unipolar" signals. In the way he describes it, aren't all di gital transmissions "unipolar", i.e. zeros and ones?

I considered using something like this in photoacoustic spectroscopy some 4

0 years ago. But at the time I don't know the technology would have allowe d it and I didn't have the background in electronics to invent the needed e quipment. We were using a xenon light source which was then swept by a mon ochromator and chopped by a rotating disk, yes, a rotating disk. The resul ting vibrations were picked up by the microphone and interfered with the si gnal. So the source had to be decoupled from the sample. I wanted to use a disk with many small holes with varied spacing and width so the signal co uld be decoupled at least from the harmonics. But we were getting good eno ugh results by using two lab benches. If we had been able to modulate the light source at will this would have worked much better and been able to in clude in a single instrument.

This was actually my first class in any electronics in college. I was in t he chemistry department at U of Md and the professor was big on working the technology. He gave me a small 8 bit computer one of his grad students co uldn't get to work. It was based on the 8008 CPU, yes, the 8008! I got it to work when I found the clock circuit which used a pair of monostables wa s crap, but would work with more decoupling.

Rick C.

Hi Tim,

It is a new kind of DSSS, and the patent covers the general signal processi ng method. I only used optical medical imaging as an example. The first cla im of the patent starts out:

1. A method of transmission from at least one transmitter to at least one r eceiver and subsequent recovery of a transmitted signal, the method compris ing the steps of:

...

CDMA (code division multiple access) uses DSSS (direct sequence spread spec trum). It's probably clearer in instrumentation to call the method code div ision multiplexing. Lots of people have used DSSS for measurement systems. The high processing gain of the method is very attractive, and gives basica lly the same advantages as a lock in amplifier, but is also pretty immune t o narrow band interference. Mine is different in being useful for uni-polar signals or in systems where you can't get the phase information. It's got to do with the code signal having all 1's and 0's, while normal DSSS uses 1 's and -1's. The -1's are needed to get the different codes to cancel each other so that there is no inter-channel interference.

I manufactured the optical scanners using the method for a few years. It wo rks.

ChesterW

DSSS signals use 1 and -1. The -1 represents a phase shift.

ChesterW

You went to a lot of trouble to obscure the gmail address you used to post three follow-ups.

On Apr 14, 2018, ChesterW wrote (in article ):

I?m sure it works, but I?m also surprised that the patent was granted.

People have been doing these kinds of things in radar for decades, and the approach has been extended to many unrelated fields since.

For instance, for interplanetary radar:

Ref: Antennas and Radar for Environmental Scientists and Engineers, By David Hysell, Cambridge U Press, 2018.

People also use linear FM chirps a lot. Both signal types depend on correlation gain using a wideband signal.

For a general background, start with ?Radar Principles? by Clayton Z Peebles. Hmm. It?s now out of print, and getting expensive. This was a widely used textboopk. There may be pdfs available on the internet.

Joe Gwinn

On Apr 14, 2018, snipped-for-privacy@gmail.com wrote (in article):

The classic remedy is to subtract 0.5 from the unipolar signal and multiply by 2 to obtain +1 and -1. Phase, while useful, is not required. In radar, with and without phase data is called coherent and noncoherent processing respectively.

Joe Gwinn

You are correct about the holes changing the frequency spectrum of the ligh t signal. That's excellent intuition for a first year guy. The light signal would still have been harmonically related to the interference from the me chanics though, since the hole pattern repeats every revolution of the whee l, so there would have still been problems. Like you said, a non-mechanical way to flash the light would have fixed the coupling, but it's hard to fla sh that kind of light quickly, as I'm sure you already know.

Using my system, you could not only replace the chopper wheel, but also the monochromater. The light source can be made of multiple emitters, each emi tting a different part of the light spectrum. All of the emitters can flash simultaneously, each using a different code, and then the amplitude of eac h could be separated mathematically from the received signal. Since the mon ochrometer only allows one color through at a time, this would dramatically increase the throughput of the spectrometer. In fact, the first test I did to confirm the idea was using RGB LEDs in a reflection spectrometer. I mea sured paint-chip samples and compared the results to a commercial color rea der.

ChesterW

[snip] [snip] [snip]

I'm not surprised that a patent was issued. The USPTO has become polluted by Vietnamese bureaucratic toadies who have no clue. I've had a few recent exposures that demonstrate that they are village idiots.

What _will_ determine patent validity will be some hard-ball suits when people like Chester try to enforce their "patents" and get their ass handed to them on a shingle.

(I really enjoy being an expert witness in patent court ;-) ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 

           To those of us in my age bracket... 

           GREEN means inexperienced and/or incompetent.

I definitely don't know why it's possible to have a patent granted on a design for a system which intrinsically relies for its function on "prior art" that has itself not been invented yet. as an absurd example e.g "here's a new design for an interplanetary spacecraft which if a space-folding antimatter drive existed would employ it in a novel fashion."

But there are definitely patents in that vein that have been granted to large multinational corporations with names like Sony, Apple, etc.

ght signal. That's excellent intuition for a first year guy. The light sign al would still have been harmonically related to the interference from the mechanics though, since the hole pattern repeats every revolution of the wh eel, so there would have still been problems. Like you said, a non-mechanic al way to flash the light would have fixed the coupling, but it's hard to f lash that kind of light quickly, as I'm sure you already know.

The unequal spaced holes would produce a signal that is still related to th e fundamental rotation rate, but many of the vibrations would be harmonics and the holes could have been designed to be insensitive to any of the inte ger harmonics of the fundamental.

As to the light, I knew nothing about it at the time (or even now really, I barely remember what it was). It was actually a year later that I had the thoughts about irregular holes. My first year was taking senior level cou rses to qualify for getting into a graduate program in EE since my BS degre e was in chemistry. The next year I was a TA sharing a large office with a bunch of others. Lots of opportunity to bounce ideas around. I had a fri end who I think was working on the GPS system for his PhD. I recall he tal ked about figuring out how to get the data from a number of sats without kn owing where they were and figuring out exactly where they were. He was str uggling with it at the time. I don't think he called it GPS at the time, i t was still just an idea for research.

he monochromater. The light source can be made of multiple emitters, each e mitting a different part of the light spectrum. All of the emitters can fla sh simultaneously, each using a different code, and then the amplitude of e ach could be separated mathematically from the received signal. Since the m onochrometer only allows one color through at a time, this would dramatical ly increase the throughput of the spectrometer. In fact, the first test I d id to confirm the idea was using RGB LEDs in a reflection spectrometer. I m easured paint-chip samples and compared the results to a commercial color r eader.

I don't think "multiple emitters" would do the job. The monochromator swee ps the frequency across a range with a very narrow bandwidth. I don't see how this could be approximated with multiple fixed light sources.

But today there are LCD shutters which would produce some EMI perhaps, but not mechanical vibration.

Rick C.

On Apr 14, 2018, snipped-for-privacy@gmail.com wrote (in article):

This is a coded aperture system:. Note that in these applications, there is no carrier phase information. It?s strictly on or off.

It is not true that correlation requires plus and minus, to yield zero for uncorrelated signals. The threshold can be placed wherever it needs to be.

The granddaddy is the Dicke Radiometer:.

Dicke went on to invent the lock-in amplifier, and to found Stanford Research Systems (SRS) to commercialise the lock-in amplifier. SRS is still around, still making lock-in amps, now based on DSPs.

Instead of coded apertures, one could have used a number of sinewaves modulated optical carriers, where the sine wave frequencies are not harmonically related (and thus orthogonal), and used a bank of synchronous mixer downconverters to isolate the signals.

.

That may be the innovation, the speedup due to parallel operation at multiple wavelengths. But the claims of US 8,693,526 don?t say that. I suspect that Claim one would cover prior art such as coded-mask IR seekers in air-to-air antiaircraft missiles like AIM-9 Sidewinder, which have been around forever. This is a pretty good defense against Claim one.

.

Coded apertures seem to have come in when defenses against active IR countermeasures (intended to confuse missile seekers) came to the battlefield. They have the idea of using different mask codes for different IR bands.

.

Joe Gwinn

You?re right. Crap.

ChesterW

Thunderbird quit, so I used google.

On Apr 15, 2018, snipped-for-privacy@gmail.com wrote (in article):

There may be something patentable in there. Use of an old approach to a new or different problem, yielding advantages over current practice is patentable. If you go that route, start a whole new patent; let the old one sleep unmolested.

Joe Gwinn

It?s 3 dB down from the simpler standard method, and takes a bit lo nger to calculate. It doesn?t look hopeful. Thanks for pointing it out though. It saved me time.

ChesterW

On Apr 15, 2018, snipped-for-privacy@gmail.com wrote (in article):

What is 3db down from what? Context needed.

Joe Gwinn