Reflective thinking - signal processing

[OptimisticCynic]

Reflection of signals within the circuitry is generally considered to be a bad thing in radio electronics and computer hardware. I suspect that there are some very interesting applications to be discovered that involve intentional reflections of an input signal. Full and partial reflections, linear, 2D, & 3D reflections both open and within a network all deserve consideration.

 

For a simple start, let's determine the frequency of a sine wave signal by using a linear array of amplitude detectors. The signal enters at one end of the array and is reflected back through the array from the other end. At some point along the array, the reflected signal will interfere with the incoming signal, for frequencies within the range of the apparatus. Which amplitude detectors along the array show the highest and lowest spikes tells us what the frequency of the incoming signal is. Possible variations abound. The reason I give this example is because it seems to have some correlation to the circuitry within biological brains.

 

Jumping ahead and sideways several steps, I ask for input from people familiar with the neural type microchips presently being studied. How much effect does additive and subtractive interference from signals reflecting in different directions at branches in the circuitry have on the way inputs are processed within the neural circuits?

 

Backing up to interferometry; what kind of patterns do multiple digital input signals form as they reflect within a network?

 

Please help me find examples of the use of reflection of signals within the detection apparatus. Plus any thoughts you might have on reflective interferometric signal processing?

 

Thank you for exploring with me.

[waitforufo]

Reflection of signals within the circuitry is generally considered to be a bad thing in radio electronics and computer hardware. I suspect that there are some very interesting applications to be discovered that involve intentional reflections of an input signal. Full and partial reflections, linear, 2D, & 3D reflections both open and within a network all deserve consideration.

 

For a simple start, let's determine the frequency of a sine wave signal by using a linear array of amplitude detectors. The signal enters at one end of the array and is reflected back through the array from the other end. At some point along the array, the reflected signal will interfere with the incoming signal, for frequencies within the range of the apparatus. Which amplitude detectors along the array show the highest and lowest spikes tells us what the frequency of the incoming signal is.

Reflection of signals within circuitry is used all the time in radio electronics. Such circuits are called filters. No radio would work without them. Reflections are also used in common antennas. A parabolic dish reflects transmitted signals to a focused point at the receiver module. A Yagi antenna only has one active element. The rest are parasitic and spaced such that signals from the active element reflect off the parasitic elements and combine to increase gain or antenna aperture in a particular direction.

[OptimisticCynic]

The filter circuits, that I know of, in electronics are feedback loops. Feedback loops give results very different from reflection.

 

The Yagi-Uda antenna is an excellent example. Thank you, waitforufo. I notice that the Yagi itself is a filter, since it has a narrow frequency range.

 

The directional effect of reflection from elements of slightly varying lengths could lead to single location spikes of interference within a network. Has anyone seen this happen, perhaps within a wired computer network, where one node of the network has an otherwise unexplained interfering signal? This could change drastically with small changes in the wiring.

 

I was thinking of 180 degree phase changes along a linear array producing 90 degree changes in the path of information through a network. Now I see that even phase changes of small degree could lead to an input being modified and routed differently within a network. This makes the variety of possible results more complicated while simplifying the actual physical construction of the network. Natural selection loves taking advantage of that kind of thing.

Edited August 22, 2015 by OptimisticCynic

[RuthlessOptimism]

I don't know if this adds anything helpful to this discussion but awhile back a representative of a company looking to hire co-op students gave a talk at my university about their company and some of the work they do. One interesting thing he talked about is their "Enigma" team wherein they study "Emergent Behavior" in electronics. The example he gave without giving any actual details (I guess they are in the process of patenting this device) is an electronic circuit consisting of only resistors that between input and output will change the frequency of a signal. He said this is very helpful for the design of their chips because it simplifies the design and saves them money. This company creates silicon chips for audio processing.

 

To me it just sounds like there is stray capacitance in the circuit, but since this company produces some of the most widely used chips for audio processing I assume they know what they are talking about. The only other thing I could think of to explain this is some kind of internal reflection in the circuitry but I don't know how you'd use that to get the result they claim.