First, many thanks for taking time to read my previous post (re sense of 
direction--whole new idea). I am very pleased that a few people found 
the concept useful. Below is an extension to the idea. Dave Malham 
alluded to this idea (sort-of read my mind), and he also mentioned a 
problem that deserves careful consideration. Here's Dave's post:

**Great idea, the only problem being the presence of un-synced digital 
electronics near ultra sensitive analogue electronics in the microphone 
which would necessitate some very, very careful design to avoid 
interference. Might be better to develop a smartphone app that took a 
photo of the mic, geo-tagged it and associated that with the recorded 
file, perhaps via a Bluetooth link to the recorder (easy if it's a 
computer based one, not so easy with anything else, Apart from the 
environmental recordings case, the situation where it 
would really come in handy (at least, with an attached unit rather than a 
photo) is for a mic suspended on a single cable, which won't 
necessarily stay oriented the same way all the time due to subtle 
"untwisting" of the cable over time. I can vouch for that from 
experience of suspending mics from the central tower of York Minster. 
Dave**

First, I'll address noise and unsynced digital signals. The device I had in 
mind is a passive sensor with associated electronics (mixed signal). 
Using mu-metal or similar shielding would preclude the sensor's ability 
to detect magnetic poles. The electronics, however, would be shielded. 
Most of the requisite electronics are tiny and battery-operated.
The output signal is analog (and, of course, balanced line), so no 
interference issues here. Briefly, a single PIC chip could do the 
direction-to-frequency conversion. Diodes can easily linearize a 
pulse-width modulated signal, thus creating a sine wave with less than 3 
percent THD. A center frequency of say 3 kHz would be 0 degrees. 
Compass (sensor) rotation in one direction would yield frequencies less 
than 3 kHz, while rotation in opposite direction yields frequencies 
higher than 3 kHz. The problem is 180 degrees--this could end up lower 
or higher than center frequency. The important idea here is that the 
output could be recorded on a conventional recorder (such as the TASCAM 
D680) without need for piggy-backing a GPS unit or other DAQ hardware 
needed to record time/place/orientation. However, having GPS on the 
recorder itself would be an asset--and would easily be isolated from microphone 
circuitry.

Next idea (re suspended or moving mics): Just after posting my idea, it 
occurred to me that the analog frequency (translated to give absolute 
direction re Earth's magnetic poles) could also be used for error 
correction. I believe this may be what Dave had in mind (my apology if 
I'm wrong here). As a mic rotates on its vertical axis, the sensor 
output will inevitably change. On the recorder, this is a mere change of 
frequency for a continuous tone with fixed amplitude. Decoded, this 
change of direction can be used to compute the mic's orientation. A more 
sophisticated approach would entail FM: The frequency deviation 
provides a voltage (or equivalent bits) that is proportional to 
direction. The rate of change is directly related to mic's rotation 
frequency in rad/s (probably very slow unless you're running with the 
mic).

Taking this demodulated frequency a step further would require linking it to 
Ambisonic software (A- or B-format). An 'analog' approach I had 
considered is to apply the direction-dependent voltage to a MIDI 
controller I had previously built using an Arduino board. It's a simple 
volume controller with an analog input (voltage), but the MIDI channel 
can be assigned to a panpot or other function. The panpot would be mic 
direction (horizontal plane). A more direct approach would be to use a 
mathematical 
expression with, for example, VVMic to 'steer' the mic to its proper 
orientation. I'm sure others will have better ideas--I'm simply coming 
up with this stuff as I write.

Re GPS: Along similar lines, I had posted an idea to a forum (National 
Hearing Conservation Association on LinkedIn): The idea relates to noise 
dosimetry. My idea was to use a GPS tracker in lieu of the dosimeter 
worn by workers to map their locations and time spent at these 
locations. Fixed recordings for a variety of critical locations within a 
workplace could be done without need for strapping a sensitive mic / 
electronics to the worker. In other words, recordings and SPLs 
are made independently of workers being at a specific place. This could 
result in high-quality recordings for post-processing as well as 
multi-channel (or Ambisonic) recordings. When combined with 'typical' 
time spend in locations (a la GPS), noise dose, or Leq, could be 
ascertained. It wouldn't eliminate need for dosimetry, but could be 
useful in some situations.

Thanks to everyone for your help and time.
Ciao,
Eric C.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 
<https://mail.music.vt.edu/mailman/private/sursound/attachments/20131007/1326a0d6/attachment.html>
_______________________________________________
Sursound mailing list
[email protected]
https://mail.music.vt.edu/mailman/listinfo/sursound

Reply via email to