On Thu, Nov 7, 2013 at 5:38 PM, Andy Isaacson wrote:

... Or, you know, just look on github. https://github.com/piannucci/blurt

please tell my week ago self about this project ;) how does one constrain the frequency bands used? blurt_cpp_80211/blurt.cc is a bit opaque, and i'd love to try the max rate benchmark with just a 18-20kHz channel...

On Thu, Nov 07, 2013 at 06:56:44PM -0800, Taral wrote:

On Thu, Nov 7, 2013 at 5:53 PM, coderman wrote:

...

how does one constrain the frequency bands used? blurt_cpp_80211/blurt.cc is a bit opaque, and i'd love to try the max rate benchmark with just a 18-20kHz channel...

Based on a quick glance, it's using OFDM -- you should be able to constrain it to the appropriate sidebands. Although for only 2kHz I'm not sure how much you get out of OFDM in the first place.

FWIW OFDM is multicarrier at low symbol rates per carrier (compared to total information rate in the signal, not just "low" by some arbitrary standard). The lower the symbol rate, the longer the coherence distance, which gets to be important to reduce ISI (inter symbol interference) in a multipath environment. And indeed sending modulated tones over a speaker to a remote microphone is likely wickedly full of multipath including multiple paths due to reflections off hard surfaces that result in some serious delay spread even compared to the relatively low speed data you can stuff into 2 KHz from 18-20 KHz. Sound does not propagate all that fast compared to useful data rates after all. So OFDM is good. -- Dave Emery N1PRE/AE, die@dieconsulting.com DIE Consulting, Weston, Mass 02493 "An empty zombie mind with a forlorn barely readable weatherbeaten 'For Rent' sign still vainly flapping outside on the weed encrusted pole - in celebration of what could have been, but wasn't and is not to be now either."

On Thu, Nov 07, 2013 at 11:32:13PM -0800, coderman wrote:

On Thu, Nov 7, 2013 at 8:44 PM, David I. Emery wrote:

...

... sending modulated tones over a speaker to a remote microphone is likely wickedly full of multipath ... Sound does not propagate all that fast compared to useful data rates after all.

So OFDM is good.

you can even run OFDM in 900Mhz with the Ubiquiti SR9 radios, oldies but goodies

combine MIMO with OFDM for even more multipath resistance and throughput. (MIMO audio channels ala dolby applied is amusing ;)

SDR MIMO kit leaves much to be desired, however...

Ettus has some new lower-cost SDR boards that are getting approving glances from several of my radio-savvy friends: https://www.ettus.com/product/category/USRP-Bus-Series And the new kid on the block is Nuand BladeRF, with a half-the-cost design that seems pretty promising: https://nuand.com/ Both of these theoretically support MIMO, with clock distribution available as an added-cost option to sync up multiple boards' Tx/Rx. -andy

On Thu, Nov 7, 2013 at 5:53 PM, coderman wrote:

... how does one constrain the frequency bands used? blurt_cpp_80211/blurt.cc is a bit opaque, and i'd love to try the max rate benchmark with just a 18-20kHz channel...

more info for week ago self: the edu.mit.csail.wami.audio EncodePipe and adjust bitrate, bandpass (extremely high attenuation), etc at this point for emission only in the inaudible range. DecodePipe also needs same filtering, with amplification, etc at this point for the analog to digital conversion. (and you can divert a copy to WaveContainer for viewing as spectrogram[0][1]) 0. "Audacity - Spectrograms Preferences" http://manual.audacityteam.org/man/Spectrograms_Preferences 1. "Exploring Audacity's Spectrogram View" http://www.youtube.com/watch?v=7WYw3qoTdU4