@Posman, @Firqby and everyone else that contributed to this topic, thanks for the engaging and in-depth conversations. Seeing how passionate you are about our product is extremely rewarding to the team here at goTenna!
On to the issue at hand, I think we have a convincing answers to present. We analyzed the FCC report of the Motorola i335 phone, which can be found at this link: https://apps.fcc.gov/oetcf/eas/reports/ViewExhibitReport.cfm?mode=Exhibits&RequestTimeout=500&calledFromFrame=Y&application_id=exRI7AR26L4TaEEdbdjA9w%3D%3D&fcc_id=IHDT56HG2
Although the Motorola i335 nominal output power for the ISM transmitter (902-928 MHz, same as goTenna MESH) is stated at 890 mW, the effective peak power it transmits is actually 1 W as stated in Exhibit 6c “the modulation scheme employed can cause peak fluctuations in output power up to 0.5 dB from maximum pulse average power, which is 890 mW (29.5 dBm)” So the transmit power of the Motorola i335 is the same as goTenna MESH, at 1W.
MSK, FSK, were mentioned as possible alternatives to the GFSK. Back during the architecture phase, we have weighted the pros and cons of these modulations schemes and have concluded GFSK is the best choice. Here is why, MSK (Minimum Shift Keying) effectively reduces the modulation index to 0.5, while improving spectral efficiency, the sensitivity will be degraded by 1.5 to 2.5 dB due to the compromised modulation index (rule of thumb ideal index is 1). FSK, although improves sensitivity, it creates high levels of spurious contents and side lobes, which causes regulation standard violations. The best of both worlds is the GFSK (Gaussian Frequency Shift Keying), it applies a Gaussian filter to the symbols before the frequency modulated signal to suppress the spurs and side lobes, at an expense of 0.5 dB reduced sensitivity.
So by the above data, goTenna MESH and Motorola i335 should have the same range performance, but it does not appear so, and we have a hypothesis - it has to do with the network protocol and the choice between optimizing a point to point system or a scalable mesh network.
Digging into the FCC report of the i335, we found that during transmission, the i335 floods the entire 902 to 928 MHz during a transmission by cycling and repeating through all 50 channels. i.e. Tx channel 1, then 2, then 3, to 50 and repeat. This is a simple but brute force approach to gain frequency diversity and continuous re-transmissions, which effectively improves range at expense of congesting the network. For a point to point, none shareable system, this is the right approach. However, for a scalable mesh network, this approach will cause significant issues - as the number of nodes scale, the overheads and complexity of the routing table scales exponentially, without allowing proper amount of time slots / quite time of nodes in the system, the defining meshing feature of goTenna will fail. So this is a compromise we are willing to make.
Of course, all of the above are concluded from paper studies. We can acquire a few of the Motorola radios and conduct measurements on them in our lab to confirm our hypothesis.
I hope this clears things up a little. Please let us know if you have any questions or found any mistakes in our analysis, and thanks again for being so passionate about our product!
JC and goTenna team