One of the main advantages to reclassifying your operations to Part 97 is interference protection from unlicensed operations (see Sec. 15.5) Licensed services have priority over unlicensed operations.
Reclassifying under Part 97 also provides a way around the Part 15 certification/authorization problem. (see Sec. 15.204) Which states in short that an authorized system includes its marketed antenna and other components and must always be used in its original configuration in which it was authorized, and should not be modifed. Amateur equipment does not require certification / authorization, and hams may use equipment that has been certified for another service on amateur frequencies. [Sec. 97.315] Under Part 97 you can legally modify these devices, using (homebrew) pre-amps, RF amplifiers, and high gain antennas.
Another fairly big advantage is the amount of achievable radiated power under Part 97.
As you may know Part 15 operation has effective radiated power (ERP) limits. It also has different (lower) limits for omni-directional antennas.
Amateur Radio has never had or has any type of ERP limits. The wording of Part 97.311(d) which regulates Amateur Spread Spectrum uses the words "transmitter power" which imply peak envelope power (PEP) or carrier power (CP) not effective radiated power (ERP). (If it said radiated power then ERP would be implied). Non-Spread Spectrum modes such as 802.11a or 802.11g have the 1500 Watt PEP limit. Note there are no ERP / EIRP limits for Part 97 operation. There is also no difference for omni-directional or directional setups. The directional antennas sited in the table below are for example purposes. And the EIRP's listed below are only limited by the gain of your antenna.
So some example calculations show that the following are achievable and permissible under Part 97 if necessary (97.313):
Achievable under Part 97: | Max. PEP RF power | Ant. gain | EIRP* |
900 MHz (spread spectrum) | 10 watts | 14 dBd yagi | 411.9 watts |
2.4 GHz (spread spectrum i.e. 802.11 or 802.11b) | 10 watts | 24 dBi partial parabolic | 2511.89 watts |
2.4 GHz (non spread spectrum i.e. 802.11g) | 1500 watts (per 97.313) | 24 dBi partial parabolic | 376.8 Kilo-watts |
5.7 GHz (non spread spectrum i.e. 802.11a) | 1500 watts (per 97.313) | 28 dBi dish | 946.4 Kilo-watts |
Reclassifying under Part 97, imposes the standard amateur restrictions.
Obviously the communications you make under Part 97 must be to and from
licensed amateurs. These communications cannot be made with a pecuniary
interest and may not be obscene or indecent.
Other misc. Part 97 clarifications
Permissible under Part 15: | Max. Transmitter RF power | Ant. gain (dBi) | EIRP (W) |
900 MHz | 30 dBm (1 W) | 6 | 3.98 |
2.4 GHz omni-directional | 30 dBm (1 W) | 6 | 3.98 |
2.4 GHz directional | 29 dBm (800 mW) | 9 | 6.35 |
28 dBm (640 mW) | 12 | 10.14 | |
27 dBM (500 mW) | 15 | 15.81 | |
26 dBm (400 mW) | 18 | 25.23 | |
25 dBm (320 mW) | 21 | 40.28 | |
24 dBm (250 mW) | 24 | 62.79 | |
23 dBm (200 mW) | 27 | 100.2 | |
22 dBm (160 mW) | 30 | 160.0 | |
5.15-5.25 GHz | 16 dBm (40 mW) | 0 | 0.16 |
5.25-5.35 GHz | 23 dBm (200 mW) | 6 | 0.80 |
5.725-5.825 GHz omni-directional | 30 dBm (1 W) | 6 | 3.98 |
5.725-5.825 GHz directional+ | 30 dBm (1 W) | 28 | 630.9** |
*Estimated maximum achievable EIRP, FCC does not limit
antenna gain in this situation.
**Operating as a Part 15.247 (ISM) device this is an estimated
maximum achievable EIRP, you are allowed unlimited antenna gain for point
to point. Under Part 15.407 (U-NII) it is 32dBi EIRP. Check your devices
classification.
Why high power?
The higher power will overcome smaller obstructions, like trees and leaves, but it still won't cut through that hill in the middle of your link path. Microwave is line of site no mater what. If you are not familiar with microwave propagation I suggest reading; VHF/UHF/Microwave Radio Propagation: A Primer for Digital Experimenters. Do keep in mind that the above is to point out the maximum achievable and permissible. As with anything in ham radio you should use the minimum transmitter power necessary to carry out the desired communication. Obviously there is a lot that plays into this. Such as available power level settings on your equipment and fade margins. In shared bandspace those margins probably need to be higher to overcome potential inference.
It boils down to good amateur practice and engineering.Here is an interesting note about FHSS:
Technically, the fact that frequency hopping systems don't spread their signal results in no processing gain. Processing gain comes from the increase in power density when the received signal is despread. This helps to improve the received signal's signal-to-noise ratio. In other words, frequency hopping systems will need to put out more power in order to have the same signal-to-noise radio as a direct sequence system.
Note update:
Are 802.11a/g or other OFDM modes classified as Spread
Spectrum?- No. A few years ago a company called WiLAN submitted an OFDM
802.11g device to the FCC for certification under 15.247, arguing that, since it
could pass the "CW jammer test" (intended to assure that devices
exhibited the then-required 10 dB of process gain), it met the requirements of
15.247.
The FCC labs rejected the request for certification, stating that OFDM did not
meet the definition of DSSS and therefore the device did not comply with the
rules. See: http://hraunfoss.fcc.gov/edocs_public/attachmatch/FCC-01-158A1.pdf