From OpenCircuits
Jump to navigation Jump to search

What should I say here about frequency hopping spread spectrum (FHSS)?

... say a few words about adaptive frequency-hopping spread spectrum (AFH)...

Is it true that adaptive FHSS can give higher bandwidth and use less power (longer battery life) than any fixed-frequency broadcast system?

... mention systems that combine FHSS with mesh networking ...

  • slow frequency hopping: after hopping to the new frequency, the transmitter sends several symbols before hopping to the next frequency. This gives the fastest data rate when running on hardware that takes a long time to switch to a different frequency.
  • fast frequency hopping: The transmitter hops to several different frequencies, sending the same symbol on each one, before transmitting the next data symbol. This is more tolerant of single-frequency interference.

Acquisition and tracking

When a receiver is turned on, what frequency should it listen to first, and how long should it listen before trying some other channel?

There seem to be 4 possibilities:

  • Both systems start at the same "acquisition channel", and stay on that channel until they hear from each other, then start hopping. This gives the quickest time-to-lock, if there is no interference. But if narrowband interference happens to be on that channel, then they never lock.
  • The receiver picks a random acquisition channel (a random point on its hop sequence) -- perhaps the quietest channel with the least noise on it -- and stays on that channel until it hears the transmitter. The transmitter always goes through the hop sequence in the standard order, sending a packet on each channel. Eventually the transmitter gets to the channel the receiver is listening on, and the receiver hears that packet, synchronizes its time, and starts hopping.
  • The transmitter picks a random acquisition channel (a random point on its hop sequence) -- perhaps the quietest channel with the least noise on it -- and transmits its time signal over and over on that channel. The receiver goes through the hop sequence in the standard order. Eventually the receiver gets to the channel the transmitter is sending on, and the receiver hears the packet and synchronizes its time. After the transmitter has given the receiver plenty of time to sample every possible channel, both should now be synchronized, and start hopping.
  • Both the transmitter and the receiver step through the hop sequence in the standard order (starting at any arbitrary or random channel). However, the receiver stays on each channel longer (2 times as long?) as normal, taking longer (2wice as long?) to step through the whole sequence. Eventually, the transmitter "catches up" to the receiver in the hop sequence, the receiver hears that packet, synchronizes its time with the transmitter, and starts hopping at the normal hop rate.
  • ...

Legal considerations

In addition to the physical limitations mentioned above, there are also legal considerations, such as, for example:

(h) The incorporation of intelligence within a frequency hopping spread spectrum system that permits the system to recognize other users within the spectrum band so that it individually and independently chooses and adapts its hopsets to avoid hopping on occupied channels is permitted. The coordination of frequency hopping systems in any other manner for the express purpose of avoiding the simultaneous occupancy of individual hopping frequencies by multiple transmitters is not permitted. --47CFR15.247 : FCC Part 15 Section 15.247

Paraphrased in SWRA041 as

It is not permitted to synchronize transmitters of different frequency hopping systems to each other. This could lead to a super system that occupies individual frequencies for more than the allowed 0.4 seconds in the 10s or 20s period. Notwithstanding this, any system may recognize others in the frequency band and adapt its hop set independently to avoid hopping into an occupied channel.

Some of the "ETSI EN 300 440" regulation is paraphrased in swra060 as

7.5 Additional FHSS requirements FHSS modulation shall make use of at least 20 channels separated by the channel bandwidth ... The dwell time per channel shall not exceed 0.4s. ... During operation every channel shall be used at least once within a period of 4 times the dwell time per hop multiplied with the number of channels. While the transmitter is performing hops between channels, the transmit power level shall be attenuated ...

Is this a hard limit on adaptive frequency hopping? Or does it merely mean that a system that skips over some frequency for longer than 4*N*D seconds no longer qualifies under the "frequency hopping" section, but must fit under the (more restrictive) "other digital modulation" test limits?

... What other legal considerations and FCC policies are relevant to designing intelligent radio communication systems? ...

Direct sequence spread spectrum

(FIXME: make new article for other kinds of spread spectrum? Do we need a general spread spectrum article?)

Spread Spectrum Link has complete schematics, photos of hardware, and design tips for a DSSS communication link.

With any kind of spread spectrum, "The most difficult area is the receiver path, ... because the receiver must be able to recognize the message and synchronize with it in real time." -- Maxim application note 1890: "An Introduction to Direct-Sequence Spread-Spectrum Communications"

Further reading