3 ms·

The point of spread-spectrum signal is that it's indistinguishable from noise unless you know the key. With frequency-hopping you still have spikes in the spec

by pps43 8y ago

The point of spread-spectrum signal is that it's indistinguishable from noise unless you know the key. With frequency-hopping you still have spikes in the spectrum around the frequencies you hop between, but with most modern systems you no longer have anything distinctive to find.

garmaine 8y ago
It's not indistinguishable from noise, if it is powerful enough to be received at all. If you filter based on the background noise threshold, you will discover a signal that quickly jumps frequencies in a pattern that is unpredictable without knowing the secret key. If frequency hopping is all that is used, you could even recombine the pieces to put the original message back together again. But in any case it will be very plainly obvious that the signal exists. A radio with a hardware limitation of only tuning to a specific frequency at a time would be unable to pick up on it, and would only hear transient blips as the signal enters and leaves the channel, which is what people mean when they say spread spectrum "looks like noise". You could defeat this analysis by actually broadcasting noise on all frequencies, so the transmission frequency is indistinguishable, assuming the encryption used is also indistinguishable from a random oracle. But now you have a noisy point source in the sky, even if you can't decipher the contents of the message. Succinctly put, obfuscation techniques like frequency hopping spread spectrum are tactical techniques that try to hide transmissions from actual radios in widespread use, not hide the existence of the radio signal in some physical sense. FHSS radio transmissions look like noise sources, not silence. Which is basically what SETI is looking for -- sources of radio noise unexplained by our understanding of astronomy.
- pps43 8y ago
  It is possible for the signal to be below the background noise. For example, GPS works this way. Sonars on submarines do the same thing, they send out a noise-like signal that can be detected only if you know what to correlate it with. There's always some noise to hide the signal in.
  - garmaine 8y ago
    This isn’t true. If the signal was below the noise threshold, and no mechanism provided for detecting noise separately (as twisted pair provides, for example) it would be literally impossible to separate signal from noise. That’s what the noise threshold means. GPS is not below the level of background noise. Not is sonar.

by pps43 8y ago

garmaine 8y ago

It's not indistinguishable from noise, if it is powerful enough to be received at all. If you filter based on the background noise threshold, you will discover a signal that quickly jumps frequencies in a pattern that is unpredictable without knowing the secret key. If frequency hopping is all that is used, you could even recombine the pieces to put the original message back together again. But in any case it will be very plainly obvious that the signal exists. A radio with a hardware limitation of only tuning to a specific frequency at a time would be unable to pick up on it, and would only hear transient blips as the signal enters and leaves the channel, which is what people mean when they say spread spectrum "looks like noise". You could defeat this analysis by actually broadcasting noise on all frequencies, so the transmission frequency is indistinguishable, assuming the encryption used is also indistinguishable from a random oracle. But now you have a noisy point source in the sky, even if you can't decipher the contents of the message. Succinctly put, obfuscation techniques like frequency hopping spread spectrum are tactical techniques that try to hide transmissions from actual radios in widespread use, not hide the existence of the radio signal in some physical sense. FHSS radio transmissions look like noise sources, not silence. Which is basically what SETI is looking for -- sources of radio noise unexplained by our understanding of astronomy.

pps43 8y ago
It is possible for the signal to be below the background noise. For example, GPS works this way. Sonars on submarines do the same thing, they send out a noise-like signal that can be detected only if you know what to correlate it with. There's always some noise to hide the signal in.
- garmaine 8y ago
  This isn’t true. If the signal was below the noise threshold, and no mechanism provided for detecting noise separately (as twisted pair provides, for example) it would be literally impossible to separate signal from noise. That’s what the noise threshold means. GPS is not below the level of background noise. Not is sonar.