1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Frequancy choice for echolocation & sonar

  1. Jan 18, 2013 #1
    Hi. My understanding of sonar is that the choice of frequency of the pulse, or of the frequncy sweep, is related to the size of what is being sought. That is, a frequency resonant with the range of likely objects, will reflect more strongly, and a non-resonant one be scattered more.

    Is this true? And does this govern frequency-choice for bats with echolocation?

    I looked at the relevant Wikipedia pages: http://en.wikipedia.org/wiki/Sonar and http://en.wikipedia.org/wiki/Animal_echolocation but couldn't find anything explicitly about frequency choice in this regard, or resonance.
  2. jcsd
  3. Jan 18, 2013 #2


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    The choice of frequency will be a complicated one. High frequencies are absorbed more in humid air, so the range can be compromised but the directivity of a short wavelength radiator can be better. Then the frequency response of the sensors will be relevant; there will be some sort of upper limit to this - even for small creatures with tiny cochlea (like bats).
  4. Jan 18, 2013 #3
    Is there a relationship between frequency and objects being sought, among the other factors?

    Maybe take an ideal condition of lossless medium etc. - how would it be chosen then?
  5. Jan 18, 2013 #4


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    The following is very arm waving but will get some of the basic numbers involved. I am not sure how informed you are about this but I am just using my general knowledge of Physics and Engineering and what I have found with Google.
    This link gives a lot if information about the absorption of various u/s frequencies - which is very relevant at distances above a few metres.
    Bats use up to and around 100kHz. At 100kHz, the wavelength is 3.4mm. This would be the sort of resolution that you could expect (twice the width of a single pulse). But, to avoid confusion, you need to leave a gap between pulses. To detect a target 17m away, you need to leave a gap of 0.1s between pulses (total transit distance = 34m) so you would need a very high peak power (100 times the mean power) compared with the mean power that the bat could produce. Chirping type signals improve this quite a bit because they can last longer and carry more energy than a single pulse. A bat, being intelligent, can increase the pulse rate as the target gets closer, of course.
    The amount of energy reflected back (target cross section) depends upon target shape and size and also on wavelength. See this link.

    I think you need to read round if you want more info. There's no way, unfortunately, you will find a single source that will give you many of your answers all in one place, I'm afraid.
  6. Jan 19, 2013 #5


    User Avatar
    Gold Member

    Not essential, I found this article that seems to answer your questions about frequency choices by insectivorous bats for echolocation. I’ve copied only the first paragraph of section three here for your perusal.

    Proc Biol Sci. 2007 April 7; 274(1612): 905–912.
    Published online 2007 January 16. doi: 10.1098/rspb.2006.0200
    PMCID: PMC1919403
    Bat Echolocation Calls: Adaptation and Convergent Evolution
    Gareth Jones* and Marc W Holderied

    “3. Frequency
    The range of frequencies exploited by echolocating bats makes perfect sense from an acoustics perspective. Bat echolocation calls vary in their dominant frequency approximately between 11 kHz (e.g. Euderma maculatum; Fullard & Dawson 1997) and 212 kHz (Cloeotis percivali; Fenton & Bell 1981). Most insectivorous bats call with dominant frequencies between 20 kHz and 60 kHz (Fenton et al. 1998). Lower frequencies are avoided because echoes from insect-sized targets are weak when the wavelength is longer than the insect wing length (Houston et al. 2004). For example, target strength (the ratio between incident and echo sound pressure) is reduced by approximately 25 dB at 1 m when the ratio of target wing length/sound wavelength drops from 1 to 0.2 (Houston et al. 2004). High frequencies are therefore necessary to detect small targets such as aerial insects. However, atmospheric attenuation is frequency-dependent, and limits the effective range of echolocation at high frequencies (Lawrence & Simmons 1982).”


  7. Jan 20, 2013 #6
    This is an interesting subject. I've actually been working on learning how to use echolocation myself. It's used by blind people to a greater or lesser degree, but there are many people out there who can use it to see just as well as any sighted person.
    The frequencies used are certainly dependent on the target object. Some frequencies are absorbed completely (meaning no signal response) while others are more useful. Here is a list of frequency absorption coefficients from my blog indicating which objects might be able to be more easily seen. For humans 3kHz is generally a good frequency because it resonates in the ear canal and requires less amplitude to be distinguished. But you can easily see from this chart that creating a signal which includes some lower frequencies will make it easier to see certain objects.
    For instance, the first line of the chart, carpet absorbs 45% of a 4kHz signal, but only 1% of a 125Hz signal, leaving more signal to be reflected and detected by the ear.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: Frequancy choice for echolocation & sonar
  1. Wave/Sonar Questions (Replies: 1)

  2. Radar and Sonar (Replies: 2)