Signal Localization

I am working on a problem involving signal localization and would like 
suggestions on signal processing alternatives.

Imagine that you are in the hallway of a hotel with a receiver in hand.  
Down the hall there are ten doors leading to ten suites.  Some dors are 
open, some are closed. Somewhere inside of ONE of the suites is a sound 
source that is producing an acoustic signal that sweeps linearly between 
2.0kHz and 6.0kHz.  The sweep rate is approximately 8kHz/sec.  Each sweep 
lasts approximately half a second and repeats approximately every two 
seconds.  

The task is to utlizize an optimal signal processing scheme in the receiver 
that will lead the receiver to the source. There is considerable multipath 
as well as attenuation between the source and the receiver. A constant-
amplitude template of the linear sweep is available. 

Any/all suggestions for optimal signal processing are welcome.  

   

Jimmy J. wrote:

> I am working on a problem involving signal localization and would like 
> suggestions on signal processing alternatives.
> 
> Imagine that you are in the hallway of a hotel with a receiver in hand.

Can you tell about the real problem, not an imaginary example?

> The task is to utlizize an optimal signal processing scheme in the receiver 
> that will lead the receiver to the source.

Do you know the direction or just the intensity of the signal?
Do you have the map of the area?
Do you know the current receiver location wrt the map?
How much of the processing power you can afford?

> There is considerable multipath 
> as well as attenuation between the source and the receiver. A constant-
> amplitude template of the linear sweep is available. 
> 
> Any/all suggestions for optimal signal processing are welcome.  

It depends.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

On 19 Aug, 02:58, "Jimmy J." <nos...@hotmail.com> wrote:
> I am working on a problem involving signal localization and would like
> suggestions on signal processing alternatives.

Practical DSP is based on that the signal is localized
at the inpot to the sensor device. If there is no signal
present near the sensor, there is little reason to proceed.

> Imagine that you are in the hallway of a hotel with a receiver in hand. =
=A0
> Down the hall there are ten doors leading to ten suites. =A0Some dors are
> open, some are closed. Somewhere inside of ONE of the suites is a sound
> source that is producing an acoustic signal that sweeps linearly between
> 2.0kHz and 6.0kHz. =A0The sweep rate is approximately 8kHz/sec. =A0Each s=
weep
> lasts approximately half a second and repeats approximately every two
> seconds. =A0
>
> The task is to utlizize an optimal signal processing scheme in the receiv=
er
> that will lead the receiver to the source. There is considerable multipat=
h
> as well as attenuation between the source and the receiver. A constant-
> amplitude template of the linear sweep is available.
>
> Any/all suggestions for optimal signal processing are welcome. =A0

This is a trivial student excercise. The key is to get
the problem formulation right, which happens to be where
you have failed.

Rune

Vladimir Vassilevsky <nospam@nowhere.com> wrote in
news:WOWdnWrDv9C26BbXnZ2dnUVZ_tCdnZ2d@giganews.com: 

> 
> 
> Jimmy J. wrote:
> 
>> I am working on a problem involving signal localization and would
>> like suggestions on signal processing alternatives.
>> 
>> Imagine that you are in the hallway of a hotel with a receiver in
>> hand. 
> 
> Can you tell about the real problem, not an imaginary example?

The term "imagine" was a bad choice of words.  The example that I 
described is a representative example of an actual real-world situation.   

 
>> The task is to utlizize an optimal signal processing scheme in the
>> receiver that will lead the receiver to the source.
> 
> Do you know the direction or just the intensity of the signal?

Neither. No apriori information is available to the user other than the 
existence of the signal.  The purpose of the receiver is to determine and 
provide directional and signal strength information to the user as the 
user hunts for the location of the signal source.  The receiver may 
contain a highly directional microphone or a pair of spaced microphones 
(or both) to determine and idicate the direction and relative strength of 
the signal as the user hunts for the location of the signal source. The 
receiver needs to be battery-powered, hand-held device.  Other than that, 
all options in terms of transducers, signal conditioning, signal 
processing, display, etc are on the table.
     


> Do you have the map of the area?
> Do you know the current receiver location wrt the map?

No maps are available. Also, the geography of the area can vary from one 
situation to another situation.  The only thing that common among the 
different situations is that the space is enclosed and is subdivided into 
various rooms of different shapes and sizes, sound absorbing 
characteristics, etc., as opposed to being an unobstructed outdoor 
situation.   


> How much of the processing power you can afford?
 
For the sake of discussion, assume that of an IPOD.


>> There is considerable multipath 
>> as well as attenuation between the source and the receiver. A
>> constant- amplitude template of the linear sweep is available. 
>> 
>> Any/all suggestions for optimal signal processing are welcome.  
> 
> It depends.


Then let me add that the signal exists in the presence of non-stationary 
background noise.  If it were not for the multipath, I can see how a 
matched filter would improve the S/N of the signal at receiver.  What I'd 
like to know is whether or not the linear sweep characteristic of the 
signal source is still beneficial in some way.  If it is, I'd like to 
know what type of signal processing can be utilized to provide that 
additional benefit.
   
 


 
> Vladimir Vassilevsky
> DSP and Mixed Signal Design Consultant
> http://www.abvolt.com

Jimmy J wrote:
> Vladimir Vassilevsky <nospam@nowhere.com> wrote in
> news:WOWdnWrDv9C26BbXnZ2dnUVZ_tCdnZ2d@giganews.com: 
>>
>>Jimmy J. wrote:

> No apriori information is available to the user other than the 
> existence of the signal.  The purpose of the receiver is to determine and 
> provide directional and signal strength information to the user as the 
> user hunts for the location of the signal source.

So your only option is going in the direction of the maximal signal 
strength.

>  The receiver may 
> contain a highly directional microphone or a pair of spaced microphones 
> (or both) to determine and idicate the direction and relative strength of 
> the signal as the user hunts for the location of the signal source. The 
> receiver needs to be battery-powered, hand-held device.  Other than that, 
> all options in terms of transducers, signal conditioning, signal 
> processing, display, etc are on the table.

So far, it is fairly basic. Where is a problem?

> Then let me add that the signal exists in the presence of non-stationary 
> background noise.  If it were not for the multipath, I can see how a 
> matched filter would improve the S/N of the signal at receiver.  What I'd 
> like to know is whether or not the linear sweep characteristic of the 
> signal source is still beneficial in some way.

The wideband signal averages out the signal strength variation due to 
the constructive and destructive interference; it is also less sensitive 
to the narowband interferrors. What kind of wideband signal to use is 
the matter of the application convenience.

>  If it is, I'd like to 
> know what type of signal processing can be utilized to provide that 
> additional benefit.

Textbook stuff.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

On 19 Aug, 19:46, Jimmy J <s...@spam.com> wrote:
> Vladimir Vassilevsky <nos...@nowhere.com> wrote innews:WOWdnWrDv9C26BbXnZ=
2dnUVZ_tCdnZ2d@giganews.com:
>
>
>
> > Jimmy J. wrote:
>
> >> I am working on a problem involving signal localization and would
> >> like suggestions on signal processing alternatives.
>
> >> Imagine that you are in the hallway of a hotel with a receiver in
> >> hand.
>
> > Can you tell about the real problem, not an imaginary example?
>
> The term "imagine" was a bad choice of words. =A0The example that I
> described is a representative example of an actual real-world situation. =
=A0
>
> >> The task is to utlizize an optimal signal processing scheme in the
> >> receiver that will lead the receiver to the source.
>
> > Do you know the direction or just the intensity of the signal?
>
> Neither. No apriori information is available to the user other than the
> existence of the signal. =A0The purpose of the receiver is to determine a=
nd
> provide directional and signal strength information to the user as the
> user hunts for the location of the signal source.

Use a simple hand-held recorder, possibly with an external
microphone, but shield the mic. Press it against each door
and see where the signal is strongest.

Rune

Rune Allnor <allnor@tele.ntnu.no> wrote in
news:d21916a2-d422-4fdb-aff9-79720c54debd@s6g2000vbp.googlegroups.com: 

> On 19 Aug, 19:46, Jimmy J <s...@spam.com> wrote:
>> Vladimir Vassilevsky <nos...@nowhere.com> wrote
>> innews:WOWdnWrDv9C26BbXnZ 
> 2dnUVZ_tCdnZ2d@giganews.com:
>>
>>
>>
>> > Jimmy J. wrote:
>>
>> >> I am working on a problem involving signal localization and would
>> >> like suggestions on signal processing alternatives.
>>
>> >> Imagine that you are in the hallway of a hotel with a receiver in
>> >> hand.
>>
>> > Can you tell about the real problem, not an imaginary example?
>>
>> The term "imagine" was a bad choice of words. &#4294967295;The example that I
>> described is a representative example of an actual real-world
>> situation. 
> &#4294967295;
>>
>> >> The task is to utlizize an optimal signal processing scheme in the
>> >> receiver that will lead the receiver to the source.
>>
>> > Do you know the direction or just the intensity of the signal?
>>
>> Neither. No apriori information is available to the user other than
>> the existence of the signal. &#4294967295;The purpose of the receiver is to
>> determine a 
> nd
>> provide directional and signal strength information to the user as
>> the user hunts for the location of the signal source.
> 
> Use a simple hand-held recorder, possibly with an external
> microphone, but shield the mic. Press it against each door
> and see where the signal is strongest.
> 
> Rune

Aside from the inconvenience, that will only get you past the first 
closed door. In my residence, for example, which has both upper and lower 
floors, there are no closed doors except for a bathroom that may 
occasionally be in use. So once you get past the front door, the simple 
hand-held recorder with shielded mic are useless. In fact, unless the mic 
is highly directional, you won't know whether to turn right and go down 
the entrance hall into the living room, or turn left toward the bedrooms 
or go take a dog leg and go down the stairs to the lower floor. And on 
top of that there's a potential S/N issue if one of the kids is screaming 
at the top of his lungs.  Additionally, in a restaurant that may not have 
carpeted floors, there may also be considerable reverberant sound which 
could even render a highly directional mic virtually useless. 
 

Jimmy J wrote:
> Rune Allnor <allnor@tele.ntnu.no> wrote in
> news:d21916a2-d422-4fdb-aff9-79720c54debd@s6g2000vbp.googlegroups.com: 
> 
>> On 19 Aug, 19:46, Jimmy J <s...@spam.com> wrote:
>>> Vladimir Vassilevsky <nos...@nowhere.com> wrote
>>> innews:WOWdnWrDv9C26BbXnZ 
>> 2dnUVZ_tCdnZ2d@giganews.com:
>>>
>>>
>>>> Jimmy J. wrote:
>>>>> I am working on a problem involving signal localization and would
>>>>> like suggestions on signal processing alternatives.
>>>>> Imagine that you are in the hallway of a hotel with a receiver in
>>>>> hand.
>>>> Can you tell about the real problem, not an imaginary example?
>>> The term "imagine" was a bad choice of words.  The example that I
>>> described is a representative example of an actual real-world
>>> situation. 
>>  
>>>>> The task is to utlizize an optimal signal processing scheme in the
>>>>> receiver that will lead the receiver to the source.
>>>> Do you know the direction or just the intensity of the signal?
>>> Neither. No apriori information is available to the user other than
>>> the existence of the signal.  The purpose of the receiver is to
>>> determine a 
>> nd
>>> provide directional and signal strength information to the user as
>>> the user hunts for the location of the signal source.
>> Use a simple hand-held recorder, possibly with an external
>> microphone, but shield the mic. Press it against each door
>> and see where the signal is strongest.
>>
>> Rune
> 
> Aside from the inconvenience, that will only get you past the first 
> closed door. In my residence, for example, which has both upper and lower 
> floors, there are no closed doors except for a bathroom that may 
> occasionally be in use. So once you get past the front door, the simple 
> hand-held recorder with shielded mic are useless. In fact, unless the mic 
> is highly directional, you won't know whether to turn right and go down 
> the entrance hall into the living room, or turn left toward the bedrooms 
> or go take a dog leg and go down the stairs to the lower floor. And on 
> top of that there's a potential S/N issue if one of the kids is screaming 
> at the top of his lungs.  Additionally, in a restaurant that may not have 
> carpeted floors, there may also be considerable reverberant sound which 
> could even render a highly directional mic virtually useless. 
>  

I'm going to shock the group by agreeing with Mr. Vassilevsky.
This seems to have too much "imaginary example".
Your problem description just doesn't seem to "hang 
together"/"have focus"/"????".

There was a recent thread (which I can't find at the moment) 
discussing good tech writing, it may have useful info/links.

Richard Owlett <rowlett@pcnetinc.com> wrote in
news:K-CdnYy_Xr7h9RHXnZ2dnUVZ_gCdnZ2d@supernews.com: 

> Jimmy J wrote:
>> Rune Allnor <allnor@tele.ntnu.no> wrote in
>> news:d21916a2-d422-4fdb-aff9-79720c54debd@s6g2000vbp.googlegroups.com:
>> 
>>> On 19 Aug, 19:46, Jimmy J <s...@spam.com> wrote:
>>>> Vladimir Vassilevsky <nos...@nowhere.com> wrote
>>>> innews:WOWdnWrDv9C26BbXnZ 
>>> 2dnUVZ_tCdnZ2d@giganews.com:
>>>>
>>>>
>>>>> Jimmy J. wrote:
>>>>>> I am working on a problem involving signal localization and would
>>>>>> like suggestions on signal processing alternatives.
>>>>>> Imagine that you are in the hallway of a hotel with a receiver in
>>>>>> hand.
>>>>> Can you tell about the real problem, not an imaginary example?
>>>> The term "imagine" was a bad choice of words.  The example that I
>>>> described is a representative example of an actual real-world
>>>> situation. 
>>>  
>>>>>> The task is to utlizize an optimal signal processing scheme in
>>>>>> the receiver that will lead the receiver to the source.
>>>>> Do you know the direction or just the intensity of the signal?
>>>> Neither. No apriori information is available to the user other than
>>>> the existence of the signal.  The purpose of the receiver is to
>>>> determine a 
>>> nd
>>>> provide directional and signal strength information to the user as
>>>> the user hunts for the location of the signal source.
>>> Use a simple hand-held recorder, possibly with an external
>>> microphone, but shield the mic. Press it against each door
>>> and see where the signal is strongest.
>>>
>>> Rune
>> 
>> Aside from the inconvenience, that will only get you past the first 
>> closed door. In my residence, for example, which has both upper and
>> lower floors, there are no closed doors except for a bathroom that
>> may occasionally be in use. So once you get past the front door, the
>> simple hand-held recorder with shielded mic are useless. In fact,
>> unless the mic is highly directional, you won't know whether to turn
>> right and go down the entrance hall into the living room, or turn
>> left toward the bedrooms or go take a dog leg and go down the stairs
>> to the lower floor. And on top of that there's a potential S/N issue
>> if one of the kids is screaming at the top of his lungs. 
>> Additionally, in a restaurant that may not have carpeted floors,
>> there may also be considerable reverberant sound which could even
>> render a highly directional mic virtually useless. 
>>  
> 
> I'm going to shock the group by agreeing with Mr. Vassilevsky.
> This seems to have too much "imaginary example".
> Your problem description just doesn't seem to "hang 
> together"/"have focus"/"????".
> 
> There was a recent thread (which I can't find at the moment) 
> discussing good tech writing, it may have useful info/links.


It would be a lot more productive if you and others would dispense with 
the criticism and simply ask for whatever additional information you 
think that you require.  The situation is a modified version of hide and 
seek that everyone has played as a child. The two modifications are: 1) 
the person who is hiding chirps every couple of seconds, and 2) the 
seeker is deaf and therefore needs to use an acoustic receiver with 
appropriate signal processing and display to assist him in finding the 
person who is hiding.  The objective is to find the chirping person who 
is hiding.  I have described the chirp signal, I have described the 
environment, and I have stated the two primary complicating factors which 
are 1) unkunknown amounts of miltipath, and 2) an unknown amount of 
uncorrelated noise that can have spectral content that may partially 
overlap the spectral content the signal.  If you and others feel that 
this description is ambiguous or inadequate, then tell me specifically 
what is ambiguous and what is missing and I will fill in the missing 
details and/or provide whatever clarification is necessary.  Simply 
asserting that the problem description doesn't seem to "hang together" or 
"have focus" amounts to engaging in the very lack of specificity that you 
are criticizing.
 

  

What EXACTLY is your question?



Jimmy J wrote:
> Richard Owlett <rowlett@pcnetinc.com> wrote in
> news:K-CdnYy_Xr7h9RHXnZ2dnUVZ_gCdnZ2d@supernews.com: 
> 
> 
>>Jimmy J wrote:
>>
>>>Rune Allnor <allnor@tele.ntnu.no> wrote in
>>>news:d21916a2-d422-4fdb-aff9-79720c54debd@s6g2000vbp.googlegroups.com:
>>>
>>>
>>>>On 19 Aug, 19:46, Jimmy J <s...@spam.com> wrote:
>>>>
>>>>>Vladimir Vassilevsky <nos...@nowhere.com> wrote
>>>>>innews:WOWdnWrDv9C26BbXnZ 
>>>>
>>>>2dnUVZ_tCdnZ2d@giganews.com:
>>>>
>>>>>
>>>>>>Jimmy J. wrote:
>>>>>>
>>>>>>>I am working on a problem involving signal localization and would
>>>>>>>like suggestions on signal processing alternatives.
>>>>>>>Imagine that you are in the hallway of a hotel with a receiver in
>>>>>>>hand.
>>>>>>
>>>>>>Can you tell about the real problem, not an imaginary example?
>>>>>
>>>>>The term "imagine" was a bad choice of words.  The example that I
>>>>>described is a representative example of an actual real-world
>>>>>situation. 
>>>>
>>>> 
>>>>
>>>>>>>The task is to utlizize an optimal signal processing scheme in
>>>>>>>the receiver that will lead the receiver to the source.
>>>>>>
>>>>>>Do you know the direction or just the intensity of the signal?
>>>>>
>>>>>Neither. No apriori information is available to the user other than
>>>>>the existence of the signal.  The purpose of the receiver is to
>>>>>determine a 
>>>>
>>>>nd
>>>>
>>>>>provide directional and signal strength information to the user as
>>>>>the user hunts for the location of the signal source.
>>>>
>>>>Use a simple hand-held recorder, possibly with an external
>>>>microphone, but shield the mic. Press it against each door
>>>>and see where the signal is strongest.
>>>>
>>>>Rune
>>>
>>>Aside from the inconvenience, that will only get you past the first 
>>>closed door. In my residence, for example, which has both upper and
>>>lower floors, there are no closed doors except for a bathroom that
>>>may occasionally be in use. So once you get past the front door, the
>>>simple hand-held recorder with shielded mic are useless. In fact,
>>>unless the mic is highly directional, you won't know whether to turn
>>>right and go down the entrance hall into the living room, or turn
>>>left toward the bedrooms or go take a dog leg and go down the stairs
>>>to the lower floor. And on top of that there's a potential S/N issue
>>>if one of the kids is screaming at the top of his lungs. 
>>>Additionally, in a restaurant that may not have carpeted floors,
>>>there may also be considerable reverberant sound which could even
>>>render a highly directional mic virtually useless. 
>>> 
>>
>>I'm going to shock the group by agreeing with Mr. Vassilevsky.
>>This seems to have too much "imaginary example".
>>Your problem description just doesn't seem to "hang 
>>together"/"have focus"/"????".
>>
>>There was a recent thread (which I can't find at the moment) 
>>discussing good tech writing, it may have useful info/links.
> 
> 
> 
> It would be a lot more productive if you and others would dispense with 
> the criticism and simply ask for whatever additional information you 
> think that you require.  The situation is a modified version of hide and 
> seek that everyone has played as a child. The two modifications are: 1) 
> the person who is hiding chirps every couple of seconds, and 2) the 
> seeker is deaf and therefore needs to use an acoustic receiver with 
> appropriate signal processing and display to assist him in finding the 
> person who is hiding.  The objective is to find the chirping person who 
> is hiding.  I have described the chirp signal, I have described the 
> environment, and I have stated the two primary complicating factors which 
> are 1) unkunknown amounts of miltipath, and 2) an unknown amount of 
> uncorrelated noise that can have spectral content that may partially 
> overlap the spectral content the signal.  If you and others feel that 
> this description is ambiguous or inadequate, then tell me specifically 
> what is ambiguous and what is missing and I will fill in the missing 
> details and/or provide whatever clarification is necessary.  Simply 
> asserting that the problem description doesn't seem to "hang together" or 
> "have focus" amounts to engaging in the very lack of specificity that you 
> are criticizing.
>  
> 
>   
> 
> 
> 
>