[600MRG] Receiver bandwidth for WSPR

[Brian Pease]

WSPR is ~6Hz BW.
I agree with what you say.  I think WSPR looks only over it's 200Hz band 
to search for signals and determine the noise level.
I just finished a lab test using a U3 as a WSPR source and my 3586C as 
the receiver, with no outside antennas, hence no atmospheric noise.
At -21dB s/n reported by WSPR, there is zero difference in decoded s/n 
between 3100Hz and 400Hz receiver bandwidths.
At -3dB and -6dB s/n, the 400Hz BW was very slightly better, perhaps 
0.5dB.  I didn't run long enough to get a good average.
I will now try a "real world" test with the real atmospheric and 
man-made noise that I have.
On 3/21/2014 11:17 AM, Edward R Cole wrote:
> WSPR and most of K1JT's sw programs look over a wide spectrum.  After 
> conversion to digital the spectrum is divided into several small 
> slices of bw which are then examined for signal.  I don't recall what 
> the RBW is for WSPR but I believe it is much smaller bw than used by 
> JT65, hence its superior sensitivity.
>
> But signal level calculations are done in reference to 2.5 KHz of 
> noise.  MDS specs for many ham radios are specified at 500-Hz which 
> makes the sensitivity spec look 5 times better:  10Log (2.5/0.5) = 
> 6.99 dB better.
>
> If RBW = 10 Hz, the improvement over 2.5 KHz is roughly 24 dB. Note 
> that JT65B decodes signals down to about -29.
>
> CW and JT65 occupy about the same bandwidth (CW speed widens its 
> spectrum proportionate to speed).  JT65 uses patented noise reduction 
> routines to improve sensitivity further (Reed-Solomon). Plus since 
> JT65 and WSPR are synchoronous codes this adds to detection SNR.  CW 
> is asynchronous.
>
> 73, Ed - KL7UW
>
> At 05:24 AM 3/21/2014, you wrote:
>> But is WSPR really looking at noise outside the 200Hz WSPR band?  I 
>> have assumed that the software just looks at noise over a small 
>> bandwidth very close to each signal's frequency, then simply converts 
>> it to what it would be at 2500 Hz.  If it averaged the whole 2500Hz 
>> I.F. bandwidth then the 400Hz BW s/n numbers would be about 8dB 
>> better since the noise drops towards zero outside +/- 200Hz.
>> On 3/21/2014 9:08 AM, jrusgrove at comcast.net wrote:
>>> WSPR s/n measurements will be incorrect if the receiver bandwidth is 
>>> set to less than the expected SSB bandwidth.
>>>
>>> Jay W1VD  WD2XNS  WE2XGR/2
>>>
>>>
>>>
>>> ----- Original Message ----- From: "Brian Pease" 
>>> <bpease2 at myfairpoint.net>
>>> To: <600MRG at w7ekb.com>
>>> Sent: Friday, March 21, 2014 8:48 AM
>>> Subject: [600MRG] Receiver bandwidth for WSPR
>>>
>>>
>>>> I have found that my WSPR reception seems slightly better using the 
>>>> wide 3100Hz bandwidth on my HP3586C receiver compared to 400Hz. 
>>>> Have others tested this? My theory is that the static bursts reach 
>>>> the sound card sharp and crisp without the ringing from the narrow 
>>>> physical filter.  There should be no effect on the desired signals. 
>>>> The extra noise on other audio frequencies may not matter. Of 
>>>> course it could be related to my particular man made 
>>>> noise.................................
>>>>
>>>> _______________________________________________
>>>> 600MRG mailing list
>>>> 600MRG at w7ekb.com
>>>> http://w7ekb.com/mailman/listinfo/600mrg_w7ekb.com
>>>
>>>
>>> _______________________________________________
>>> 600MRG mailing list
>>> 600MRG at w7ekb.com
>>> http://w7ekb.com/mailman/listinfo/600mrg_w7ekb.com
>>
>>
>> _______________________________________________
>> 600MRG mailing list
>> 600MRG at w7ekb.com
>> http://w7ekb.com/mailman/listinfo/600mrg_w7ekb.com
>
> 73, Ed - KL7UW
> http://www.kl7uw.com
>     "Kits made by KL7UW"
> Dubus Mag business:
>     dubususa at gmail.com
>