ASSESSING THE PERFORMANCE OF THE RSP RECEIVER SYSTEM IN HF

General discussions
glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: ASSESSING THE PERFORMANCE OF THE RSP RECEIVER SYSTEM IN HF

Post by glovisol » Thu Oct 04, 2018 12:38 pm

5. ANTENNA DATA AND NOISE FLOOR EVALUATION

The Noise Power delivered to the Receiver by a loss free antenna is equal to the flux density of the electromagnetic radiation impinging on the antenna times the antenna Effective Aperture. As will be shown in the Appendix, the Antenna Effective Aperture is nothing else than the Antenna Gain Factor G.

Table 4 provides data for several antenna types in common use in HF. We can now evaluate noise performance of a practical receiving system, leaving out only the influence of antenna elevation. The wideband Beverage antenna data cannot be tabulated because gain depends on wavelength ratio and is shown in the graph of Figure 1. Equations shown in Figure 1 can be used to calculate gain for any Beverage antenna length, different heights above ground and at any frequency. The graph is valid for average dry soil.

Examples amended 05/10/2018

EXAMPLE 1
A half wave dipole, operating at 14.3 MHz (20m Band) measures a noise floor of -120 dBm. Coaxial line / balun loss amounts to 1.2 dB.
From Table 4 we find r’ = -2.15 dB.
Therefore the corrected antenna nose floor Anf= -120 + 1.2 + 2.15 = -116.6 dBm
Applying the decile deviations: -113.6 > Anf > -119.6
In conclusion (Table 3) we are in a “Quiet rural” situation.

EXAMPLE 2
If the same half wave dipole, at the same frequency, exhibited an Anf of -98 dBm, it would be positioned in a “City” situation.

EXAMPLE 3
An L = 135 m Beverage antenna, at an average height of 6 m operates at 3.6 Mhz and measures an average noise floor of -104 dBm. Balun loss is negligible. The Beverage is a wideband antenna and gain depends on the ratio of length L to wavelength W.
At 3. 6 Mhz Wavelength W = 300/3.6 = 83 m.
L/W = 135/83 = 1.62
From Figure 1, r’ = -2.8 dB.
Therefore the corrected antenna nose floor Anf= - 104 + 2.8 = - 101 dBm
Applying the decile deviations: -98 > Anf > -104
In conclusion (Table 3) we are in the bottom end of the “Quiet rural” situation.

Some data in Table 4 has been taken from: "RDF METRIC"
http://www.seed-solutions.com/gregordy/ ... Metric.htm

Beverage antenna data has been derived from:
BEVERAGE ANTENNAS FOR HF COMMUNICATIONS,
DIRECTION FINDING AND OVER-THE-HORIZON RADARS
By J. Litva and B.I. Rook / Report No. 1282 /Dept. of Communications / Ottawa, August 1976
Attachments
Beverage Gain.jpg
Figure 1 - Beverage Antenna Gain Vs. Wavelength ratio
Beverage Gain.jpg (284.38 KiB) Viewed 16853 times
Antenna data table.jpg
Table 4 - Antenna data for several antenna types
Antenna data table.jpg (95.8 KiB) Viewed 16853 times

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: ASSESSING THE PERFORMANCE OF THE RSP RECEIVER SYSTEM IN HF

Post by glovisol » Fri Oct 05, 2018 1:46 pm

6. APPENDIX – DEMONSTRATION OF THE ANTENNA CORRECTION PROCEDURE

We rewrite here again the expression developed in Part 3:

Pn = Fa + B – 204 [dBm] (8) Solving for Fa, effective antenna Noise Figure, we have:
Fa = Pn + 204 – 10*LOG10(b) [dB] (8bis)

As already stated, the Noise Power Pn delivered to the receiver by a loss free antenna is equal to the flux density of the electromagnetic radiation impinging on the antenna, times the antenna Effective Aperture. By using the well known universal relationship:

Pn = ((e^2)/(120*π))*((g*(λ^2)/(4*π) (12)

120*π - free space impedance [Ω]
e - noise electric field strength, r.m.s. value [uV/m] on a bandwith b [Hz]
g - antenna gain referred to an isotropic radiator
λ - wavelength in meters, where λ = 300/F and F frequency in Mhz

Analogously to what we have done before, we define:

G = 10*LOG10(g) [dB] (13) antenna gain factor
E = 10*LOG10(e^2) [uV/m] (14)

and (12) by substitution, becomes:

Pn = E – 120 – (20*LOG10(F)) + G + 10*LOG10((300^2)/(4*(π^2)*120)) [dbm] (15)

Substituting (15) into (8bis) and calculating:

204 – 120 + (10*LOG10((300^2)/(4*(π^2)*120))) = 96.79 dB which is the gain constant for the Isotropic Antenna, we obtain the following expressions for Fa and for E:

Fa = E – 20*LOG10(F) – B + G + 96.79 [db] (16)

E = Fa + 20*LOG10(F) + B – G – 96.79 [uV/m] (17)

Both Fa and E depend on the antenna gain factor G. Therefore the noise behavior of any antenna can be specified by modifying the constant 96.79 of (16) with the gain of that same antenna referred to the isotropic. No additional compensation is necessary for system impedance, as all calculations are done in dB and dBm.


7. CONCLUSION

In this thread, based on ITU Recommendation ITU-R P.372-13, we have learned how to classify Receiving Systems with respect to antenna noise coming from various sources, using the precision measurement of Noise Floor provided by the RSP class receivers. The fundamental instrument for classification and comparison is the data provided by Table 3. We have also learned how to correct the result to accommodate different antenna characteristics and/or line feeds.

However we conclude that, apart from a few exceptions, the corrections for individual antenna characteristics are small and not significant, being often smaller that the known decile corrections also available from Table 3.
Likewise (CCIR 322-2) experimental information on the effects of directivity points to the fact that variations will be usually less than 6 dB. Furthermore variations due to differences in polarisation will be even lower. Therefore unmodified Table 3 data can be directly used in the majority of cases.

References:
- Recommendation ITU-R P.372-13, ITU 2016.
- CCIR Report 322-2.
- "Techniques for Estimating the effect of man-made radio noise on distributed military systems", D.B. Sailors.
Ocean and Atmospheric Sciences Division - Naval Ocean Systems Center, San Diego, CA.
- "The effective antenna noise figure Fa for a vertical loop antenna and its application to extremely low frequency/very
low frequency atmospheric noise", Anthony C. Fraser-Smith. Radio Science, Vol. 42, 2007.
- "Beverage antennas for HF communications, direction finding and over-the-horizon radars", J. Litva & B.I. Rook, Report 1282, Dept. National Defence, August 1976, Ottawa.

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: ASSESSING THE PERFORMANCE OF THE RSP RECEIVER SYSTEM IN HF

Post by glovisol » Wed Oct 10, 2018 11:37 am

The enclosed graph summarises Radio Noise evaluations, data and measurements discussed in this thread.
Attachments
Noise Floor Summary.jpg
Noise Floor summary graph
Noise Floor Summary.jpg (88.14 KiB) Viewed 16772 times

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: ASSESSING THE PERFORMANCE OF THE RSP RECEIVER SYSTEM IN HF

Post by glovisol » Tue Oct 23, 2018 6:26 pm

One week noise floor variation in Northern Italy. Data taken with already mentioned 135 m Beverage antenna at 17:00 local time (15:00 GMT) each day. This is not local noise, as verified by radio ham reports in QSO's, but a general situation, especially today, when a remarkable noise increase was registered on all frequencies.
Attachments
One week noise floor .jpg
One week's noise on HF bands
One week noise floor .jpg (50.26 KiB) Viewed 16575 times

Reason: No reason

glovisol
Posts: 662
Joined: Thu May 10, 2018 6:42 pm
Location: Piedmont, Italy

Re: ASSESSING THE PERFORMANCE OF THE RSP RECEIVER SYSTEM IN HF

Post by glovisol » Tue Nov 27, 2018 12:29 pm

CONCLUSION BASED ON MEASUREMENT RESULTS

In the thread:

https://www.sdrplay.com/community/viewt ... f=5&t=3436

We have seen the values of the Noise Floor in the frequency range 1.8 to 29 MHz over two periods of 15 days: period #1, last 15 days of October and period #2, first 15 days of November. Even if this looks as a very brief time period, still, with the added experience of the balance of November, we can reach some conclusions.

1) A specific antenna in a given area can be regarded as a remarkably stable noise source until there is the extraneous intervention of (a) electrical noise caused by atmospheric activity, such as rainstorms and/or of (b) electrical noise caused by local disturbances.

2) The quantitative level of the antenna noise can be accurately determined using a quality instrumentation system, such as the RSP Receiver / SDRuno over a period of several days: the longer the period, the higher the quality and reliability of the level obtained.

3) This result can be used to obtain a comparative value of the receiving system quality, using the classifications and rules given and examined in this thread. For the system used, we have obtained the following values:
F………1.8 MHz…………-103 dBm………QUIET RURAL
F………3.6 MHz…………-103 dBm………RURAL / QUIET RURAL
F………7.2 MHz…………-104 dBm………RURAL
F…….14.2 MHz…………-110 dBm………RURAL
F…….21.2 MHz…………-110 dBm………RURAL
F…….29.0 MHz…………-114 dBm………RURAL

4) It is true that measured Noise Floor values for October are higher, but these were provoked by much higher than normal, indeed quite exceptional, electrical activity due to stormy weather. This is additional proof of how an exceptional situation can become immediately apparent.

5) Once reliable Floor Noise values for a given Receiving System are obtained, they may serve as reference to verify local noise variations, to help in tracking down possible local noise radiation issues.

6) Finally attention must be devoted to antenna characteristics and their relationship to noise.
• Higher gain / larger aperture antennas (the 135 m Beverage is an example) deliver higher noise level and, when present, higher wanted signal level. Both may significantly exceed the threshold sensitivity of the RSP receiver: in these cases using attenuators between antenna and receiver is a good and advantageous practice.
• Zero gain antennas are in the borderline where attenuators may or may not be used to obtain a significant advantage.
• Lossy / small aperture antennas, on the contrary, will need specialized, low noise and distortion proof amplification in presence of high level signals, to achieve an acceptable level of system sensitivity.

Reason: No reason

Post Reply