Polarisation diversity reception on 50MHz

[page updated 02 Sep 2016]

See my Diversity presentation given at G3WOS BBQ 08/08/2015 (pdf, 1.8MB)

Here are some recordings of dx heard here in IO81QC on 6m using polarisation diversity. Polarisation diversity is often used on the LF and HF bands where it can contribute significantly to the readability of the signal. (For recordings of its use on other bands, see for example the websites of EI6IZ or N1EU or W8JI)

The recordings here are in mp3 format in stereo with 11kHz sampling, the left hand channel being horizontal and the right being vertical polarisation. The recordings are best heard using stereo headphones.

Date dd/mm/yy	UTC	Content	km	Propagation mode	File size	Duration	Comments
NEW!		2015 season
14/05/15	1529	3B9FR in MH10rh	10144	F2 +Es?	373kB	47sec	Diversity working LZ2DF. Strong and then fades out.
14/05/2015	1516	3B9FR in MH10rh	10144	F2 +Es?	696kB	1m 29s	Diversity working G3ZSS and me. Very weak on vertical.
02/05/2015	1310	ZS4TX/6 in KG47ur	8745	TEP + Es	275kB	35sec	Diversity, working G4FJK.
		2014 season
24/09/14	1901	ZS4TX/6 in KG47ur	8745	TEP + Es? (no Es evident though)	2460kB	5mins	Diversity but very weak on vertical. Rapid phase shifts between H nd V.
23/06/14	1928	ZD7VC/B in IH74	7456	Es + TEP or nEs?	679kB	1m 27s	Slightly rough sounding note, with phase changes typical of Es
05/04/14	1643-1647	C5YK in IK13pk	4373	F2	671kB	1m 25s	Diversity in ssb with gaps edited out. Quite fast phase changes between H and V.
04/04/14	1621	TZ6BB in IK62ao	4305	F2	309kB	39s	Diversity in ssb, working G4FJK. Fairly stable phase shift between V and H.
04/04/14	1552	TZ6BB in IK62ao	4305	cw, F2	238kB	30s	Mono only, working me.
20/03/14	1511	6V7SIX/B in IK14lk	4274	F2	424kB	54s	Diversity. Beacon with slowly drifting phase between H and V.
		2012 season
02/06/12	1627	6O3A in LJ29an	6261	cw, Multi hop Es	456kB	58s	Mono only. Working G4PWA
15/05/12	1549	FR4NT in LG79pa	9792	Es + TEP	315kB	40s	Diversity. Working G5WQ who is audible by backscatter as well as direct
05/05/12	1720	Z21BB in KH52ne	8333	cw, Es + TEP	287kB	37s	Mono only.
09/03/12	1123	6V7SIX in IK14lk	4274	F2	610kB	1m 18s	Diversity. Remarkably stable with virtually constant phase shift between H and V
		2011 season
23/10/11	1701	6V7Q in IK14LL	4270	F2	648kB	1min 22s	ssb. Much slower fading and more correlation between H and V than Es propagation
21/10/11	1706	3XY1D in IJ39dl	4735	cw, F2	626kB	1min 20s	cw cq. Ditto
26/09/11	2147	GB3LER beacon in IP90jd	1008	Auroral-E	545kB	1min 09s	Rapid flutter, stronger on h pol
26/09/11	2131	OY6BEC beacon in IP62oa	1239	Auroral-E	999kB	2min 07s	Rapid flutter, signal strong on both polarisations
26/09/11	1713	LA8AJA in JP50hp	1338	cw, Aurora	566kB		Mono only - 57A
25/09/11	1645	5N7M in JJ39sb	4763	cw, F2 - 1st qso on 6m F2 for 9 years.	544kB	1min 10s	Mono only - weak signal. 559 then fades out.

16/06/11	2023-2028	PJ4E in FK52ud	7264	cw, Multi hop Es	2MB	2min 10s	Edited highlights - working G3WOS, GM3WOJ, G4RRA, G3TCT
18/06/11	1335	TF/DL3GCS in HP93dt	1800	cw, Es	1.3MB	1min 22s	Working G4EZT. TF/DL3GCS was using a vertical antenna.
18/06/11	1416	VO1FU beacon in GN37ol	3587	Multi hop Es	491kB	31s
25/06/11	1656	K1TOL in FN44vg	4935	cw, Multi hop Es	967kB	2min 4s	Channels swapped: LH=Vert, RH=Horiz. I am transmitting on the vertical.
27/06/11	1930	PJ2/DJ9ON in FK52lf	7308	cw, Multi hop Es	591kB	1min 16s	with QRM
30/06/11	1834	PJ76 in FK88	6399	Multi hop Es	551kB	1min 10s	ssb wkg G0VQY
30/06/11	1942	PJ76 in FK88	6399	cw, Multi hop Es	513kB	1min 6s	cw split in mono, working M0BKL, G3TCT
01/03/11	1800 approx	MM0AMW in IO75ej	516	cw, Aurora	444kB	57s	working G4WJS - audible by both Ar and direct

		2010 season
08/04/10	1243	CS5BLA beacon in IM57px	1534	Es	356kB	45s	Early in Es season.
25/05/10	1650	A92IO in LL56fe	5223	Multi hop Es	553kB	1min 10s	working split, intervals between transmissions edited out
25/05/10	1717	1B1AB in KM65pi	3345	Multi hop Es	624kB	1min 20s	in QSO with G0CHE
04/06/10	1633	A92IO in LL56fe	5223	Multi hop Es	225kB	29s	working me
05/06/10	0821	E4X in KM71oq	3758	Multi hop Es	465kB	1 min	working split up 5
10/06/10	1918	FY1FL in GJ35qd	6873	Multi hop Es	398kB	51s	working me
22/06/10	1507	A71EM in LL55rj	5356	cw, Multi hop Es	485kB	1min 2s	Working me, mono recording only
22/06/10	2041	JW9SIX in JQ94lm	2784	Multi hop Es	361kB	46s	Bear Island beacon, vertical ant
24/06/10	0845	ST2AR in KK65er	5009	cw, Multi hop Es	465kB	59s	Callers not being heard
28/06/10	1612	FS/W6JKV in FK88	6354	cw, Multi hop Es	519kB	1min 6s	Working G5WQ
28/06/10	1619	FS/W6JKV in FK88	6354	cw, Multi hop Es	426kB	54s	Working me
30/06/10	2035	JW7QIA in JQ68tb	3067	cw, Multi hop Es	446kB	57s	Working me, mono recording only
19/07/10	1924	YO4RYU/MM in IM38dh	1665	Es	415kB	52s	Recording with AGC off. G0JHC heard via backscatter on both pols, and others calling heard on V pol.
19/07/10	1946	CS5BLA beacon in IM57px	1534	Es	712kB	1min 31s	Recording with AGC off. Breaks in RH channel caused by me checking sub rx. 3 second period sinusoidal fading evident.
05/08/10	1457	GB3LER beacon in IP90jd	1008	Es	290kB	37s	Short skip Es, recorded with AGC off
05/08/10	1505	GM8LFB in IO88kk	815	Es	346kB	44s	Short skip Es, AGC on
06/08/10	1339	N3DB in FM18su	5707	cw, Multi hop Es	351kB	45s	working me

In general the signals (which started out as horizontal polarisation*) show a significant vertical component. Very rapid changes in the H-V relative signal strength are evident, even several times a second. There are some instances of the vertical signal being much better than the horizontal. The signal from G0CHE (tropo, 139km) is only heard on H pol. It will be interesting to see in the next few years whether signals via F2 propagation show similar characteristics.
*except JW9SIX - vertical

The 3 second period sinusoidal fading of the CS5BLA beacon on 19/07/10 can be seen here (screen shot of Audacity, upper trace is LH channel = H pol)
Audacity screen shot of CS5BLA beacon

The effect we are seeing is called "Faraday rotation". In the ionosphere and in the presence of the earth's magnetic field, the linearly-polarised wave from the transmitter is split into two essentially circularly-polarised waves (called the ordinary and extraordinary waves) with opposite senses of rotation. (Note that a linearly-polarised wave can be defined as the vector sum of two circularly-polarised waves.)
LH circular polarised wave

Left hand circularly polarised wave

Right hand circularly polarised wave

The two waves in the ionosphere have different wavelengths and phase speeds. Therefore, the polarisation of the resultant linearly-polarised wave appears to rotate (Faraday rotation). The effect came to prominence when the first satellite transmissions on frequencies from 20 to 136MHz were received on the ground. The characteristic fading appeared as a rectified sine wave with deep nulls, similar to that seen here.

The rapid changes in received polarisation can be demonstrated rather well using an oscilloscope with one channel fed to the x axis and the other to the y axis. This can be achieved very simply on a PC by downloading and installing "Soundcard Oscilloscope". Open the scope and then play the mp3 file - I use "realPlayer". Select X-Y Graph on the scope and see the effects visually!

Or see a quick video of the effects here (37sec, .avi format, 6MB). This is the GB3LER beacon as recorded above on 5 August 2010, range 1059km.

Lissajous figure of JW9SIX beacon

Equipment used

Elecraft K3 with subreceiver in diversity mode - see my K3 page.

Recordings - made with a Zoom H2 recorder, uploaded and then edited and processed with Audacity to produce mp3 format.

Antennas used - 5 ele M²yagi for H Pol and a dipole with reflector for V Pol - the dipole and reflector are mounted on the 4m yagi above the 6m yagi.

yagis and vertical dipole

The calculated gain of the M² yagi as installed is 9.2dBd. The calculated gain of the dipole with its reflector is 2.8dBd.

Polarisation of auroral backscatter at 50MHz

In working stations by Auroral backscatter, it is interesting to speculate whether a polarisation diversity receiver might offer some advantage in improved signal to noise ratio. Auroral backscatter occurs in the E region of the ionosphere where ionised particles move along lines of the geomagnetic field. The scattering from such field-aligned particles requires a particular geometry between transmitter and receiver, and this has been covered elsewhere in detail.

A search of the academic literature online reveals a number of experiments, some early ones in 1954 as well as more recent work eg Ref 1. This reference includes some interesting visualisation of the received signal polarisation and its variation over quite short periods of time.

There are two rather nice graphs showing first a highly polarized return, and in the second an almost completely depolarized one.

Here is their diagram of a strongly polarised return at 0430 to 0435UTC

Strongly polarised Ar return

They interpret the strongly polarised return as implying

...a small scattering volume and/or a small number of discrete scatterers located close to one another...

Here is a depolarised return at 0345 to 0350UTC on the same day – ie 40 mins earlier.

Depolarised Ar return

They interpret this to mean that the depolarized wave is

...coming from a number of individual scattering volumes within the scattering region (i.e., the effective radar viewing region), which are each influenced differently by Faraday rotation to and from the scattering region

How does this help us to work Aurora on 50MHz? It means there can sometimes be benefits in using diversity reception but this will not perhaps be typical.

Ref 1. Polarization of auroral backscatter at 50 MHz G. C. Hussey, J. A. Koehler, and G. J. Sofko; Radio Science, Volume 32, Number 2, Pages 541-555, March-April 1997

Equipment used

Further reading and interesting links

Polarisation of auroral backscatter at 50MHz