Tuesday 11 September, 2007, 14:07 - Radio RandomnessOK, so it's not as snappy as 'spot the ball' but the idea is the same. Look at the picture below and see if you can spot the HF antenna. Clue: Look for the 'x's.
Posted by Administrator
Posted by Administrator
A bit silly perhaps, but the thing of interest is that this building is the Admiralty Building as seen from Horseguards Parade in Central London and that those really are HF antennas (see the expanded picture below). Whilst it's easy to dismiss good old fashioned short wave communications as outdated, especially in the age of satellites and mobile phones, it's very satisfying to see that those who need some assuredness of communication think it worthwhile to defile historic buildings in the centre of an area of tranquility and beauty with whopping great big, ugly HF aerials.
It doesn't necessarily follow, but it might be fair to presume that the people who did this (let's call them the 'military' for want of a better word) will be keen to ensure continued, low-interference access, to HF spectrum for some time to come. Which has to be good news for short-wave listeners and radio amateurs alike.
Tuesday 31 July, 2007, 15:26 - Radio RandomnessAnyone who really knows their stuff when it comes to radio propagation and stuff like that will tell you that any aerial (or an antenna for that matter) has a certain capture area.
Its capture area defines, in effect, how much of the signal that is in the ether it can capture and thus present to an attached receiver. The bigger the aerial, the greater its capture area and the stronger the received signal. Obvious really.
For some antennas, such as satellite dishes, the capture area is relatively obvious. A satellite dish which covers an area of 1 square metre has a capture area of roughly 1 square metre (ignoring edge effects). For linear aerials, however, the capture area is less obvious. Which has the greatest gain - a 4 element yagi or a 4 wavelength long colinear? Not so easy is it. However, it seemed to me that there ought to be a simple rule-of-thumb which allowed simple comparison between antennas. I therefore hypothesise that:
* The gain of a linear aerial ought to be somehow related to the amount (length) of metal it represents. Thus if you add up the length of the elements on a yagi and stretch them out into a colinear, the two should have the same gain.
* A doubling of the amount (length) of metal should double the overall gain (i.e. increase it by 3dB).
For now, I'll leave my first hypothesis and concentrate on trying to test the second one. Taking the quoted gain figures for over 20 amateur radio colinear antennas I've plotted their length in multiples of half a wavelength against their gain (in dBi). The length is done logarithmically, so '1' represents a length of half a wavelength, '2' represents one wavelength, '3' represents 2 wavelengths, '4' represents 4 wavelengths, '5' represents 8 wavelengths and so on, doubling in length for each increase in the index. Some of the antennas are multi-band (and thus might represent a compromise of gain balanced across more than one band), some are single-band.
The dotted line represents my hypothesis - i.e. that the gain rises by 3dB for each doubling of length. The solid line represents a 'best fit' line. As the antenna gets long (in terms of wavelengths) there is a noticable drop in gain with respect to my hypothesis. It is probable that as the antennas become long, the losses in the 'metal' itself cause a fall away from the my theoretical figure. Alternatively my hypothesis is incorrect. Either way, the original premise isn't that far off!
Next job is to see whether my hypothesis about the total length of elements of a yagi and the length of 'wire' in a a colinear holds any water.
Monday 18 June, 2007, 15:31 - Radio RandomnessIn a previous entry, I discussed the massive security hole presented to their neighbours like a baboon's bottom, by those still using analogue cordless telephones, as they can be easily received with cheap radio scanners, over quite large distances. However this is as nothing compared to the relatively common practise of bugging one's own house; otherwise known as 'installing a baby monitor'. The majority of these low power (10 mW) transmitters operate using basic analogue FM modulation on frequencies between 49.820 and 49.980 MHz (a low-power, short-range, licence-exempt band in the UK). These devices, like their cordless phone counterparts, can be picked up over several hundred metres, if not further. And whilst the owners often switch off the receivers when their child isn't in range of the monitor, they rarely switch off the transmitter meaning that it's often possible to tune-in to your neighbours going ons all day long (though such activities are strictly illegal in the UK and should not be entered into).
Around the wireless waffle HQ, there are several such baby monitors clearly audible on frequencies of 49.830, 49.840, 49.890, 49.930, 49.940, 49.950 and 49.962 MHz (the latter possibly intending to be on 49.960 MHz but is off-tune). There are also carriers on several other frequencies in this range but which are too lost in noise and interference (caused by other transmitters on the same frequency) to clearly make out. Some devices just produce a steady carrier, modulated with audio, others transmit data too, either as a 'warble' every second or so, or as a continous 'chuff-chuff-chuff-chuff' type noise. The latter types typically revert to being audio transmitters once the microphones detect any sound.
As well as allowing anyone with a cheap receiver to tune in to your private moments, many of these devices are poorly designed or built and have the capacity to cause significant amounts of interference to nearby radio frequencies, in particular the 6 metre (50 MHz) amateur band. It was as a result of such interference that my attention was drawn to the use of these 49 MHz frequencies in the first place, as reception from around 50.000 to 50.200 MHz suffers from out-of-band emissions from these devices (especially the warble and chuff-chuff-chuff-chuff models). What effect a 200 Watt SSB transmission on 50.150 MHz has on reception on neighbouring devices, I have no idea but it's to be hoped the receivers are as bad as the transmitters and that mummy and daddy are startled to find their 6 month old baby calling 'CQ'. It seems that the power supplies used for these transmitters are often badly regulated or smoothed meaning that there's broadband 50 Hz powerline noise, or worse, switch-mode noise emitted along with the intended transmissions.
There are some newer digital baby monitors available which operate either in the 900 MHz range (US models only, not licenseable in the UK), the 2.4 GHz and 5.8 GHz range. These models are virtually impossible to eavesdrop upon as they use digital modulation and are usually spread-spectrum. That being said, they're not encrypted so I guess it would be possible for some enterprising brainbox to figure out how to listen in, but that's hardly a Sunday afternoon activity (an activity which we do not condone, as it demonstrates utter contempt for people's privacy and is, at least conceptually, even more illegal than listening to the analogue ones). If it were me with a young child, I'd opt for a digital system safe in the knowledge that (a) it offers better features, (b) it's much more difficult to intercept and (c) it causes less noxious emissions that have the capacity to cause severe damage a young humans brain and body tissue. This latter point is exceptionally important, or completely made up, I'm not telling: you decide!
Friday 27 April, 2007, 14:51 - Radio RandomnessAnother train journey, another chance to run good ole Netstumbler and do a survey of channel occupancy for 2.4 GHz (that's 802.11b, g and n and not 802.11a in case you were wondering) to see whether my previous analysis of which are the best WiFi channels to use still holds.
For those who haven't (or can't be bothered to) read my previous article, I came to the conclusion that if you lived in an area of high WiFi penetration, channel 1 was the best channel to use as it was the least likely to suffer interference from other Wireless LAN users. In areas where there was unlikely to be any other wireless LAN activity, channel 11 (or 12, or 13) would be best, as these are the most free from other interferers (e.g. the military, microwave ovens, radio amateurs and so forth).
So what are the results of this train journey? I've plotted them above. I've shown the outbound journey separate from my return journey. As it's highly possible that if I picked up a LAN in one direction, I might have equally picked it up in the other, I've filtered the return numbers to take account of this. Also, I kind of half forgot to switch my system on on the outbound journey so, as you can see, the results for the return journey show many more LAN's than the outbound!
The upshot remains exactly the same as before (phew!) Channel 1 continues to be the best channel to use if you are in an area saturated with other users. Remember when looking at the above graph that channels 2 to 5 interfere with channel 1 and as such are not independent - equally they interfere wich channel 6 - only channels 1, 6 and 11 (or 1, 7 and 13) are actually free from interference from each other. My arguments about channel 11, 12 or 13 being the best to use in quiet areas remain unchallenged.
As a postscript, I though you might enjoy one or two of the network SSID's (names) that I found during my journey. Here are my favourites:
'GARY BARLOW' (was it really...?!)
'Ideal Cleaning Wireless'
and my absolute favourite: 'FRAUDULENT'...! Also, a few other vaguely interesting facts and figures:
Number of networks called 'BTVOYAGER': 12
Number of networks called 'BTHomeHub': 45
Number of networks called 'SKYxxxxx' (where xxxxx is a 5 digit number): 30
Number of networks called 'Belkin54g': 10 (and 7 of them were open)
Number of networks called 'default': 5 (all of them open)
Number of networks called 'linksys': 10 (4 of them were open)
Number of networks called 'NETGEAR': 16 (10 of them were open)