Let’s use MURS!
There is a band allocation available in the USA that doesn’t get the attention it deserves. I’m talking about the MURS band. Five narrowband spot frequencies around 151-154MHz (yes: VHF) where up to two watts of transmit power are allowed.
For those of you brought up on 2.4GHz, that means “range” with a capital “R”.
Now VHF narrowband radios are pretty different to the ‘single chip integrated RF section plus big processor’ setup familiar to a Bluetooth or Zigbee implementer. Most MURS radios use double conversion super heterodyne architectures. Selectivity comes from dedicated narrow (crystal or ceramic) filters. Channel frequency is derived from a discrete component low noise VCO locked to a highly stable TCXO reference by a PLL. There are a lot more parts in the line-up.
On the other hand, the interface is likely to be quite simple, as there is no use for a complex MURS radio interface layer (with high data overheads and processing power demands) when the data rate rarely exceeds 5kbit/s. Generally you see either a simple serial port, with some kind of elementary coding and packet formatting, or the user connects directly to the module’s “audio” interface, and handles the coding/decoding themselves.
They also tend to be quite large (compared to a single chip) and comparatively expensive, since the parts needed to build a good narrowband radio come with a price-tag
Receive filters require narrow bandwidth and good shape factor and stop-band rejection.
A high sensitivity (20-30dB more sensitive than a Zigbee receiver) radio calls for proportionately better selectivity, and that calls for crystal and high-performance ceramic filters. Frequency accuracy is also proportional to the width of the occupied channel. A MURS radio needs 5ppm centre frequency accuracy (by the conditions of the FCC spec.) so while a wideband unit can use a low spec, inexpensive crystal, our narrowband unit needs a TCXO.
(For reference, a typical MURS transceiver capable of putting out 500mW RF power, such as the SHX1, fits into a 67x30x9mm footprint, for under $50. There is a built-in 1200 baud data modem, or access to the baseband input and output signal paths. The whole thing runs from a +5v rail)
So, having identified a class of radios with almost polar opposite characteristics to the more familiar 2.45GHz radios, where can they be used, and for what ?
In these VHF bands only narrowband radios are permitted. At these lower frequencies there is simply less bandwidth to be had. The spectrum is already very crowded with a variety of existing users and a wide modulation bandwidth (requiring MHz of spectrum) could not be permitted. (Even 25KHz channels are now seen as unacceptably wide, and 12.5KHz is the norm.)
Compare the 80MHz width of the 2.45GHz band with the miserly five MURS channels
Narrowband VHF techniques do not replace the now-familiar high data rate wireless network devices, but they do complement them. There are applications where the inherently long-range (low path loss, high sensitivity) VHF radio can still show considerable advantages in long distance low power consumption, low data throughput tasks.
But when a 10mW transmitter must reach out to 1000 yards, nothing but VHF will do.
If there was not a tangible advantage in the use of the VHF bands, there would be no point in even looking at a MURS radio. Why use a more expensive, bigger MURS radio to send your data at a much slower bit-rate? The answer is the path loss calculation, and the range.
But before we get into this subject, we need to take notice of a warning:
The actual, achievable, range of a wireless data link can be a highly unpredictable thing.
In theory, and in un-obstructed free space, you can meaningfully calculate a path-loss, relate that to the performance of the radio components, and get an absolute range. But not in the real world. Firstly, the actual propagation will be nothing even vaguely like the free space model (compare the equations in notes 1 and 2). The presence of obstructions, the curvature of the earth limiting the line of sight “horizon”, the proximity of the ground itself (causing diffraction losses) all mess up the propagation of your radio waves.
A wireless link will function if the link budget is greater than the path loss.
Link budget = transmitter power – receiver sensitivity – margin
(‘margin’ is an un-apologetic fudge-factor. It takes into account random degradations, such as fading and moving obstacles. 10 to 20dB is usual)
Path loss can be measured, or estimated from a propagation model (there are a great many path loss models in existence. I’ve quoted the “Egli model”, which is reasonably reliable for VHF and UHF propagation through cluttered rural and suburban environments. In an industrial jungle of steel framed buildings no model can be relied on)
At this point I’m going to cheat. It gets pretty theoretical for a few paragraphs after here, so I’m going to pretend we all have previously read and understood what’s coming, and I’m going to pull the most important bit out ahead of time:
Path loss is related directly to frequency (actually proportional to 10log(1/f^2). )
This means that (even if aerial gain, transmit power, and rx sensitivity were the same, and they really aren’t) 500mW at VHF (154MHz, MURS), 4 watts at UHF (450MHz), 12 watts at 915MHz, and 100 watts at 2.4GHz would exhibit similar ranges.
But receiver sensitivity (hence range) is also related to channel bandwidth (and hence data rate). Each doubling of the signal bandwidth drops the link signal-to-noise (S/N) ratio and hence sensitivity by 3dB. So, while a narrowband 12.5KHz unit might have a sensitivity of -118dBm (at 2.5Kbit/s), a comparable wideband unit might achieve only -107dBm at 64kbit/sec (and will require a 600KHz wide channel).
So higher speed links have even shorter ranges for the same transmit power.
This is why your domestic WiFi gives you coverage through your house (and maybe into the garden if you’re lucky), but a decent VHF link can provide coverage across your whole town
Am I getting through here?
A simple MURS radio, pumping out a watt or so into any halfway decent aerial, is going to have an operating radius that you measure in miles. Give it an optimal path (sea, lakes, flat prarie) and you could easily reach to the visible horizon!
Let’s go back to the classroom
For all you maths addicts out there, here is the Egli irregular terrain path loss model, expressed in dB terms : (remember it refers to a path gain, so the answer is always negative)
(J. J. Egli, “Radio Propagation Above 40 Mc Over Irregular Terrain,” Proc. IRE, Oct. 1957)
path gain (dB) = 32.4 - 40 x log(d) - 20 x log(F) + 20 x log ((Hr x Ht)) + Gt + Gr
F = frequency in MHz d = distance in meters
Gt, Gr = transmit and receive antenna gain (dBi)
Ht, Hr =height above ground of transmit and receive aerials (in meters)
Phew !
Lets put some numbers in: Take an imagined 154MHz radio link with a 125dB link budget (-115dBm rx, +10dBm tx). Egli’s model gives a range at about 2500 feet (assuming 0dBm aerials at 1m elevation), which is typical for practical links in this performance category.
That 10mW radio only has a 1.5 x 1 inch footprint
Fortunately , two useful observations can be drawn quickly from Egli’s model:
- Path loss (and so the amount of transmit power needed to achieve a given range) is proportional to channel frequency
- It takes a 12dB improvement in link budget to double the operating range (which ties in well with an empirical rule of thumb for VHF radio: multiply transmit power by ten and you double the range)
Want to know more about MURS?