What Can You Do With 50 bits per second

Some things go up and some things go down.

Force sizes are definitely going down, as we know only too well.

Demand for sensor coverage is going up, in all theatres the need to have a ubiquitous and persistent network on which to overlay communications and sensor feeds is growing at an alarming rate.

Despite the rapid increases in wireless bandwidth, compression and associated technologies an interesting new development is taking the opposite approach.

Ultra Link Processing concentrates on range and persistence, eschewing high bandwidth for low power usage and the ability to operate for extreme time intervals with a minute amount of power. Bandwidth is tiny, about 50 bits per second, but the power management allows devices to remain in use for 20 years. Onramp Wireless are targeting smart energy grids, with a range of nearly 50 miles and ability to penetrate underground it is hoped that it will support transmission of low bandwidth data such as meter readings.

A trial network in San Diego has demonstrated that just 35 base stations can collect data from a 4,000 square mile area.

 

These are the key system highlights, which set this technology apart:

  • Greater than 40 miles (65 km) line-of-sight range, and up to 10 miles (16 km) to underground locations.
  • RPMA, a unique multiple access method with order of magnitude capacity improvements, and demodulation of up to a thousand incoming signals, below the noise floor, every frame.
  • Each Access Point supports up to 64,000 endpoints.
  • Access Point aggregate throughput of 120 kbps (10 Mb/day).
  • Each ULP Gateway can support hundreds of Access Points.
  • Ultra low power consumption, allowing up to a 20-year battery-life for water meter applications.

From the On Ramp website

The ULP system provides a greater than 600x coverage advantage and a 25-40x capacity advantage over competing systems, but in many deployment scenarios, the advantage is substantially larger, particularly due to ULP’s ability to be deployed in highly elevated locations. This is not the case for Carrier Sense Multiple Access (CSMA) based systems, such as 802.15.4, 802.11, and 900MHz FHSS networks. CSMA assumes that endpoints in the network have a radio link between each other to detect contention. As coverage increases, the likelihood of that being true decreases, and the multiple access scheme collapses. Additionally, due to the short range of these systems, there typically aren’t enough favorable locations to place gateways, making this approach theoretical at best.

Just thinking out loud, how about the following;

  • Container tracking, no sniggering at the back please. The tracking of the whereabouts of container based and sub unit loads is a significant challenge and tends to get really difficult once beyond the secure and bandwidth rich rear locations. Combine the ULP with a GPS device and across a large area you could on demand, interrogate the whereabouts of all equipment, spares packs and other supplies
  • Movement sensors, one of the most significant challenges in the counter IED space is we simply do not have enough people to monitor vulnerable locations, attaching a simple movement sensor to one of these will allow us to flood an area with motion detectors. When a movement is detected it could cue another sensor to either eliminate it as a false positive (farm trucks, animals etc) or task other assets with taking a closer look and no doubt false alarm rejection software could be developed
  • Smart energy grids, every single generator in Afghanistan needs fuel and this fuel has to be driven in at great cost and danger. Using a low bandwidth but long range and persistent network would allow all generators and meters to be networked together. Remote management of generators operating parameters, making sure generator sizing is appropriate (large generators being run at below optimal output for example) and remote fault diagnosis could reduce energy usage, a valuable financial and operational result
  • CBRN agent detection network, instead of deploying vehicle based sensing and analysis a network of sensors could be deployed and monitored remotely. A centralised picture of chemical agent plumes and other contaminants could be rapidly analysed and disseminated for early warning purposes
  • Perhaps it might even be used for so called ‘blue force tracking’ of individuals and vehicles without connectivity to BOWMAN

Anyone think of any others?

9 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Michael (Civ.)
Michael (Civ.)
June 30, 2011 7:22 pm

Could it be used for automatic or faster reporting of casualties to medical fast response personnel/bases, especially if you’re there as a long term/occupation force?

RW
RW
June 30, 2011 7:50 pm

If I’ve understood this correctly they have used a very slow (long number of) cycle(s) to encode a digital signal on a carrier wave, having surrendered bandwidth they acquire very high likelihood of correctly interpreting the signal.

Then they dramatically reduce the packet size (against UDP TCP) via their own protocol to rebalance, to some degree, the loss not sure how they will progress with a protocol that isn’t a standard

@TD
Re distributed sensors and data transmission I have worked on an alternative idea which looks to use RFID tags for the physical layer, thereafter the data is parsed into XML and this is sent by ordinary networks to the assigned receivers.
Same general idea but the advantage is you can monitor and rewrite RFID tags and XML is a universal language, I think your idea of possible uses is good but I think you don’t have to invest in as much GPS technology as you think. Doesn’t matter exactly where something is (it does matter roughly how far past RFID readers, it has travelled) So map GPS to RFID and reduce costs.

Also you are right that this type of approach is suitable for smart energy grids where the aim is I think described as “distribution pricing” – work is being done on this in various places though I think there has been some notable work done in and around Bristol.

This ULP network does have value but I would add it to a range of networks rather than give it a monopoly.

S O
S O
June 30, 2011 8:49 pm

Reading about such a tiny bandwidth, I immediately thought you’d write about ELF radio for subs…

Rupert Fiennes
Rupert Fiennes
June 30, 2011 8:57 pm

Probably being paranoid, but at the data rates specified any form of PKI exchange would be impractical. That being said, a one-way hash should be OK provided the hardware was reasonably “FIPS” compliant. Just key management to worry about.

I would say blue force tracking would not be practical as the data rate is too low. Given the transmissions would be easily intercepted, it’s probably best suited to counterinsurgency: movement sensors sounds like a practical idea. That being said, the profusion of 3G/GPRS networks even in conflict zones, you wonder why anyone would want to roll this out when the latter was available.

I think the key management is what would kill this for military uses….

ArmChairCivvy
ArmChairCivvy
June 30, 2011 8:58 pm

Hi RW,

RE “Re distributed sensors and data transmission I have worked on an alternative idea which looks to use RFID tags for the physical layer, thereafter the data is parsed into XML and this is sent by ordinary networks to the assigned receivers.
Same general idea but the advantage is you can monitor and rewrite RFID tags and XML is a universal language, I think your idea of possible uses is good but I think you don’t have to invest in as much GPS technology as you think”
– I’ll write a comment on this, about how you can implement security by tracking thousands of assets and people, over a fairly large area

Hi S O,

I thought exactly on the same lines (even wrote a comment, but the ether eat it)

ArmChairCivvy
ArmChairCivvy
July 1, 2011 9:45 am

As promised:

Last night I thought I better not try to fit all this into the comment window, so drafted a few further thoughts on TD’s two points below, from his broader list. Having a low power, high density, low bandwidth network cover a sprawling area, like Camp Bastion, and combining it with another radio frequency technology, RFID tagging and readers, could be quite useful.

Container tracking, no sniggering at the back please. The tracking of the whereabouts of container based and sub unit loads is a significant challenge and tends to get really difficult once beyond the secure and bandwidth rich rear locations. Combine the ULP with a GPS device and across a large area you could on demand, interrogate the whereabouts of all equipment, spares packs and other supplies

Perhaps it might even be used for so called ‘blue force tracking’ of individuals and vehicles without connectivity to BOWMAN

Such an RFID (Radio Frequency ID) tag wirelessly sends bits of data whenever it is triggered by a reader. In addition to such active tags that have a battery, you can mass produce passive tags very cheaply. i.e any type of consumable (we are not talking bar codes here) can be tagged without undue cost The chosen frequency determines the range: from a few cm to 10m. So a container on top of another one; no problem. 500 items in the container: no problem. Part depleted: contents info automatically updated. High value items within the 500: no problem, put an active tag on them and no need to open the container to do checks.

You would also tag the operatives (embed in mobile or a wristwatch) This is part of the answer to the low bandwidth being OK: all static data (contents or identity and tasking, for instance, can be kept on a server and only actions are transmitted over the limited bandwidth. Whether the action is simply moving, or perhaps “consuming “ an item… so you can track at item level, and they all still aggregate, in real time, to a container load, a shipment, a separately secured storage area… whatever.

So let’s look at not just the items stacked up somewhere in Camp Bastion, but the whole chain.

The operatives can be identified as well as their tasking; in addition they can always be located. A duplicate in two different places raises an automatic alert, as the mobile and the wrist watch should be on the same person. Not all need both; they just have batteries inbuilt!

Add movement sensors and when there is no reading, then potentially you have a security breach (or a stray dog!).

As the area is extensive and you have to build in redundancy anyway (so that any incoming will not knock out the system as it will be well distributed), you would put in filtering servers. Trivial actions and location tracking would typically stay at this level and only true transactions (impacting on inventory) would be passed on to the Integrating Server, still physically in Camp Bastion. No form filling when returning for tea as all transactions would also be confirmed as recorded to the operative and his tasking superior (bi-directiional traffic, but very few bytes per message).

The next levels would be the logistics systems (the ones attempted several times, with a cumulative cost of £ 800m, I seem to remember) and even the (participating) suppliers’ business systems. For certain types of items you would mandate participation through the procurement contracts.

Quess who uses this sort of thing?

The Germans, of course.
Don’t know about their armed forces, but Lufthansa Cargo.

ArmChairCivvy
ArmChairCivvy
July 1, 2011 9:48 am

You can guess whether I have written to Bernard Gray; nothing heard back, so must be under control then

Paragon
Paragon
July 1, 2011 11:02 am

My immediate thought was unattended ground sensors, so definitely agree on that one. Presumably as this system involves emissions, it can be detected by the opposition? That would make it less suitable for operations against a technically advanced opponent, so I’m cautious about tagging your logistic chain in this way.

One thing I thought about was mines. If you have mines which emit in this way you can track incursions into minefields (as emitters shut down), and it would facilitate post-conflict cleanup. Of course detection by the opposition would be an issue (although possibly not, as any clearance attempts would be detectable). The emission could be selectively activated by a coded transmission, if friendlies were moving through the area, or at the end of hostilities.
With landmine bans and degradeable explosives this idea is probably fairly redundant though.

So, how about putting these into every vehicle or aircraft we export to the middle east. Maybe selectively activated on demand, as and when needed :-)

Dave
Dave
July 1, 2011 3:59 pm

Re Container tracking. You will be better of going with CONT-TRAK.
http://telecom.esa.int/telecom/www/object/index.cfm?fobjectid=28461