In the scope of an international deployment for remote monitoring of machines, we would like to use USSD as transmission channel. The reasons are: it's fast, it's for free and it requires only little transmission power and network coverage. My questions:
- Do you recommend using USSD for an international deployment? What are the pros and cons?
- How reliable is USSD?
- Do you have a reference of such deployments?
- Is USSD standardized and supported globally by every GSM operator? Can I expect that this works in every country, where my operator has GSM roaming deals (I want to rely on a single operator)?
- Is USSD supported in non-GSM standards?
- Is it possible that some operators block USSD because they want to prevent free usage of data traffic on their network?
- What do I need from my operator in order to use USSD (access to USSD server, short-number, etc.)?
- How can server-initiated communication be done using USSD?
- Can I be sure that my operator will not charge anything for this roaming traffic even if at some point in the future I will have hundreds or thousands of machines roaming outside my operator's network free of charge?
Oli - even if the technical fit is there, i would strongly urge you to reconsider basing your business on trying to cheat the operators out of revenue. And that is how they will view it. If they find out, they will shut you down immediately, with no warning. There's a famous case study out of there of an MVNO-type company in the U.S., who shall remain nameless, who tried to use the control channel for free and were successful for a while...until the operator caught them and threated to shut them down unless they paid multiple millions of $$. By that point their entire business model depended on this network access and they had no choice. Now they are barely in business.
USSD was created for the operators to have a back-channel to the phone. To have a third-party use it for core application data is an un-natural act. And those rarely have long-term viability.
This is an interesting question, and while I agree the carriers should frown upon this exploitation it does bring up the question: "Why is SMS so exorbitantly priced?" SMS, like USSD, uses the Control Channel. Upto a level of activity requiring the addition of new resources, the cost to operators should be close to zero. In a competitive market the price should approach the marginal cost. Does anybody think there is collusion among the carriers to artificially keep SMS prices so high? Obviously, they are salivating at the explosion in social networking services such as Twitter and view SMS as a gold mine. At the same time, many Telemetry applications require only short, infrequent bursts of data--a perfect match for SMS (anybody remember Cellemetry, which did much of this with only 32 bits of information?)
If SMS was FREE, or nearly free (as it should be in a competitive market), I think we would see huge growth in M2M and the Internet of Things. Many applications simply do not require GPRS / Data connections, and with SMS BOM costs can be reduced by using cheaper microprocessors without the need to host TCP/IP stacks.
"many Telemetry applications require only short, infrequent bursts of data--a perfect match for SMS (anybody remember Cellemetry, which did much of this with only 32 bits of information?)"
You may not have intended to imply that Cellemetry is past tense - it is definitely not. Numerex converted Cellemetry from AMPS to digital a couple of years ago, and has expanded their various networks, including satellite services, nearly world wide.
I recall there was an Analog Sunset a couple of years back, in which carriers were no longer required to suppor the older analog technologies.
My assumption regarding Cellemetry was that it would simply be converted to SMS, gaining the benefit of a much larger payload and more universal interoperability with both carriers and devices.
How is the digital Cellemetry different from SMS, then?
I am far from being an expert in these matters, but have followed Numerex for a very long time. Here is a link to a white paper on the NMRX web site that, although written several years ago, seems to describe the migration of Cellemetry from AMPS to both CDMA and SMS services. I hope this answers your question.
I noticed that Syniverse was mentioned in a message below. Here's a recent press release from them that describes their relationship with Numerex
While carriers seem to want to reserve USSD for themselves, at least for now, there does seem to be a valid/enduring business in writing USSD applications for the carriers. "White-label" operator interactive service applications include replenishment, self-care, notification including personalized advertising, LBS, voting, callback including offers from major players including Syniverse and AMDOCS
http://www.syniverse.com/business-solutions/solutions/USSD-Callback-Serviceand now social networking
There also seem to be rogue callback services using USSD (for rogue carriers as well as rougue individuals). Apparently USSD is tough for carriers to block http://www.apexvoice.com/index.php/APEX-Blog/ussd-callback.html as most carriers leave USSD ports open for roaming traffic.
Here's a Uruguayan telecom software developer, Leib ICT, that seems to have significant USSD expertise as well as infrastructure applications including a USSD developer toolkit/gateway simulator
Leib ICT's list of USSD gateway customers includes Movistar Colombia
More details on the pros, cons and mechanics of USSD, including how a USSD gateway can use USSR requests to initiate a session to the mobile terminal, here
While USSD is only available in GSM networks, here's an attempt to provide USSD-like interactive services for CDMA carriers http://www.cboss.ru/products/cbossussdLike.html?locale=en
Hope it helps,
My understanding is that USSD is not guaranteed to be delivered, especially in roaming scenarios. That solely depends on your mobile network operator. If you do not have an agreement with the operator, he might firewall away your USSD messages, especially if he detects high volumes circumventing his pay-messaging services.
Good point Rob, and here's an industry expert's blog post from 2009 backing it up:
"If you want to deploy a USSD application, then you need to be a mobile operator, or you need to do it through a connection to a mobile operator's USSD gateway. You can't just use a GSM modem. You can't just go to a bulk SMS provider and get a connection...SMPP is very widely used as an interface into USSD systems."
The article also backs up Morgan's point higher in this thread:
"So, when we get asked questions about USSD support, sometimes it is a technical curiosity question ... people looking for potentially cheaper alternatives to SMS. In those cases, generally, it is not worth pursuing."
Hope it helps,
Here we are!
We enable businesses to use USSD globally. Roaming free options are possible.
International businesses of any size can insource mobile data as an application for the services or hardware.
See more: networkservice.biz or contact me.
We have been successfully building and utilizing USSD in a variety of devices including GPS tracker, lake level flood warning system, storage tank level monitoring and weather monitoring stations.
We can send data alphanumeric data by USSD from any of 220 countries on 550+ networks to a user defined URL via our own gateway and Global SIM.
Our Tracker was used in the London Marathon on a runner. The tracker sent positions using GPRS, SMS and USSD. USSD out performed GPRS and SMS in terms of the number of positions received and latency.
Our service is available for use in applications today.
And I can confirm tha USSD is working very well. 400 toiletts in the Netherlands can be opened by a call to a service center. The service center will charge Euro 0,60 from your account and will send a USSD back to the toilet door. A USSD is traveling estimated 1 second. 1 second later the door is open. So fast you can´t open your wallet and take out the coin.
BTW, the GSM/GPS tracker of Ed was developed by a business friend of mine in Germany. I was talking care on the design in of the embedded GSM and GPS antennas. More about antennes here:
The remote access to the toilet was based on embedded GSM antennas as well:
Who is interested in a M2M design based on GSM, UMTS, HSPS or LTE is welcome to email to harald.naumann (at) gsm-modem.de
Haralds Antenna in our GPS tracker has proved to be excellent in all tests.
We have tested the tracker in a number of situations that would normally interfere with mobile and GPS signals and it has excelled in all situations. The tracker maintained tracking and data delivery whilst in "horrible" locations such as vehicle glove boxes, in the boot of cars under luggage and ladies hand bags!
The robust design of the antenna and tracker allowed tracking to be maintained in all instances, even inside houses.....although the GPS position accuracy and repeatability did deteriorate.
We also proved the reliability of USSD as a data/position delivery method where a GSM GPRS signal was weak or not available on a Yacht in Greece.
In the Netherlands, a new billing system for public toilets which is based on mobile wireless technology has been developed by “NeoNumus”. The costs incurred for provision of the facilities, cleaning and maintenance are paid quickly and simply by the customer by “mobile phone” via a cashless system. This system reduces the waiting time for service and maintenance and improves protection against break-ins.
Most people take their Smartphone with them wherever they go. What could be easier than using your phone to open a door? The concept was designed so that even ancient mobile phone models can do this. USSD is used rather than SMS. A number of services in the GSM network are based on USSD. The call, or the request to make a connection, is already a USSD in itself. The round trip for the USSD out to the server and back takes around 2 seconds. A text message needs 6 to 10 seconds in one direction only. However, sometimes it can even take anywhere between 6 minutes to 6 days, or the SMS is not delivered at all. With USSD, the acknowledgement arrives immediately as the door is opened. The service – the opening of the door – must also be billed. The USSD is sent by the network operator in two directions to a “NeoNumus” server. This means that the SIM card must be billed somehow. There is also further access to other servers for billing of the service via the network operator. Inexpensive GSM modules are employed, however, inexpensive in this case was not quite cheap enough. The most cost-effective GSM module was still way too costly. The costs of the module have been balanced out by an intelligent installation concept which uses an integrated antenna. 18 years ago, the author was a remote vending project manager for a service provider of vending machines. This company covered everything, from filling the vending machines to their installation. Back then, it was established that over 90% of machines could be equipped with wireless technology and this was also the case when the antennae were built into the vending machines. Since then, the availability of wireless has improved even further. If the antennae need to stay inside the machines, there is no need to drill any holes. This practical experience has undergone even further refinement.
The GSM antenna, the GSM module and the remaining electronic components are located on the same PCB. Standard enclosures generally have four domes with a ***** in each corner. These enclosures have now been optimized to just three screws below the antenna. This eliminates the influence of the screws on the antenna (even plastic screws can affect the antenna). In addition, the antenna ground plane is adapted to the required physical size at 45 mm x 90 mm. For systems without a power supply, an antenna toggle switch to the external antenna socket is available. The scan operation of the GSM module is launched twice when the PCB is first put into service. Once with an external magnetic antenna installed on a large metal plate, and a second time with the internal antenna on the PCB. The result: the internal antenna was able to receive more base stations than the external model. This concept saves on external antennae and significantly reduces installation costs. The savings versus the costs for the GSM module are neutral. Besides, the PCBs were developed for vending machines, not for toilets. Standardized interfaces can help with the use of these PCBs for vending applications.
(Origin story and further M2M stories on page 8 @ http://www.m2m-alliance.com/fileadmin/user_upload/pdf/2013/M2M_Journal-19_EN.pdf)
|Are you a mobile operator interested in being featured here? Find out how.|