A different kind of question ;)

[qtommer]

this is not so much a technical question but it is a question of experience as an engineer.

in developing a project, for example say a precision parking robot incorporating the use of an ADC from an analog sensor and the control of motors using PWM, the scope of analysis would be:

1)the resolution of the ADC-how much resolution really is needed for the specified precision?
2) the timing analysis between the reception of an input to the actuation of the motors at the output to carry out the precise parking whilst avoiding accidents.
3) the usual power consumption calculations
4) the actuation of PWM control- the manipulation of PWM to acquire the required actuation of the motors..

and etc.......


Now let's look at a totally different field. say the design of a data communication link between a GSM Module and a microcontroller. Besides, the rate at which the communication is conducted, the hardware involved for the required communication, what are the aspects of analysis that should be performed to show the feasibility of the project?

to be honest, im doing this analysis for a dissertation and would really appreciate the advice of the "seasoned" engineers out there in this matter.

Many thanks=)

 

[Diver300]

The data communication between a GSM module and a microcontroller is usually just a receive line and a transmit line. There might be a couple of handshaking lines if you wanted. It is very much as the GSM module manufacturer demand.

The GSM communication between the module and the GSM network depends on access to the GSM network, both in terms of signal and having paid for access to the network. Obviously you need to know where this will be used.

In terms of feasibility, there are millions of GSM modules communicating with microcontrollers already, so I don't see what else you need to look at. Data rate and the cost of sending data over the mobile network are the things that could make a project viable or not.

 

[qtommer]

The data communication between a GSM module and a microcontroller is usually just a receive line and a transmit line. There might be a couple of handshaking lines if you wanted. It is very much as the GSM module manufacturer demand.

The GSM communication between the module and the GSM network depends on access to the GSM network, both in terms of signal and having paid for access to the network. Obviously you need to know where this will be used.

In terms of feasibility, there are millions of GSM modules communicating with microcontrollers already, so I don't see what else you need to look at. Data rate and the cost of sending data over the mobile network are the things that could make a project viable or not.

thank you for your reply=) basically I have actually already developed this system and it's definitely a project that can be done as you have said...
I'm trying to provide the "Engineering Analysis" to my system now...I have considered the data rate of the GSM module and the MCU and from there implemented the required programming and circuitry...After I have done that, it seems that nothing else can really be done to analyze the data communication system. Besides data rate, and the encoding methodology and protocol , are there any other aspects in which I can evaluate data communication such as this?

thanks=)

 

[Diver300]

I think that it is far too easy to waste a lot of time on something like that. The GSM network and the GSM modules are there and they work. If you start to analyse them in a report, you are only going to get managers start asking themselves why you think it is necessary to ask the questions.

I assume that you have to use mobile communications of some sort, so you really don't have a choice what to use. You are using the network and modules as they are supposed to be used, so they are likely to work. If you are not careful, your analysis can get mission creep and you're going to try to justify 1% resistors, every single decoupling capacitor and the use of FR4 circuit board. If that happens, your report will be 10,000 pages and the project won't go anywhere.

In my experience with vehicle tracking devices, what matters is the protocol's ability to recover from data loss without the user seeing it.

Also the mobile network prices or contracts are what could destroy a project like that. In vehicle tracking, a very small amounts of data are sent by GPRS, so the minimum cost of a connection is the most important thing. Vodafone started by charging 1p, then they waived the minimum and let the data amount get up to 10 kB before taking 1p, then they "reduced" the price to 50 p per day. In fact, we found that the connections were 8p a time with a 50p per day limit.

The point is that they changed the prices by over 10 times, which is easily enough to make a project viable or not.

Different counties have different rules. PAYG in America has a $100 per year minimum.

What does is the project to do?