Developing a digital radio communications circuit design for PCB

[SanctePieDecime]

So, I am working on a project that will involve transmitting data with a range of about a half mile (and actually I pretty much just need to transmit 1 bit). I am trying to make it all one circuit board. Unfortunately, all the research I have done has been fruitless. All I can seem to find are pre-made modules (such as the HC-12 and NRF24). I would like to know and understand how to design a circuit for the ICs that I keep seeing on these boards so that I can just directly implement the IC in my circuit board without having to use the full module footprint, but I have not been able to find anything. Would anyone be able to point me in the direction of a resource that explains how to accomplish this?

[Andy Chee]

Need more information.

Half a mile through water? Air? What about terrain and obstructions?

Also what frequency band are you interested in? 3.5MHz? 433MHz? 2.4GHz?

Also what do you mean by 1 bit data?  I assume you just want a simple on and off function.  Which raises the question of data speed, how fast do you want to switch things on and off?

If you don't know the answers to any of these, best to stick with using pre-made modules.

[SiliconWizard]

Yes we need to know more about your requirements. First, is this a half mile line of sight, or can there be obstacles, etc.
As a very general thought, 2.4GHz RF is not ideal for long range like this unless seriously amplified. Those modules based on a nrf24l01+ can do it in the variant with an added power amplifier, no way they can without this external PA. And even so, they can do it outdoors without obstacles, but will have a hard time otherwise.

Given your low throughput requirements and the distance, I would, again as a general recommendation, suggest going for a sub-GHz solution. You have cheap ICs, such as CC1101 or CMT2300A, which should work here. Alternatively, you can go for LoRa, which is much more robust and can cover much longer distances yet. LoRa chips are a bit more expensive.

Either way, if you have no prior experience with RF stuff, I'd suggest sticking to pre-made modules rather than implement the RF stage yourself on PCB - it requires some routing care for properly tuning it and is probably not worth the trouble, especially if you're not going to build a lot of these. Modules can be had for a few bucks, guaranteed to work, and some will take very little space on another PCB (like 10x15mm, something like that.)

If you're curious, you can still find schematics for some of these modules and see for yourself how they are designed.

[SanctePieDecime]

My specific application is a mailbox sensor. There are ones that exist but they all tend to rely exclusively on when the door opens and closes and as such I am trying to develop a solution that won't be prone to unnecessary notifications or errors (for instance when you open the mailbox to put mail in).

Half a mile through water? Air? What about terrain and obstructions?

So it is just a half-mile through air (so that it could be applied in settings where there are particularly long driveways). I am not terribly worried about terrain but ideally it would not be hindered too much by trees (for instance in an application at a forested home).

Also what frequency band are you interested in? 3.5MHz? 433MHz? 2.4GHz?

The frequency band doesn't really matter to me as long as it gets the job done. I have not been able to find any resources for integrating any of them directly into a PCB.

As a very general thought, 2.4GHz RF is not ideal for long range like this unless seriously amplified. Those modules based on a nrf24l01+ can do it in the variant with an added power amplifier, no way they can without this external PA. And even so, they can do it outdoors without obstacles, but will have a hard time otherwise.

Given your low throughput requirements and the distance, I would, again as a general recommendation, suggest going for a sub-GHz solution. You have cheap ICs, such as CC1101 or CMT2300A, which should work here. Alternatively, you can go for LoRa, which is much more robust and can cover much longer distances yet. LoRa chips are a bit more expensive.

Thank you for recommending sub-GHz radio implementation. I should have put more consideration into this, but it makes sense so I am thinking that I would probably want a lower frequency.

Also what do you mean by 1 bit data?  I assume you just want a simple on and off function.  Which raises the question of data speed, how fast do you want to switch things on and off?

The one bit is just sending the notification to the base station that the mail has been delivered (which is determined through a combination of sensors, comparators, and logic gates). Data speed isn't terribly important, but it will hopefully be fast enough to be able to use the base station to create a log file on an SD card with proper date and time stamps. Then again I will be using a power latch circuit to shut everything down when not in use to conserve battery and as such the faster the transmission is then, theoretically, the more battery can be saved because it would not have to be on as long. But this doesn't really make too much of an impact when considering that the system will have to implement a delay allowing for the mailbox flag position to change after the lid closes in cases of putting mail in the mailbox.

Either way, if you have no prior experience with RF stuff, I'd suggest sticking to pre-made modules rather than implement the RF stage yourself on PCB - it requires some routing care for properly tuning it and is probably not worth the trouble, especially if you're not going to build a lot of these. Modules can be had for a few bucks, guaranteed to work, and some will take very little space on another PCB (like 10x15mm, something like that.)

As far as not using modules, this is in part so that I can learn how to apply the technology without relying on premade modules (which are in my opinion a little like training wheels). I will definitely start out my testing with premade modules, but I want to move away from them for the final design.

[SiliconWizard]

ebyte is one of the companies making such modules for relatively cheap. They have a web page there in which they describe the design of their SX1262 modules: https://www.ebyte.com/en/new-view-info.html?id=295
(The SX1262 from Semtech is one of the "best" LoRa transceiver ICs at the moment.)

Note that those modules require some microcontroller, they can't be used "standalone". Unless you already are familiar with some microcontroller you want to use, you can go for some Arduino compatible one, there are already Arduino libraries for the SX1262 (and other LoRa chips) and you'll find various blogs and videos about that.

[pcprogrammer]

Another option is to look into modules like these on Aliexpress. No need for a micro on the mailbox side. Just one or more switches connected to the module.

Should not be to hard to find what kind of IC's are used on these boards, and make your own PCB using them if wanted. But like some already wrote, with little experience on RF it is best to stick to the pre made modules.

On the receiver side use a simple MCU setup to log the received input on your SD card. In the mailbox you can put a switch on the door, the handle or on the bottom to sense mail lying in there.

[BennoG]

You also have to consider identification. If the neighbour has the same unit then you will be notified when he gets his mail and also when you get your mail.
without knowing what signal is from what mailbox.

Benno

[radiolistener]

in your case the most easy solution is just to buy some Chinese remote doorbell and connect it's button to your mailbox sensor.

Something like this:
https://www.aliexpress.com/item/1005004701032074.html

[CaptDon]

"Determined through a combination of sensors, logic gates......" Really, the sensing needs to be that complicated, not just a switch that causes the unit to transmit one signal burst? A very simple FM carrier modulated with perhaps a 1KHz tone is all you need for the transmitter. Now it gets complicated. The transmitter requires an antenna that can not be inside the steel mailbox but could perhaps be inside a plastic or fiberglass mailbox. Are you designing this for sale or for your own use?? That will force some hard choices. Also, 1/2 mile is not trivial!! You will most likely be limited to less than 100mw D.C. input power to the R.F. final stage unless the device obtains F.C.C. certification for some of the dedicated remote control frequencies. Also, do you know the part number for the I.C.'s used on the modules?? Why not just Google their datasheets? Based on the probable low transmit power you may require an outdoor directional YAGI antenna which at 2.4GHz are small and practical. I guess you are re-inventing the wheel here as opposed to buying an off the shelf solution. If it is a learning project there must be 100 better ways to learn about R.F. and signalling.

[tggzzz]

So, I am working on a project that will involve transmitting data with a range of about a half mile (and actually I pretty much just need to transmit 1 bit). I am trying to make it all one circuit board. Unfortunately, all the research I have done has been fruitless. All I can seem to find are pre-made modules (such as the HC-12 and NRF24). I would like to know and understand how to design a circuit for the ICs that I keep seeing on these boards so that I can just directly implement the IC in my circuit board without having to use the full module footprint, but I have not been able to find anything. Would anyone be able to point me in the direction of a resource that explains how to accomplish this?

The phrases you use indicate that you are probably a beginner.

Be aware that developing a radio transmitter is not something a beginner should be attempting. There are too many legal requirements to ensure your transmission doesn't screw up someone else with a receiver. Violate any, and you can't complain if neighbours become irate and you are visited by men in black suits.

OTOH, it is entirely reasonably to use a module that has already been certified for use in your locality. You will have to verify that any such modules will work over the relevant distances and with relevant power supplies. That ought to be within a beginner's capabilities, and is a very good way of starting to learn about RF.

[radiolistener]

Here is 1 km range intercom, which can be suitable for your needs:
https://www.aliexpress.com/item/1005006363485172.html

[temperance]

OTOH, it is entirely reasonably to use a module that has already been certified for use in your locality. You will have to verify that any such modules will work over the relevant distances and with relevant power supplies. That ought to be within a beginner's capabilities, and is a very good way of starting to learn about RF.

Indeed, and select a module with 433 or 868 or whichever freq. which is allowed in your area. 2.4 GHz isn't going to work over half a mile.

[BrianHG]

I've used these transmitters in the past, by passing the onboard IC, for some occasional serial communications.  They can achieve 1/2 mile without problems:

https://www.aliexpress.us/w/wholesale-433mhz-transmitter-4km.html?spm=a2g0o.productlist.search.0

If you cannot code, you can get them with a relay receiver module:
TX-RX-combo

[SanctePieDecime]

The phrases you use indicate that you are probably a beginner.

You bet. That is why this is in the "Beginners" forum.
If there are things that you can tell I would definitely need to know before I continue with this project please point me to some good resources.

I've used these transmitters in the past, by passing the onboard IC, for some occasional serial communications.  They can achieve 1/2 mile without problems:

https://www.aliexpress.us/w/wholesale-433mhz-transmitter-4km.html?spm=a2g0o.productlist.search.0

ebyte is one of the companies making such modules for relatively cheap. They have a web page there in which they describe the design of their SX1262 modules: https://www.ebyte.com/en/new-view-info.html?id=295
(The SX1262 from Semtech is one of the "best" LoRa transceiver ICs at the moment.)

Thank you for these.

Another option is to look into modules like these on Aliexpress. No need for a micro on the mailbox side. Just one or more switches connected to the module.

Thank you. I was actually wondering if there was a module that I wouldn't need to use a microcontroller for since I wouldn't need one otherwise.

You also have to consider identification. If the nieghbour has the same unit then you will be notified when he gets his mail and also when you get your mail.
without knowing what signal is from what mailbox.

So what would your recommendation be to achieve this? Have a unique ID to transmit along with the data?

in your case the most easy solution is just to buy some Chinese remote doorbell and connect it's button to your mailbox sensor.

Something like this:
https://www.aliexpress.com/item/1005004701032074.html

I am not looking to integrate pre-made solutions, but thank you anyway.

"Determined through a combination of sensors, logic gates......" Really, the sensing needs to be that complicated, not just a switch that causes the unit to transmit one signal burst? A very simple FM carrier modulated with perhaps a 1KHz tone is all you need for the transmitter.

Well, the logic gates and sensors are just to determine if there is indeed mail to be picked up.

• There is a door sensor that would allow me to know when to check for the presence of mail (only check right after the door is closed - with a small delay to allow the flag to go up if it does)
• There is a sensor for the flag to make sure it isn't just that I put mail in the mailbox to be picked up
• There is the actual mail presence sensor (which I am leaning towards IR but I would be interested in a better solution)

The logic circuits wouldn't be complicated but they would tell the transmitter whether to transmit a 1 (for if the mail has arrived) or a 0 (for if I just picked up the mail so that the base station returns to its original "awaiting mail" state). I did think about FM briefly. Do you think that would be better for my purposes?

Are you designing this for sale or for your own use??

It is for my own use, but I want to refine it to the point where it could be sold (although I don't really intend to). Again, it is in part to help me learn, and in this case, I am trying to learn how to develop a PCB for my project that doesn't use premade modules (which take up space and look unprofessional to have these just smacked on top of a circuit board that they could have just been integrated into).

I guess you are re-inventing the wheel here as opposed to buying an off the shelf solution. If it is a learning project there must be 100 better ways to learn about R.F. and signalling.

That is partly the idea. I am the sort of person who wants to thorughly understand how something works and as such I have found that building what already exists from scratch helps me learn a lot. But I would not be entirely reinventing the wheel because I would still be using a radio IC and not creating my own from scratch (and I am not sure I would be able to anyway considering that they tend to use non-standard, faster-switching transistors which I assume would be hard to get a hold of even if I wanted to).

Be aware that developing a radio transmitter is not something a beginner should be attempting. There are too many legal requirements to ensure your transmission doesn't screw up someone else with a receiver. Violate any, and you can't complain if neighbours become irate and you are visited by men in black suits.

OTOH, it is entirely reasonably to use a module that has already been certified for use in your locality. You will have to verify that any such modules will work over the relevant distances and with relevant power supplies. That ought to be within a beginner's capabilities, and is a very good way of starting to learn about RF.

I wasn't planning on developing the whole thing from scratch. I was looking to do something more along the lines of integrating the circuit used in the module directly into my PCB design without the extra module circuit board. Ideally, I want to know what pin connects to what on the IC that I use and what it does, as well as how to shift things around on the circuit board to minimize the size required for the circuit board (with consideration taken for parasitics). I do intend to do my initial prototyping with a pre-made module, but after that, I want to collapse everything onto a single circuit board.

[tggzzz]

The phrases you use indicate that you are probably a beginner.

You bet. That is why this is in the "Beginners" forum.
If there are things that you can tell I would definitely need to know before I continue with this project please point me to some good resources.

How many years are you prepared to spend? Seriously.

I found Sklar’s Digital Communications: Fundamentals and Applications useful. I suspect you won't. http://www.amazon.co.uk/Digital-Communications-Fundamentals-Bernard-Sklar/dp/1292026065

Please don't interleave replies to multiple people making different points. It makes it difficult to understand you, and difficult to strip points out.

Be aware that developing a radio transmitter is not something a beginner should be attempting. There are too many legal requirements to ensure your transmission doesn't screw up someone else with a receiver. Violate any, and you can't complain if neighbours become irate and you are visited by men in black suits.

OTOH, it is entirely reasonably to use a module that has already been certified for use in your locality. You will have to verify that any such modules will work over the relevant distances and with relevant power supplies. That ought to be within a beginner's capabilities, and is a very good way of starting to learn about RF.

I wasn't planning on developing the whole thing from scratch. I was looking to do something more along the lines of integrating the circuit used in the module directly into my PCB design without the extra module circuit board. Ideally, I want to know what pin connects to what on the IC that I use and what it does, as well as how to shift things around on the circuit board to minimize the size required for the circuit board (with consideration taken for parasitics). I do intend to do my initial prototyping with a pre-made module, but after that, I want to collapse everything onto a single circuit board.

You may not be planning on that, but that is the consequence of what you are saying.

The last part of your response makes this sound like an "XY problem" where you aren't giving us all the information. I don't have time to extract everything from you, one point by one point. (Example: what is the benefit of re-engineering it onto a single PCB?)

[SanctePieDecime]

Please don't interleave replies to multiple people making different points. It makes it difficult to understand you, and difficult to strip points out.

I am sorry. I am just trying to answer everyone's questions (in what seems to me to be the most logical order which is mostly chronological).

Why wouldn't I find Sklar’s Digital Communications: Fundamentals and Applications useful? Is it because it doesn't actually cover anything that would answer my questions? If it does have the information needed to answer my questions I would love to order it. I am in the USA and was able to change the link to view it for purchase in the USA. I see that the back of the book has a notice that it is not authorized to be sold in the USA so I don't know if I will have trouble ordering it.

I am a little disappointed in most of the replies so far. I get that this is something difficult, and time-consuming, and knowledge-intensive. I merely asked for resources on how to integrate existing radio technology directly into a circuit board without using modules but so far most people are just busy pointing out how I am a beginner, and I would be better off using a pre-made module, and how this is something I shouldn't be bothering with. I'll admit that I am quite naive and definitely have a lot to learn, but that is precisely why I am asking for these resources. So far I have not learned anything from this besides that I should probably use a MHz-range transmitter to achieve longer transmission distance.

I am well aware there are legal requirements and I don't suspect that I would have much issue with them since all I am trying to do is basically transfer the existing circuit on the module boards directly onto my circuit board while developing an understanding of how the components integrate. As I said, I would be using existing radio ICs.


Almost all the responses don't even come close to starting to point me in the right direction. Taking that into consideration and the fact that I have not found anything in my research to start me on the right path I am beginning to wonder if any of you actually know how to do this. This is not meant as an insult, but a legitimate observation that all I can find as answers to my question both in my research and in this forum is: "modules".

So I will ask the question again:
Would anyone be able to point me to a source explaining how to integrate existing radio ICs directly into my PCB design (without using modules)?

This is my question and it doesn't really need more elaboration. Keeping it general means that I can learn how to do this for more than just my specific project.

If I am really so naive that I am doomed to fail, then you have nothing to worry about because I won't even be competent enough to get this to work.

If you do have any questions that are really necessary to point me in the right direction then ask away.

[radiolistener]

I am not looking to integrate pre-made solutions, but thank you anyway.

This is in vain. Such solution will be much more easy, cheaper and more reliable.

From your question it is clear that you're don't have enough knowledge to even understand suggestiongs, so it's hard to explain things for you. I'm sure if you're want to design RF transmitter as a newbie, you will fail, because it requires some knowledge and experience in this area. And obtaining such knowledge needs pretty much time for learning. So, it's much more effective and cheaper solution for you to use already designed modules.

For example, I can design and tune such transmitter, I have experience and many measurement equipment required to tune and test it, but if I needed such a solution, I would use ready-made modules, just because it's easier and cheaper.

Would anyone be able to point me to a source explaining how to integrate existing radio ICs directly into my PCB design (without using modules)?

I'm afraid there is no simple "point to" way. If you want to start, the most easy way is to learn some ready to use modules. When you will be familiar on how to use it, then it will be more easy to learn how they work and then how to implement the same circuit. Unfortunately there is no simple way to design some complex thing which requires specific knowledge in the area with no learning basic things and with no experience.

[SanctePieDecime]

I am not looking to integrate pre-made solutions, but thank you anyway.

This is in vain. Such solution will be much more easy, cheaper and more reliable.

From your question it is clear that you're don't have enough knowledge to even understand suggestiongs, so it's hard to explain things for you. I'm sure if you're want to design RF transmitter as a newbie, you will fail, because it requires some knowledge and experience in this area. And obtaining such knowledge needs pretty much time for learning. So, it's much more effective and cheaper solution for you to use already designed modules.

For example, I can design and tune such transmitter, I have experience and many measurement equipment required to tune it, but if I needed such a solution, I would use ready-made modules, just because it's easier and cheaper.

Like I said. I am not designing anything new I am basically just transferring an already existing design onto my PCB design.

Here is the best analogy I can use. Using an Arduino is good for prototyping, but eventualy you want to move away from it (and use the barebones ATMEGA328 without any of the circuitry that you don't need) in a final product. So too with the radio module but in a slightly different way. Just as using the ATMEGA328 directly on your circuit is more space-efficient and cost-efficient than using the whole Arduino, so too with the radio communications system.

[globoy]

So I will ask the question again:
Would anyone be able to point me to a source explaining how to integrate existing radio ICs directly into my PCB design (without using modules)?

Probably there is no one single source to explain this since it touches a number of topics.

• You can probably understand what circuitry is needed to support any particular radio IC by looking at its user manual and/or by looking at the schematics for modules that use it.
• The circuitry needed to operate the radio IC will vary depending on what radio IC you choose.  Most are designed to be controlled by an external (or internal) micro-controller via an industry standard interface such as I2C or SPI.  Your particular application will determine other aspects of the other circuitry such as interface to sensors and power supply/power management.  The devil is in the details here and you haven't shared enough specific information to suggest specific solutions.
• You will probably need to tune the RF Output circuitry of your design.   This is the aspect that is causing most people to suggest using a pre-built module.  The tuning itself depends on the circuit (matching circuitry, frequency, antenna type, etc) and is a process that requires some expertise and specialized measurement tools.  Do a search for "tuning pcb antenna matching circuits" or something like that to get an idea.  Tuning can mean the difference between a transmitter that can transmit 10 feet or one that can transmit 1000 feet.
• Maybe I missed it but is this a "one-off" (just for you) or you want to design something that you will manufacture, market and sell?  If it's going to be for sale then you will need to contend with regulatory approvals (e.g. FCC, CE) as well - and probably you should put in place a whole validation/verification effort to make sure your device works as intended in all the various situations it will find itself.  If it's a one-off then you can get by without sweating any of those issues.

Engineering design is a process that involves setting requirements/goals and then looking for solutions to implement those and dealing with all the various compromises/decision points that come up, as well as considerations for the regulatory and manufacturing requirements of hardware products.  It's great (IMHO) that you're interested in pursuing this for your own edification but often someone who is beginning doesn't really understand what they are asking for and either doesn't know to include certain details in their questions  that are necessary to get good answers or significantly underestimates the problem they are trying to solve.  Add that to the semi-anonymous nature of a medium such as this one and you get a lot of responses such as the ones you've just seen.

If this were my project - lol, and I have thought about automating my mailbox but am just too lazy - I'd probably settle on a LoRa radio module and some simple micro-controller capable of very low power operation.  Or if I really wanted to do the radio part too, I'd include the LoRa IC on my board (understanding I'd just bitten off more work and expense).

[radiolistener]

Just as using the ATMEGA328 directly on your circuit is more space-efficient and cost-efficient than using the whole Arduino, so too with the radio communications system.

Even in this case using ready to use module will be the most efficient and cheaper solution. Designing RF transmitter is not easy like designing hobby digital arduino circuit. It requires to understand many things like transmission lines, impedance matching, etc.  As other said here, this is definitely not a task for beginner. If you're want to start such design with no knowledge and experience, I bet you will fail.

[armandine2]

you might be able to avoid the thousand page book, or some of it at least

[tggzzz]

Please don't interleave replies to multiple people making different points. It makes it difficult to understand you, and difficult to strip points out.

I am sorry. I am just trying to answer everyone's questions (in what seems to me to be the most logical order which is mostly chronological).

Why wouldn't I find Sklar’s Digital Communications: Fundamentals and Applications useful? Is it because it doesn't actually cover anything that would answer my questions? If it does have the information needed to answer my questions I would love to order it. I am in the USA and was able to change the link to view it for purchase in the USA. I see that the back of the book has a notice that it is not authorized to be sold in the USA so I don't know if I will have trouble ordering it.

I am a little disappointed in most of the replies so far. I get that this is something difficult, and time-consuming, and knowledge-intensive. I merely asked for resources on how to integrate existing radio technology directly into a circuit board without using modules but so far most people are just busy pointing out how I am a beginner, and I would be better off using a pre-made module, and how this is something I shouldn't be bothering with. I'll admit that I am quite naive and definitely have a lot to learn, but that is precisely why I am asking for these resources. So far I have not learned anything from this besides that I should probably use a MHz-range transmitter to achieve longer transmission distance.

I am well aware there are legal requirements and I don't suspect that I would have much issue with them since all I am trying to do is basically transfer the existing circuit on the module boards directly onto my circuit board while developing an understanding of how the components integrate. As I said, I would be using existing radio ICs.


Almost all the responses don't even come close to starting to point me in the right direction. Taking that into consideration and the fact that I have not found anything in my research to start me on the right path I am beginning to wonder if any of you actually know how to do this. This is not meant as an insult, but a legitimate observation that all I can find as answers to my question both in my research and in this forum is: "modules".

So I will ask the question again:
Would anyone be able to point me to a source explaining how to integrate existing radio ICs directly into my PCB design (without using modules)?

This is my question and it doesn't really need more elaboration. Keeping it general means that I can learn how to do this for more than just my specific project.

If I am really so naive that I am doomed to fail, then you have nothing to worry about because I won't even be competent enough to get this to work.

If you do have any questions that are really necessary to point me in the right direction then ask away.

Sigh.

Start with knowing the FCC regulations that apply, and understanding them. Move onto the physics of PCBs and antennas and propagation. Proceed to being able to configure and test the implementation. End up with getting regulatory approval for your device.

Yes, some of could do it. We also know how much time and effort it would take. That's why we wouldn't bother.

But, hey, go knock yourself out!

[radiolistener]

Almost all the responses don't even come close to starting to point me in the right direction. Taking that into consideration and the fact that I have not found anything in my research to start me on the right path I am beginning to wonder if any of you actually know how to do this. This is not meant as an insult, but a legitimate observation that all I can find as answers to my question both in my research and in this forum is: "modules".

In short: first you're needs to select working frequency, then you're needs to select modulation which you're want to use. Then you're needs to decide where you will do modulation - in digital domain as DSP or in analog domain as hardware analog modulator. In that way you can implement IF stage of transmitter. Then you're needs oscillator and mixer and do frequency shift from IF to working frequency. Then add RF amplifier, filters and matching circuit to match it with antenna. Then you will need to tune all things and debug it.

And before use it with antenna you will needs to measure it's spectrum and check if it comply with government regulations. Most of all it will not, so you will need to fix it by adding proper filter, change some components or properly tuning it.

If you knew how much knowledge, time and measuring equipment this would require, then I am sure you would lose interest in independent development. And since you're newbie and don't have knowledge in this area, this is why no one even don't start to suggest you on how to do it.

[tggzzz]

Almost all the responses don't even come close to starting to point me in the right direction. Taking that into consideration and the fact that I have not found anything in my research to start me on the right path I am beginning to wonder if any of you actually know how to do this. This is not meant as an insult, but a legitimate observation that all I can find as answers to my question both in my research and in this forum is: "modules".

In short: first you're needs to select working frequency, then you're needs to select modulation which you're want to use. Then you're needs to decide where you will do modulation - in digital domain as DSP or in analog domain as hardware analog modulator. In that way you can implement IF stage of transmitter. Then you're needs oscillator and mixer and do frequency shift from IF to working frequency. Then add RF amplifier, filters and matching circuit to match it with antenna. Then you will need to tune all things and debug it.

I suppose we could just point the OP to investigate the cost of a basic RF test setup that is sufficient to measure the harmonic and intermodulation products. Without that, regulatory approval will be a slow iterative (and expensive!) process

[tggzzz]

OP...

Why don't you start with the most basic design parameters: the link budget.

That means understanding the tx power, the tx antenna properties, the path parameters (loss, fading, multipath, Fresnel zone), the modulation scheme, the rx antenna, the rx noise and susceptability, the allowable error rate, the channel coding.

Or, if you use modules, you only have to understand the tx/rx antennas and the path parameters. Don't fall into the beginner trap of thinking you can increase the margin by putting a high-gain antenna on the transmitter. For an X dB antenna gain you are required to reduce the tx by X dB. That doesn't apply to the rx antenna.