High Data Rate RF Link

[thecranium]

I have a device that transmits 5Mbps serial data over a cable to another device. I would like to replace the cable with a wireless link, but am having some trouble finding a good solution. I don't need or want any encoding/decoding of data, simply to transmit ones and zeros over RF from one place to another (under ~100 ft.). I've attached a simple picture of what I'd like to do. Does anyone have any ideas? Ideally, an off-the-shelf solution is preferred, but I'm not opposed to building something if I need to.

 

[hevans1944]

A pair of Arduinos with Zigbee shields can implement a rudimentary, one-way, simplex data link, as shown on your sketch. Lots of information is available in various forums on how to do this. Learn to let Google be your best friend in finding possible solutions before coming here. With line-of-sight, and no obstructions between transmitter and receiver, it should be possible to reliably reach one hundred feet from transmitter to receiver. Cost is less than two hundred bux for the Arduino/Zigbee pair, the last time I played with this technology a few years ago. Add a few weeks of your time spent climbing the learning curve and programming the Arduino. Note that, after programming, the Arduino/Zigbee pairs are disconnected from the PC used for programming and then function as a stand alone peer-to-peer data link. Plug and pray.

If you need more range, a WiFi version is also easily supported by Arduino, but at considerably higher cost. It's been awhile since I investigated Arduino/Zigbee, although I plan to revisit this technology "real soon now" and have no plans to examine WiFi alternatives because I currently have no interest in the Internet of Things (IOT), which appears to be popping up just about everywhere. But Google "IOT WiFi datalink" to see what is currently available. We help those who help themselves here. Tell us what possible "solutions" you have found, off-the-shelf or do-it-yourself. We, being a motley crew of hands-on experimenters and makers, naturally prefer the DIY approach unless a suitable solution can be procured dirt-cheap from an Asian vendor. But do your own due dilligence and then ask us pertinent questions if there are issues you don't understand.

Good luck with this project. I think it is an ideal entry point for a motivated beginner. Much more exciting than flashing a few LEDs with a trivial Arduino program.

 

[thecranium]

The reason I posted to this forum was that I spent 3 or 4 days Googling this and coming up with very little. I came across the Arduino/Zigbee solution somewhere during that time but wasn't exactly sure it would do something as simple (albeit at a high data rate) as what I wanted. I will re-visit that, thanks for the suggestion.

While I am a beginner at RF, I'm not exactly a newbie...I have a degree in Electrical Engineering and have worked as an EE/Software Engineer for over 30 years. I just have almost no recent experience with designing hardware and, again, no real experience with RF. So, I thought this was the perfect place to post my question. If I had known that the price of an answer was to be chastised by the first person to comment on my question, I would have posted elsewhere. Lesson learned.

Thanks again for the info.

 

[hevans1944]

If I had known that the price of an answer was to be chastised by the first person to comment on my question, I would have posted elsewhere. Lesson learned.

"The price of answer was to be chastised?" What did you mean by that statement? And what lesson do you think you learned? This isn't a teaching forum, and I wasn't trying to chastise anyone, just suggesting that you haven't been very diligent yet in seeking answers before posting here.

I can remember having to spend weeks in the early 1980s, not just a few days, trying to search online databases with SQL and having to seek the help of "experts" to properly format questions. And then wait for "answers" that likely as not did not address my original query because the "expert" did not understand my problem, nor my questions seeking answers to the problem, nor the answers that came back. This was before Google, Bing, DuckDuckGo, and various other Internet search engines opened the vast abyss of unfiltered and un-vetted online data to the unwashed masses, who imagined that they didn't need or want white coats and micron-filtered, air-conditioned rooms to communicate with computers. Fortunately, personal computers came along shortly after my initial experience with database inquires, and PCs soon spawned Google and other pretenders in college dormitories all over the country.

So, you spent three or four days with Google and came up with nothing!? Try this search string "arduino zigbee peer-to-peer data link" which yielded about 97,400 results for you to pursue? Go to, and read the information at, this link first.

And what exactly were your negative search results? You didn't post any links to the sites that offered no useful information. If you had, instead of complaining about my "chastising" post, perhaps someone else here with more patience than I have would have responded to send you in the proper direction. Perhaps you do need to post your question elsewhere. Good luck with that. This is not intentionally a drive-by forum, where you post a quick question, get some sort of immediate answer, and then move on, never to be seen or heard from again. This forum requires a dialog, a back and forth exchange of ideas, to produce results. That's what I meant when I said we help those who help themselves.

So, you are a degreed electrical engineer who has spent most of the last thirty years rolling code, instead of wielding a soldering iron or trying out real circuits you have constructed with your own hands. If that's your bag of tea, you are still welcome here at Electronics Point, and you may even have something to contribute via software to those newbies among us who don't have a clue yet about programming. But just about everyone here is willing to learn.

When I decided to explore Arduino/Zigbee for a point-to-point data communications application, I knew absolutely NOTHING about either Arduino or Zigbee. RadioShack was still a brick-and-mortar business, so I purchased all the Arduino "stuff" I thought I might need at "going out of business, fire-sale" prices. The Zigbee shields required purchasing the RF hardware from a third-party source, but no RF circuit design on my part was necessary. It was all plug and play, or with my limited software skills, more like plug and pray.

Actually, I had off-and-on been experimenting with RF as a hobbyist since my early teenaged years. After joining the Air Force in 1963 and getting "properly" trained up as an electronics technician (a promotion, I think, from hacker-hobbyist), I obtained a Novice amateur radio license (KN8UTJ) in 1966, built a Heathkit SB-300 ham radio receiver, and designed and built an 80m CW transmitter rig from scratch. Lots of fun and no Internet in sight yet. But there were books available, and public libraries that would loan books for free. There are lots of ways to learn, including on-the-job-training (OJT) and a college education.

After my four-year Air Force commitment was up, I let my non-renewable Novice license expire and went to work as an entry-level technician for the University of Dayton Research Institute. As a full-time employee, I was eligible for reimbursement of tuition, so I took advantage of that and, going to school part-time, graduated from the University of Dayton with a Bachelor of Electrical Engineering degree in 1978. Working for UDRI during that period meant I received LOTS of OJT in a wide variety of disciplines. I even dabbled in embedded computers (Intel 8085 microprocessors) and rolled some code during the 1980s and 1990s. Now fast-forward to this century: divorced, retired, children out the nest and on their own... time to re-enter amateur radio and learn about the latest microprocessors!

 

[davenn]

. If I had known that the price of an answer was to be chastised by the first person to comment on my question, I would have posted elsewhere. Lesson learned.

You were a very long way off-base with a comment like that!

 

[Nanren888]

You might want to add a clock.

 

[hevans1944]

You might want to add a clock.

For "ones and zeroes"? Why would you wanna add a clock to that? Besides, the OP stated:

I don't need or want any encoding/decoding of data

Self-clocking data protocols have existed for a very long time, so adding a clock appears to fall into this category. As does error-detection and correction. So, lets apply the KISS principle and just stick with ones and zeroes, letting the bits fall where they may.

 

[Nanren888]

I wanted only to raise the question of sampling and clocks.
Fair enough perspective. So more clearly;
If the binary waveforms in your simple diagram above are of (from & to ) somethings that produce and cope with self-clocking protocols then your figure is about all you need. This may be what you intended.
If self-clocking protocols are to you more "encoding/decoding of data" and the binary needs to be sampled at the right times, stable, valid sample times, then you may want to think about adding a sample clock on the sending side so the link knows when a sample is valid and a clock at the output so the receiver knows when the output is valid.
.
Reacting to your comment about oinly wanting a heads up on what's modern and available, I guess many modern links send packets of data in contrast to sending assynchronous binary data. No mention of frequency, so I guess an ISM band is appropriate. If you turn your seraching to ISM systems, you may get a reasonable rundown on what's available. Your rate is quite high for small scale systems. Happy hunting.

 

[hevans1944]

I wanted only to raise the question of sampling and clocks.
Fair enough perspective. So more clearly;
If the binary waveforms in your simple diagram above are of (from & to ) somethings that produce and cope with self-clocking protocols then your figure is about all you need. This may be what you intended.
If self-clocking protocols are to you more "encoding/decoding of data" and the binary needs to be sampled at the right times, stable, valid sample times, then you may want to think about adding a sample clock on the sending side so the link knows when a sample is valid and a clock at the output so the receiver knows when the output is valid.
.
Reacting to your comment about oinly wanting a heads up on what's modern and available, I guess many modern links send packets of data in contrast to sending assynchronous binary data. No mention of frequency, so I guess an ISM band is appropriate. If you turn your seraching to ISM systems, you may get a reasonable rundown on what's available. Your rate is quite high for small scale systems. Happy hunting.

Thank you for your comments, @Nanren888. I visited your website and came away quite impressed with what is going on "down under," so-to-speak, in New Zealand. One of our best liked and most prolific contributors to Electronics Point went by the online moniker of @KrisBlueNZ, but he is now sadly deceased and (so far) no one has stepped up to fill the void Kris left behind. He was the first person who engaged me in conversation here, and the first person from New Zealand I had ever met... I have met others here since then. I miss him a lot.

Your comments that I have read on other topics posted here on Electronics Point appeared to be spot-on and relevant to the topics. However, I doubt the OP (Original Poster) has hung around for a discussion, much less followed what you presented, since all he or she wanted to do was to replace a simple wired data link with an RF data link. So, maybe an LED transmitter, a photocell receiver, and two telescopes aimed at each other would do the trick? All stuff available off-the-shelf and really cheap if purchased from Asia.

As a Boy Scout (Explorer Scout, actually) in the previous century, I built a wireless voice transmission system, but it used a low-voltage incandescent flashlight bulb in the transmitter, and a vacuum photo-diode from a 16mm film-sound-projector in the receiver to achieve a range of only about twenty feet or so. Maybe if inexpensive laser diodes, or even ordinary red light-emitting-diodes (LEDs), and maybe a better pair of telescopes, had been available it could have been tweaked for use at one hundred feet.. but twenty feet was "gud enuf" to demonstrate "proof of concept," which was all I wanted to do at the time.

There was no mention by the OP of how to ensure data integrity, which is usually the prime consideration of any but the least important digital data links... I used TV remote controls as an example of where data integrity doesn't really matter, but I didn't bother to mention this is only because the end-user is "in the loop" as far as feedback is concerned. The end-user knows pretty much immediately when the remote control isn't working as required. And most people know it is because the batteries need to be changed or charged.

So I guess we will see where this thread goes...

 

[Nanren888]

Many thanks for your words.
Have to agree, I also had the impression that the original poster had managed to read something negative into your answer and was unlikely to see later messages.
Optical - an interesting option.
Going off-topic.
I do admire the practical experience aspect some of you bring to answers. I have worked a while in contract R&D and so have limited experience in a few areas.
Also, updated the webist elink to the newly established technical one.

 

[hevans1944]

Most of my career was spent performing contract R&D work for U.S. Government sponsors. Everything we did was a one-off, a proof of concept that, if successful, was tossed over the transom to a real manufacturer the likes of Hughes, Northrop Grumman, Boeing, Raytheon, Texas Instruments and so on ad nauseum. Funny thing is, none of these folks ever came back and asked us for advice. And we very seldom got any feedback about the results of our efforts. I just have to assume it was my tax dollars, well spent. that got us to the Moon and scared the crap out of the Russians with our Star Wars program... and, yeah, I was a Star Warrior in the 1960s, working with the Air Force Weapons Lab in Albuquerque NM to build and test the Airborne Laser Laboratory (ALL)..It was fun while the funding lasted,