Keeping both signals in phase is an additional constraint ! ;-)
Then again, since your end result is an FFT per sec, you could use a 1s timer to trigger the start of processing both stored/accumulated signal "1 sec slices" to get your FFT result. With today's multi-Mega-samples/s processing speeds, a single DSP chip could have enough "humpf" to pull it off... but within your IC price budget ? TBD.
Yes, StephT, that was my assessment, too. With DSP chips or even and RTOS system, you have to round robin the samples. If you want to keep the vibration or audio samples in phase, you need to have concurrent processing in an SOC.
In FPGAs, you could have 2 streams processing in real-time and do with much less RAM (ongoing processing of the sample stream) ; with DSP chips, you'll have to time-share the processing meaning storing the samples while the other stream gets processed...
1s ; that's (relatively) pretty fast ; would need to re-check DSP theory to get a feel for how much RAM (to store the samples in the FFT topology) you'd need to store results... You could also look up Wikipedia for some pointers ! ;-)
Thanks StephT. I'll check that out, too. I was thinking a PSoC that would do signal amplification via op amps in parallel, then >> A/D conversion >> digital filter and FFT for signal analysis in Frequency Domain. 32 bit is okay.
@78RPM : your FFT solution will depend on how fast, how deep you want the structure to be... MAC (mult-and-accumulate) cells are what you need and even FPGAs have that (could be way faster than a DSP chip).
Would your CPU/micro need to be 32 b or less ?
I was at a Renesas trade show and they seemed to have nice chips / solutions. But have not work with those...
An ex-MicroChip-Program-mgr was saying Renesas did it right (compared to the PIC32)... so he said...
Suggestions for classes. From what I have reviewed, things are hopping around a bit, since CEC inception. Structured flow would nice, evolving around a certain product or theme. As for specifics would have to dwell on that for a bit. There is so much out there and more coming everyday, it's hard to pin point the target from the "noise" When the light comes on, I'll relay the thoughts to someone here.
Thanks for bringing out the potential for preprocessing/data reduction at the sensor node. Where you have both the power budget and the mips to support it, minimizing traffic over the transport is really desirable...
I see there point but always believed, get them in the door, we will take lic at a different point depending on the cust comlplexity.
Would think that the silicon mfgrs would provide a good well rounded software platform w/o royalies to get more folks into there products, therefore more sales. Always setting the barriers to entry high eliminates a lot of smart folks. That is where Parallax Rasp pi and now a flood of others took in some good $ and got this all started, and now everyone whats a part of it
@@ A webinar on the GPS and related systems would be great. A webinar on engineering ethics (thinking ahead ) would be great. I have seen and been involved in projects where the engineering was sucessful but the results ( colateral damage ) were pretty horible.
Biggest challenge : most probably integrating everything ! It will be case-by-case (or application-by-application) but I'm still not convinced everything need to be Big Data, Cloud or broadcasted to all the atoms in the universe ! Many applications will need (still) to be wireless (cannot be "leashed" by, say, optical fiber), so IPv6-over-the-air ?? Probably better ways to do that, but Wi-Fi solutions do exist...
If there is a class which requires some software and hardware then it should be clearly stated a few weeks ahead of time so that we can acquire them and get ready for the class. This would be most helpful. Like the course starting on July 7th Designing With ARM Cortex-M4 (& a Real Dev Board). Discount on those items will help too.
@Charles @Jennifer Charles can lecture more classes like the recent MQX/Freescale ARM class. Make something practical from start to finish. Charles presents this stuff well. I know there is an upcoming class of similar material.
What do you see as your biggest challenge in the IIoT?
Multiple challenges: convincing clients to take the security issues seriously and trying to find the "best" solution for a particular device with so many protocols (Wi-Fi, BLE, etc.) and devices (SoCs, modules, etc.) to choose from.
For me, it seems like the latest "buzz word". Seems like there is some applicability in most industries. Also seems that many companies are talking about implementing it for their customers, but no one wants to be the first to do so. Many are afraid to start down a road that ends up NOT being the standard approach.
@kchandra "Things will get so complicated in the future"... Yet that vey idea was suggested to be an asset when we want to distribute security across resouces in a network. No one node is responsible for all the security.
Seems like sensor fusion functionality could be combined with gateway functionality => device both receives sensor data from multiple sensors (possibly from a low-power, short-range wireless network), processes the data appropriately, then passes it off to a higher-speed/bandwidth network...
I think that the general IoT will be a concept. There will be different (as you said) universes, with their own standards. Most will have common fundemental rules, but different applications will require specific rules.
We deal with "Big Data" -- millions or hundreds of millions of points per run sometimes. Or thousands of projects with millions of points collected. We use Interbase SQL server -- since maintenance and setup is relatively simple.
Today in the TV News says that Inteligent Power Comsuption Counter (Tax Meter) have a blinking LED that can help thieft to know if a house in habitated or empty, simply, LED blink more slowly when there're a low comsuption: This is a great risk threat in the IoT. It's so ridiculous than the contdown timer displays of the bombs in the movies.
Charles' question was: Who is working from the "big data" side? What are your biggest challenges?
I n the past have faced similar problems. Have had great success with data compression. This however can be data dependent. For example streaming video can only send pixels that change from one frame to the next.
A good application of Big Data would be Zooniverse style web sites that help scientists find planets around stars or count animals on the African Savanna or classify tiny fragments of ancient Greek parchment.
If you experience audio interruptions and are using IE, try using FF or Chrome as your browser. Many people experience issues with IE. Also, make sure your flash player is updated with the current version.
Tower of Babel network - Yes it's an issue. Inserting intelligent nodes with lots of storage and data distilling to send along at pre-defined intervals in some systems. On other systems we're just drinking from the firehose.
Charles' question was: Are you facing the interface to legacy or 'mis-matched' protocols?
Not currently but have faced similar problems in the past. I can recall of at least one case where we had a big / little endian issue. The one group had simulated it on ASIC and all was fine but the signal was backwords and thet next group consequently processed it wrong.
Can you provide guidanec as to who might be the manufacturers to look at for getting demonstration systems rapidly deployed with as little up-front time as possible? Looking to evaluate systems of a couple hundred nodes, so setup time is an issue.. Thanks
If you don't see the audio bar at the top of the screen, please refresh your browser. It may take a couple tries. When you see the audio bar, if it doesn't start automatically, hit the play button. If you experience audio interruptions and are using IE, try using FF or Chrome as your browser. Many people experience issues with IE. Also, make sure your flash player is updated with the current version. Some companies block live audio streams, so if that is the case for your company, the class will be archived on this page immediately following the class and you can listen then. People don't experience any issues with the audio for the archived version.
The streaming audio player will appear at the top of this web page when the show starts at 2 PM Eastern time today. If the audio doesn't start automatically, click on the play button to start it. Note however, that some companies block live audio streams. If when the show starts, the audio bar doesn't appear or you don't hear any audio, try refreshing your browser. If that doesn't work, try using Firefox or Google Chrome as your browser. Some users experience audio interruptions with IE. Also, make sure your flash player is updated with the current version. If that doesn't work, your company is likely blocking the live stream. The class will be archived immediately following our live taping and you will be able to listen to it then. You shouldn't experience any problems with the audio when listening on-demand.
In many engineering workplaces, there’s a generational conflict between recent engineering graduates and older, more experienced engineers. However, a recent study published in the psychology journal Cognition suggests that both may have something to learn from another group: 4 year olds.
Conventional wisdom holds that MIT, Cal Tech, and Stanford are three of the country’s best undergraduate engineering schools. Unfortunately, when conventional wisdom visits the topic of best engineering schools, it too often leaves out some of the most distinguished programs that don’t happen to offer PhD-level degrees.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.