Bluetooth Traffic Monitoring 1

[Trffc]

My question is regarding Bluetooth traffic monitoring. My company is bidding on a project that will involve detailed data collection on average speeds. Research indicates that Bluetooth traffic monitoring devices are very reliable and much more affordable than traditional methods. I was wondering if anyone here might have some input about the pros and cons of these devices.

Additionally, we found a company called Acyclica (

http://www.acyclica.com)

and noticed they have an extremely affordable product. Has anyone here worked with them? What are your thoughts about their product, the BlueCompass?

Thanks

 

[dcnelson]

Traffax Inc has been manufaturing and selling Bluetooth-based traffic monitors for the collection of traveltimne and origin-destination data since 2009. We sell and rent units that can be used to collect data for studies, as well as for realtime monitoring applications. Please visit our website:

http://wwww.traffaxinc.com

or contact us for detailed information.

 

[Debaser]

I had a presentation from a survey company re aquiring data from Bluetooth enabled devices. They suggested that (in the UK) they could get a 30% sample rate (that is 30% of the population had such a device and could therefore be recorded). Getting an adequate sample size in a given time period could be the crux of your study.

 

[dcnelson]

The sample sizes (proportion of traffic captured) for Bluetooth-based traffic detectors vary from place to place and are influenced by a number of factors --three of which include the percent of traffic with a detectable Bluetooth signal, the configuration and placement of the data sensors, and also the methodologies/tools for used to process the data.

It is possible that the presentation you refer to may well have been conducted using our devices; however I would need to assk a few questions to confirm that.

I am new this this forum and do not want to step over the boundaries with respect to discussing specific products set for this forum.

If you send an e-mail to our generic company address (traffax@traffaxinc.com), the message will be forwarded to me. I woulld be happy to share information I have from studies in locations around the world, discuss sample size requirements associated with traveltime data for short term studies and for realtime monitoring applications, and also to provide you with reports/papers published by others using our equipment.

In general, the characteristics of the data collected using our Bluetooth-based devices, combined with the capabilities of the processing software packaged with the devices provide a very cost-effective approach to collecting travel time data on freeways, arternials, and in urban environments.

 

[Trffc]

dcnelson,

I appreciate your reply. We worked with a small firm in our local area and after some intensive research we have decided to go with Acyclica for our needs. They investigated your product but determined it was too costly for the project. One thing that appealed to us about Acyclica was not being chained to their software. With their very basic XML feed we have been able to build a simple, analytical tool that more than meets our study needs. Some of their software looks great, too, and we might look at it for the future, but for now we needed a very basic interface with the data.

Of interest to other members of the forum may be some of our results. We are currently employing the hardware in a study to attempt to understand what the density of Bluetooth travellers on our roadways might look like. We won't have the results for a few months yet, and I promise I will keep you posted.

Thanks!

 

[IRstuff]

It's certainly not as high as it should be. I still see lots of people holding their cell phones to their ear while driving, even though that's illegal in California.

How do you deal with multiple Bluetooth devices in the same vehicle? I use my own Bluetooth, but my car is equipped with one as well, just for the phone, but there might also be one or more Bluetooth headsets for MP3 players and the like.

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[Trffc]

IRstuff,

Great question. Something I had written off because I don't think it's a significant enough phenomenon that it will create an issue for sampling studies. I could be wrong. Do you have any thoughts?

For our other interests, which are origin-destination and travel-time studies it won't affect the results at all.

 

[IRstuff]

It may be sufficiently early in the Bluetooth era for something like that to not grossly skew your results. Likewise, time of day may help/hinder that.

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[EagleChu]

Trffic,

What is the result of using Acyclica? I'm also going to purchase some Bluetooth receiver to investigate the feasibility of using it in travel time indication.

 

[Drumchaser]

Sorry to be tardy on this one. It is evident that I am comparatively new to this technology. In perusing some of the proprietary systems noted above I was unable to detect much in the way of cost.
This technology is very interesting.

 

[dswfondy]

Just met with Acyclica's President and CEO today. We are looking at doing a 40 to 100 unit OD study in our area. They seem very willing to adapt product to what we want in a package that will work.

The problem with Bluetooth detecting, is that Bluetooth devices are getting more locked up to detection for security. Most new phones only are visible for 120 seconds then goes off.

The good thing is that Acylica's BlueCompass does also detect WIFI if you want. Which seems to be the future in traffic detection. A phones WIFI, if switched on stays on and pings for networks.

So far their portable units are not up to par with what our needs really are, solar panel is attached to box. Which makes it impossible to get a good angle to the sun. Antenna's need to be changed out, they supply a 12dbi omni directional which is way to powerful. Which is to powerful for a 300 ft four-lane highway study. The insides of the box has great parts, but is way to big and need to be more compact. Box could easily be half the size and fit all things currently inside.

But they are willing to adapt thats a good thing

I would also look into BlueToad by trafficCast, we have used these and Traffaxinc also. Both area great products

 

[Trffc]

Interesting ongoing discussion. Sorry for not continuing to be a part of it. EagleChu, we ended up having very good results with the Acyclica product. There were some firmware issues with some of our first units, but they were eager and responsive to take care of that.

We were looking at a suburban setting in the Southeast, and were pretty surprised to find that we saw anywhere from 6-20% of all vehicles with a detectable Bluetooth signal. We did not think we would find this at all. This was after a good deal of scrubbing, too. Acyclica actually helped us craft an algorithm to reduce the weighting of coincident MAC addresses as they went through our network.

We still have their hardware and will surely be using it again as soon as possible.

dswfondy, I find it interesting that you weren't happy with their antenna choice, because one of the other aspects we really liked was the ability to pick our own. We didn't buy Acyclica antennas and instead found some affordable ones on Amazon.com.

I've had regular contact with their sales VP and he's let me know they have several antenna options, too. An 8dBi which can be embedded in the case or directionals. He's also told me about the WIFI readers, which sound like a great option if we want to upgrade for higher sample rates. We might do this in the future. We're also looking at integrating our system into IWZ projects.

We also didn't have an issue with the solar panel because all of our mounts were to poles, in which case it was straightforward to orient them to the south.

If you end up working with them, I'd love to hear the results. It sounds like your project is a great deal bigger than anything we do, though Acyclica seems to be a valuable partner and might enable us to chase some projects of that scale.

 

[IRstuff]

"A phone's WIFI, if switched on stays on and pings for networks. "

Only if you keep that on. I've generally been very unhappy with battery consumption with WiFi on, and turn it off as I leave the house.

TTFN
faq731-376

 

[dswfondy]

Correct it does drain battery, but just making a statement that it is more likely to be visible to detection then Bluetooth is currently.

I understand that we can change our antenna type, we use different antenna's with other products depending on the study as well. I just think their standard antenna is overkill. I dont understand why you would need a antenna to pull data 1/2 mile away is all.

They are a young company and are willing to adapt to what us as customers want so far. I am excited to get their product out and try them in the field, hopefully this week

 

[Acyclica]

Hey guys, Adam from Acyclica here. A few customers directed me to this discussion, and I thought I would chime in to represent our technology. Trffc, I'm glad you've been having good results! dswfondy, your feedback is fantastic, keep it coming!

As for WIFI, we have definitely found it is a better technology to rely upon for the reasons mentioned (no "sleep" on discoverability), but we think that it is left on in vehicles more often than Bluetooth, too. The reason being is manyfold, but here are the top two:

1. Hands-free/Bluetooth users tend to activate Bluetooth when using it, not always whereas the majority of WIFI users leave it on and it is only a small subset that controls it for battery concerns. Even for those WIFI users that are typically more "on-top" of their firmware, they often forget to turn it off between gateways like during the home-to-work commute or vice-versa.

2. Apple actually suggests in their documentation that users leave WIFI on all the time to enhance geo-location services (via signal strength embedded in the SSID). Many advanced Android users have caught on to this, and some great Android Apps utilize this as well.

Also, because of the signal strength we can do very advanced "Local" filtering on the datasets. Trffc mentioned the ability to download the XML report directly and interface with it through a homebrewed analytic software. This is possible because on the device we scrub out all unnecessary duplicate MAC detections and report to XML and CSV only the instance with the highest signal strength. Thus, when you fetch your database it is Ready-To-Go, NTCIP compliant and easy to integrate with third-party applications.

In fact, we do similar Local Filtering with our Bluetooth readers, but it is based on an algorithm run over the aggregate of detections instead, so is ostensibly less accurate as to proximity of the vehicle to our device.

But this begs at one more question regarding this technology: How effective is a Bluetooth reader at determining true proximity of a detectable device? The short answer is that it is terrible. The only real way to accomplish this is by restricting the scanning range with a low dB antenna choice. This way you have a defined search area which you can shrink to reduce error. That said, antennae frequently are capable of scanning further than their specification. Also, by restricting the gain of the antenna, there is a chance that device detections will be missed if travelling at high speeds as a reader scans through all available 2.4 gHz channels that Bluetooth protocol operates on.

By using WIFI we can use a very high gain antenna to our advantage to ensure that we're capturing as big a sample set as possible. Furthermore, it is trivial for us to filter out data that is irrelevant from a network of readers(pedestrians, cyclists, other outliers, etc...), so there really is no reason to be concerned about an overly powerful antenna. That said, our readers all come equipped with RP-SMA connectors and we work with all of our customers to make sure of the nearly uncountably many antennae choices available the right one is chosen for the project.

I know I've shared a good deal of information here, so please feel free to reach out with any questions. We're very excited about this technology, and very customer focused. We look at a chance to work with you as a partnership and want to know how our technology can be made better to help you achieve new applications!

Adam Hicks
VP Sales & Distribution
ahATacyclica.com