With hidden cameras being increasingly used to snoop on individuals in hotel rooms and Airbnbs, the goal is to be able to pinpoint such rogue devices without much of a hassle.

The technology, named Lumos, is intended to “visualize their presence via an augmented reality interface,” according to Carnegie Mellon University’s Rahul Anand Sharma, Elahe Soltanaghaei, Anthony Rowe, and Vyas Sekar in a recent article.

The platform detects and identifies hidden devices by sniffing and collecting encrypted wireless traffic over the air. Then, when the user walks around the perimeter of the area, it estimates the location of each recognized device in relation to the user.

The localization module, for its part, combines signal strength measurements contained in 802.11 packets (called Received Signal Strength Indicator or RSSI) with relative user position determined by mobile phone visual inertial odometry (VIO) information.

On Apple’s iOS devices, for example, positional tracking is accomplished by ARKit, a developer API that allows developers to create augmented reality experiences by utilizing the phone’s camera, CPU, GPU, and motion sensors.

“As the user approaches each device, the RSSI values corresponding to those data points grow, and subsequently decrease as she moves away from the device,” the researchers explained. “Lumos estimates the position of each device based on spatial observations of RSSI values and their fluctuations.”

Furthermore, Lumos can locate IoT devices regardless of the user’s walking speed. A fingerprinting module is also included, which analyzes collected 802.11 traffic patterns using a machine learning model to identify devices based on MAC addresses.

The study tested Lumos on 44 distinct IoT devices of various sorts, models, and brands in six different contexts, discovering that it can detect concealed devices with 95% accuracy and locate them with a median inaccuracy of 1.5m in a two-bedroom, 1000 sq.ft. apartment in 30 minutes.

However, an expert attacker can use techniques such as MAC address randomization to avoid detection and avoid localisation by randomly changing the transmit power of the devices.

“Lumos can possibly generalize across numerous device brands and models provided it has encountered at least one device with comparable behavior in the training phase,” the researchers noted, pointing out how the system can even recognize unprofiled devices.

The post Lumos System Can Find Hidden Cameras and IoT Devices in Your Airbnb or Hotel Room appeared first on B6G.NET| for all information technology.

If your site is not configured and optimized for mobile devices, you will lose many visitors and may even lose your work soon, so make it a precedence for your work. According to a study conducted by internetretailer.com, speed in loading pages is the most preferred by Internet surfers using smart phones 16% of browsers By phones, they do not return to the site if it is slow.

Before going into these tips, I advise programmers to read this reference from Google

Tips to make your site compatible with smartphones:

• Simplify the design and make it proportional

The design for smartphones does not have to be like the design for ordinary large-screen devices, it is better to make your design for smartphones one column for the visitor to browse vertically, not to go right or left, this would distract the visitor, so it is always better to speed it up one column. The small screen should not display a lot of information that would distract the reader through the mobile phone.

1. There are several ways to do this either by purchasing optimized and ready-made themes , or add- ons that will improve your blog whether the site is based on a blog.
2. Or rely on Google’s reference to build the style from scratch and improve it for mobile. I recommend this method for programmers.

• Content Optimization

Know what the visitor needs from your site. Visitors to the B6g.net News usually need to browse the news quickly, and they can only read the headlines, so do not make them read the entire topic and put a lot of text on the screen. Give them addresses. Whether they find that the title attracts them, take them to another page, which is the topic page or The news.

• Make navigation easier

Make the navigation between the sections of your site easier and better, visitors do not want a thousand pages and a thousand links for each section, and it is always preferable to make the most important sections on the start page of the site.

Do not forget that phones have a small screen, so do not make the important links too small so that the visitor cannot click on them, or press five links together due to their smallness. Larger or use rather large icons to move between sections.

Use the moving menu (as it is in the world of technology) whether there are a lot of sections you want to show the visitor.

• Make your site faster

Whether you successfully made the first point (simplifying the design), this will really do the job and will facilitate and speed up browsing, of course Flash I advise against using it moreover to that it is not compatible with the SEO of your site, it also causes you problems browsing by phone and will slow down Browsing in general, do not use too much JavaScript either, an alternative is to use HTML 5 , which everyone likes.

Large images and high resolution also cause slow browsing, think are these images very important, provided not, how much you reduce them or even cancel them whether they are not important. Use the page speed that I talked about in a previous article of mine to make Optimized images.

• Make your website or business name visible

You should not forget the logo of your site and make it visible in browsing through phones, this will add a good impression to the visitor who visits your site for the first time, whether he sees it compatible with his device and fast, he will always prefer you over any other competitor when he is looking for something particular.

These are some simple tips for website owners, but whether you are a programmer, I advise you to look at the Google guide that explains and details more.

The post Make your site compatible with smartphones appeared first on B6G.NET| for all information technology.

The results show that NVIDIA remains the main AI accelerator, in terms of data center and widely available offerings.

David Kanter, CEO of MLCommons , the body behind MLPerf, said: “It has been an extraordinary effort on the part of the machine learning community, with so many new entrants and a tremendous increase in the number and diversity of submissions. I am particularly excited to see increased adoption of power and energy measurements, highlighting the industry’s interest in efficient AI.”

The MLPerf benchmark is held four times a year, with inference results in the first and third quarters and training results in the moment and fourth quarters. Of the two, model training is more computationally intensive and tends to fall into the realm of HPC; inference is less so, but it is still demanding. The latest round of inference had three different benchmarks: Inference v2.0 (data center and edge); Mobile v2.0 (mobile phones); and Tiny v0.7 (IoT). MLPerf divides exercises into divisions and categories to make comparisons between systems fairer and easier, as shown on the next slide.

NVIDIA was, again, the company that obtained the best results, in most tests. Quacomm had strong results, particularly in edge AI applications. Its Qualcomm Cloud AI 100 accelerator is designed not only to have good performance, but also to be energy efficient, a quality that was evident during the tests.

During an NVIDIA briefing with media and analysts, David Salvator, Product Manager for AI Inference and Cloud, recognized the great power of Qualcomm. “There are a couple of places in CNN-type networks where, frankly, Qualcomm has offered a pretty good solution when it comes to efficiency. That said, we outperformed them on both workloads and, in the case of SSD-Large, by a factor of about three or four. A really substantial performance difference, whether you put it in the context of how many servers it would take to receive equivalent performance, which really reduces their advantage per watt.”

NVIDIA’s briefing also focused on its latest Jetson AGX ORIN device and its edge performance. Software was once again a key driver of performance gains. Also featured was NVIDIA’s Triton software platform, which was used with both NVIDIA-based systems and presentations based on AWS instances that use their Inferentia processor instead of NVIDIA accelerators.

Intel, which participated in the Closed Inference Division last call, did not this time; instead, it opted for the Open Split, which allows for greater flexibility of hardware and system components.

There have been a couple of changes to the latest MLPerf Inference components and procedure. One of them was to shorten the time needed to run the tests. As explained on the MLPerf website, “we made a rule change that allows each benchmark test, more or less, to run in less than 10 minutes. And that took a lot of statistical analysis and work to get correct. But this has shortened the execution time of some of the benchmarks running on lower performance systems. You know, there are people who are presenting on Raspberry Pi. And this allows them to do it in a much more well timed manner.”

The post NVIDIA Dominates Again in MLPerf Inference Tests appeared first on B6G.NET| for all information technology.