ASSERT Yourself!

Someone once said, "To know oneself, one should assert onself." Researchers at the University of Texas at Dallas Distributed Systems Lab, took this notion and created an Advanced wireleSS Environment Research Testbed. ASSERT provides users with the true ability to know oneself. In many instances, simulation allows for rapid development and testing, but it does not provide a realistic RF environment. To compensate for this field experiments are performed yet when problems are encountered, it can be difficult to located and correct them. ASSERT provides users with the advantage of simulations and field experiments without the suffering and disadvantages of both! The goal of this testbed is to provide experimenter with the ability to run controlled, repeatable experiments that would have a similar performance to that of field experiments without the associated deployment and travel costs.

ASSERT has been a work in progress for the past two years and currently offers a forty site testbed. As wireless networking is becoming more pervasive there has been a greater desire to develop communication hardware and protocol stacks that have a number of desirable properties like increased throughput, reduced latency, reduced energy consumption, quality of service security, etc. ASSERT is a true emulation testbed. Field testing requires extensive time and energy. Changes to the location of devices require physical movement of the equipment and measurement devices from place to place making repeatability a difficult task. 

The hardware for each site consists of two mated boards, the Site board and the RF board. Each Site board consists of an ARM 200MHz processor, 256MB of on board RAM, a Xilinx FPGA, support for an external clock and various peripheries such as Ethernet and RS-232. The RF board consists of 17 SMA coaxial connectors, 16 of which are used to connect to other Site-RF board pairs and 1 dedicated to the Unit Under Test (UUT). The current system integrates the Crossbow/MEMSIC MICA2 (900MHz) Mote platform with gateways as UUTs, however any device operating in the 700MHz to 1100MHz range is natively supported. 

ASSERT emulates the distance between multiple devices by using variable attenuation - this means that identical tests can be performed in the lab compared to the field. ASSERT also allows users to conduct experiments in licensed frequencies like the cellular service bands without interfering with the services offered by the owners of these licensed bands. By changing signal attenuation levels, using programmable attenuators and amplifiers on the path between nodes, UT researchers emulated the formation/disruption of wireless links and node mobility. Using power dividers/ combiners they are able to emulate complex, multi-hop wireless sensor networks. As all communication is over coaxial cables, the network is immune to interference from other wireless networks, or allows the flexibility for users to inject the desired amount of noise into the links.

The ASSERT software performs a variety of tasks including monitoring for faults, efficiently allocating resources and providing an easy to use interface. ASSERT currently consists of forty nodes and is designed to scale to at least one thousand nodes without any design changes. Researchers at UT Dallas are converting the UI from the current small Java program to a web based application so that the experiment set up and data gathering is done by the experimenter from their browser. Through the sophisticated custom hardware and easy-to-use control software ASSERT has many valuable features that allow it to reduce the cost of testing wireless networking protocols at scale. This device independent platform for developing and testing wireless network protocols has taken experimentation to a new level!

For more information on the MICA2, click here.
For more information on ASSERT, click here.

The Contiki Operating System - Contiki 2.4 Released

In September we announced that the Contiki team had developed a port for Crossbow's MICAz platform. We are excited to inform our users that the team is proud to announce the release of version 2.4 of the Contiki operating system! Contiki 2.4 brings a number of new improvements over previous versions, several bugfixes, and an overall improved stability of the system. The low-power wireless MAC protocols have gotten an overhaul, improving power-efficiency and improved collision and interference handling. The COOJA/MSPsim simulation environment has received a significant speedup. Many improvements and bugfixes has been made to the uIP code as well as the SICSlowpan implementaion of 6lowpan IPv6-over-802.15.4.

See the changelog for full details and go to the download section to download the 2.4 release!

A Compass in Every Smartphone - Courtesy of MEMSIC

MEMSIC's capability to address new applications due to the performance vs. price points that their sensors can achieve was recently featured in IEEE Spectrum. Here is an excerpt from the article:

An old axiom says that in order to know where you're going, you first have to know where you are. To that, add that you should know which way you're facing. Makers of wireless handsets, having already installed GPS receivers, are poised to flood the market with phones containing tiny electronic compasses that allow the gadget to sense exactly what direction it's facing.

According to electronics industry analysis firm iSuppli Corp., in El Segundo, Calif., we can expect to see an increasing number of cellphones containing electronic compasses. Analysts predict the market for magnetic compass sensors will grow from the 8.7 million manufactured in 2008 to more than 540 million in 2013.

Underlying the jump is a combination of consumer demand for the whiz-bang feature of the moment and a steep drop in price fueled by manufacturing process improvements.

"Navigation is emerging as a must-have feature in smartphones," says Richard Dixon, a senior analyst at iSuppli. Consumers are already looking to their phones for turn-by-turn directions. With the addition of compasses, the beginnings of location-based augmented reality are also emerging, wherein a street map or even the phone's camera image could be overlaid with highly detailed information about what's in front of you.

The component that handset makers are exploiting to make these feats possible is the three-axis magnetometer. The sensor system's job is to home in on Earth's magnetic field and use that as a reference for determining the handset's orientation along the x-, y-, and z-axes. Three axes are important "because that third sensor allows the handheld device to correct for the orientation of Earth's magnetic field at a given location, as well as the relative position of the device," says Mark Laich, vice president of worldwide sales at Memsic, a maker of electronic compasses based in Andover, Mass. "Otherwise users would have to hold the phone precisely parallel to the ground or in some other position that may not correspond to how they normally use it."

Like the rest of the electronics industry, electronic compass makers have been making their products smaller, cheaper, and more energy efficient. And they reached a watershed moment in 2009, when the price of three-axis magnetometers dipped below US $1 per device. "Six months ago, magnetic sensors were selling at $1.50 each, but the price is coming down rapidly," says Laich. "They're now at the sub-$0.50 point and look to get even cheaper as production volumes increase."

Memsic and other new entrants into the magnetometer market for mobile devices mostly sell sensors that take advantage of amorphous magnetic alloys whose resistances change when acted upon by the planet's magnetic field. Memsic, for example, uses three chips, each containing a thin film of the alloy. The chips, which are set at right angles to each other, are incorporated into a Wheatstone bridge circuit so that when you move the phone, the change in resistance resulting from the perceived change in the strength or direction of the magnetic field is output as a change in voltage.

Aichi Steel, Honeywell, Memsic, and Sensitec have all developed magnetoresistive sensors in an effort to take a bite out of AKM Semiconductor's 95 percent share of the mobile phone compass market. AKM, in San Jose, Calif., uses a technology that relies on the Hall effect, in which electrons racing through a conductor in the presence of a magnetic field are forced to one edge of the conductor, creating a voltage difference whose orientation is perpendicular to the magnetic field. AKM's competitors claim their devices are power misers compared with Hall effect sensors, a really big deal in palm-size devices that barely have room for the batteries they stow. The companies also say that their sensors are more sensitive and have quicker response times.

The competition is making important inroads. Geneva-based STMicroelectronics, a leading accelerometer maker, chose Honeywell's magnetoresistive technology over Hall sensors to make the first single-package combination accelerometer-magnetometer on the market. One reason ST cited was that the device provides the same sensitivity in all three axes without the need for the magnetic flux concentrators that Hall effect sensors require. And as for the degaussing issue, ST found it required only a matter of microamperes.