The RFID industry depends on the semiconductor manufacturers who make the chip that is the brains of the tag to keep the performance improvements and innovations coming. I talked to several chip makers at this year’s RFID Journal LIVE! and they did not disappoint.
Alien Technology introduced a new reader, the 9680, positioned between their high end 9900 and their highly popular and compact 9650 reader. The 9680 is designed to fulfill the market’s desire for that intermediate position — a four-antenna reader in a thin chassis at a more attractive price point. They also introduced 3 new tags: the GT (Garment tag) for hanging and stacked garments; HT (HiScan tag), a small form factor (15X50 mm â 1.6” X .6”) with very high backscatter to work well with low power readers, such as handhelds, with tightly packed goods, such as a stack of jeans; and the BAT tag, designed for use on automotive batteries, but useful for many high-dielectric objects such as glass, metal, or water.
NXP’s UCODE 7
NXP introduced their UCODE 7 chip, which they are touting as the best-in-class EPC gen2 chip. It does have excellent published performance specs, with a read sensitivity of -21dBm and write sensitivity of -16dBm. The read sensitivity represents a 3.5 dBm improvement over their previous design, which translates to about a 45% improvement in performance. The write sensitivity was an even greater jump of -8dBm. They also said the tag is not as sensitive to parasitic capacitance which allows them to make smaller tags with very good performance, thereby penetrating new markets such as mascara, perfumes, and jewelry. They also decreased encoding time by over 10X, from approximately 15 ms down to about 1 ms for writing a 16 bit word. On top of this, they introduced a massively parallel version of the traditional serially processed bulk encoding, as well as automatic self-serialization. The chip also senses the level of power when it’s being written, to ensure any write that is started is reliably completed (or else it will not start the write). NXP works closely with tag partners from Avery to Motorola to Zebra and others, creating new high performance tags targeted at specific applications.
Impinj displayed several innovative products. Their xArray reader, built using Impinj’s top-of-the-line Speedway Revolution reader, has a phased array of antennas, whose beam can be steered in any direction in a ~40’ diameter circle. This enables it to sense the location of passive tags within that read zone to an accuracy of about ± 1 meter (3 feet). This gives RTLS capabilities at passive tag prices — a real breakthrough compared with active RTLS tag prices. Of course you will need a much denser array of readers compared with an active RTLS system. This could make sense for certain applications, such as a scenario where you have lots of items but a modest physical footprint (e.g. a boutique clothing store). Impinj’s customers have used the xArray in retail settings and find they are getting a few percentage points of improved inventory accuracy over using RFID with handheld readers. That’s pretty impressive considering RFID with handhelds has been shown to be one of the most accurate ways of counting inventory, improving perpetual inventory accuracy by 20%-30% over the traditional manual cycle counting. In addition to location, the xArray can sense direct too, which is very useful when loading or unloading a truck or tracking stock movements through a doorway. The xArray won the “Best in Show” award at this year’s RFID LIVE.
Impinj also is addressing the trend towards integrating RFID readers into all kinds of products, equipment, and devices (see The Blank Canvas of RFID: RFID in Product Designs) like in the engine of a big mining truck, or a medical machine to administer anesthesia, or a soda dispensing machine, or a toy. The designers of these products are not RFID or RF experts. To make their job easier, Impinj has a small form factor ‘reader-in-a-box’ — just add power and an antenna. And if you want help designing the antenna, they can help with that as well.
It’s not often that something really new and potentially game changing comes along. So, it was exciting to hear about new self-tuning technology from RFMicron. Those who are familiar with RFID understand that the impedance of an RFID tag changes when it is put near different types of materials, thereby detuning the tag, leading to a loss of energy, performance, and read range — potentially dramatically. Liquids and metals are well-known offenders.
To overcome this problem, chip-tag-antenna combinations are tuned and designed for specific applications — i.e. use on specific materials or types of products. Even this is not ideal because things change — a corrugated cardboard shipping container may get damp, causing detuning. Plus the proximity to other objects is constantly changing. Do you design the tag to work best on a single bottle of shampoo in open space, or on one that is within a case in a pallet of bottles?
What RFMicron has done is add some circuitry that automatically detects and compensates for local detuning effects. Within limits, it automatically tunes itself to the best impedance and hence performance. This means a single design can be used across a much broader range of materials.1 Furthermore, there has always been a tradeoff when designing a UHF tag that can work globally across the broad range of different UHF frequencies used around the world (i.e. ~ 865Mhz to 958Mhz). The so-called broadband antenna works modestly well (not horrible) across all geographies. Or you have a more narrowband design that works well in one geography but not others. By auto-tuning, a single antenna design can adjust itself to work as well as a narrowband design, but across all global UHF frequencies.
A Low-Cost Batteryless Sensor
Where things get really interesting with the RFMicron concept is that the tag can be used as a sensor. This is most evident when it comes to moisture sensing. Once the tag is calibrated, changes in impedance will correlate to changes in moisture content of the object that tag is on. Thus the tag can measure these changes and transmit moisture measurement information upon demand. Differentmaterials can be added on the tag, turning it into various kinds of sensors. These would include materials that change their own impedance based on changes in temperature, pressure, proximity, or other conditions that you want to sense. Potential uses for this are truly enormous in variety. The ultra-low cost compared with traditional sensors opens up many new applications. Add a small battery and it can become a data-logging sensor that records changes over time.
High Speed Hang Tag Encoding
One of the more visually impressive displays was the high speed hang tag encoding machine (you can see a short 12 second video I shot here). This system is built for speed, able to encode up to 75,000 tags/hour. That’s about 21 tags per second, as they fly by at up to 600 feet per minute. This is done using a high speed camera and four RFID readers in series. The camera, which can shoot up to 240 frames/second, takes a picture of the tag, capturing its barcode and other information. The first of the four readers validates that the tag is good. The next two each do partial writes of the 96 bits of data (sharing the job because there isn’t enough time for one alone). The fourth and final reader fills in any gaps missed by the other two readers. The production version will have a diverter to remove any tags not successfully encoded. The system is built by Kirk-Rudy, who has been designing and manufacturing paper-handling, printing, and packaging systems since 1967. They partnered with Impinj (who supplied the readers and RFID expertise) and The Solutions Group (who wrote the software).
A Better WiFi RTLS — Finally!
Redpine Signals is another company trying to change the game. Over the years, they have developed a variety of innovative 802.11 chipsets, modules, and systems. Recently, Redpine developed a WiFi-based RTLS solution stack (chips, modules, software) that appears to overcome some of the more onerous limitations of previous versions of this technology — in particular short battery life and low location accuracy.
WiSeMote,™ Redpine’s core technology, is comprised of tags (Wireless Secure Motes) and an overlay network with SuperMote™ and NetworkMote.™ These components offer the potential to reduce the cost of the infrastructure and deployment costs while improving accuracy and battery life. Redpine said the accuracy is ±1 meter and has battery life of 5 years (if you wake up the device every 5 minutes). Accuracy numbers for RTLS always have to take into account factors like the environment and density of readers.
Having said that, I have not heard of any other vendor getting close to ±1 meter. The battery life is also hard to compare apples-to-apples without digging down to understand power consumption over time for ‘similar applications’ (which itself is an open-to-definition concept) — nevertheless, Redpine claimed that their battery life is 10X that of any other tag you’ll find. Assuming these claims are verified, this represents a game-changer for RTLS using WiFi, which should enable the growth of this market into new application areas.
All in all, for those of us deeply involved in RFID, sensors, and RTLS, this year’s crop of announcements provides plenty for the industry to be excited about.
1 Within limits. You still will need different designs for radically different material types. But you end up with fewer different designs needed to achieve broader applicability with more consistent performance across different materials. — Return to article text above
To view other articles from this issue of the brief, click here.