Blog     e-Newsletter       Resource Library      Directories      Webinars     Apps
AFCEA logo

Superconductors Advance Signal Filtering

May 2002
By Robert K. Ackerman
E-mail About the Author

Unhindered electron flows clear the way for improved communications and intelligence applications.

High-temperature superconducting materials discovered only 15 years ago now are enabling signal filters that can achieve performance levels not even approached by conventional filters. Virtually any commercial or military system that must pull weak radio frequency signals out of background noise can benefit from the new technology.

These filters are especially useful in applications that mandate extremely narrowband filters, such as in 1-gigahertz signaling that requires a bandpass filter only a few megahertz wide. Another key application is for broader-band filters that have sharp skirts, which determine how fast the filter transmission drops off as a function of frequency.

The high-temperature semiconductor filters are built using fabrication technologies common to semiconductors. They do not require exotic and expensive cooling, and they can be incorporated easily into existing radio frequency systems as a front-end attachment to a receiving antenna.

In the commercial world, more than 1,000 units already are in use in cellular telephone stations. In the military arena, high-temperature superconductor filters have been tested or are in use aboard U.S. Navy ships, U.S. Air Force aircraft, and intelligence sensor platforms.

The U.S. Navy is looking to incorporate them into several Navy systems, states Anna M. Leese de Escobar, a high-temperature superconductor scientist at the Space and Naval Warfare (SPAWAR) Systems Center in San Diego. She notes that the most interested program office is SPAWAR’s Signals Intelligence (SIGINT) Office, which is looking at using tunable versions of the filters. Other significant potential applications include software-programmable radios and submarine communications.

The Navy has evaluated filters manufactured by STI, Santa Barbara, California, and Conductus Inc., Sunnyvale, California. Shipboard tests have validated many of the performance predictions for the devices, Leese de Escobar relates. One test in a SIGINT application “worked great,” she remarks. “It performed as predicted and beautifully.

“To my experience, everyone who has ever had the opportunity to evaluate these filters has been really happy with the results,” she continues. “That is why the Navy is paying attention to the technology.”

Dr. Randy Simon, vice president of government business and chief technical officer at Conductus, explains that his company is focusing on high-performance filtering for radio equipment, particularly communications systems in both the commercial and the military arenas. The commercial sector can benefit in cellular or wireless communications systems, whereas the military can see substantial improvement in its tactical communications and intelligence gathering.

In commercial cellular applications, the filters are installed in base stations where individual telephone signals are received. These filters actually improve the sensitivity of the base station, which increases the station’s sensitivity to each individual cellular telephone transmission. They also diminish interference in the signal environment that degrades cellular service.

In the military arena, the filters serve two basic applications built around the concept of being able to detect exceedingly weak or specific signals. Signal strength may be reduced by terrain, weather, logistical limitations, adversary actions or other aspects of the battlefield environment. Similarly, a receiver may need to pull a signal out of an extremely complex radio frequency environment, especially in the midst of multipurpose platforms and vehicles all emitting and receiving. Combining both of these scenarios presents the double challenge of pulling weak signals out of a particularly noisy environment.

The new high-temperature superconductor technology eliminates a performance trade-off that has been standard in traditional systems. Before the advent of these filters, receivers had to be designed either for sensitivity to weak signals or for the capability to screen out unwanted or interfering signals. More-sensitive receivers were susceptible to noise and other interference, and less-sensitive receivers that selectively screened out noise and unwanted signals also missed weaker wanted signals.

That trade-off no longer is necessary with the new type of filters, Simon states. The technology permits a receiver to be as selective and as sensitive as possible, he says. These devices also have better defined skirts and reduced insertion loss, which determines how much of the signal is wasted in heating up the filter.

The new filters are built around yttrium barium copper oxide (YBCO) superconductors. This compound is constructed in thin-film form on a wafer fabricated in the same manner as a semiconductor chip. Even though the devices are fabricated using conventional photolithography and etching techniques, the level of complexity is considerably less than that of computer chips. The superconductor chips feature internal structures that are much larger, and thus easier to design and fabricate, than those of their conventional metal oxide semiconductor counterparts. The 1-inch superconductor chip can replace a conventional filter that is about the size of a cigar box, and its superconducting material is about “1,000 times less lossy” than cryocooled copper, Simon claims.

The YBCO materials do not require special cryogenic cooling to achieve superconductivity. Instead, a closed-cycle mechanical cooler such as a Stirling or Gifford-McMahon refrigerator can reduce their temperature below the threshold. This eliminates the need for a user to resupply cryogenic coolant to a device in the field, which also enhances each unit’s portability and ruggedness. The YBCO materials achieve superconductivity at about 90 Kelvin (–297 degrees Fahrenheit), but the superconductivity properties are very tenuous at this temperature. The cooling system operates at about 70 Kelvin (–333 degrees Fahrenheit) to ensure optimum superconducting performance.

In their role as signal filters, these materials effectively are loss-free, Simon notes. Because they do not lose vital electronic or radio frequency signals, they no longer consume some of the wanted signal as they screen out unwanted interference.

“You can build a radio receiver that is as sensitive to weak signals as you like and still build a filter into that system that actually is more selective than any filter you could build using alternative technologies,” he declares.

The technology also permits construction of filters that would have been all but impossible using conventional means. Simon notes that his firm has built a filter that has the equivalent performance of a 50-pole conventional filter. These poles represent stages of filtration, and a higher number of poles leads to a more selective filter with steeper rejection skirts. Conventional filters tend to be limited to no more than 10 poles, as higher numbers produce increasingly greater losses. The 50-pole superconductor filter, which Simon describes as “a brick wall,” has only minimal losses, he claims.

For radio receivers, the superconductor filters serve as a front end connected to the antenna between it and the receiver. The filter can be attached to an existing radio receiver as an appliqué without re-engineering the original unit. Simon notes that eventually radios can be designed with the superconducting filter embedded in the unit.

Because the filter is placed between the antenna and the receiver, it filters signals before they are amplified. By amplifying the filtered signal prior to its entering the receiver, the clarity of the signal is enhanced. Amplifying an incoming signal before filtering boosts interference levels and can result in a filter being saturated by an out-of-band signal.

For military communicators, the result of using the new filter can be range extension. An extremely sensitive receiver would benefit only slightly; however, an average system could see significant improvements. These could range from a 15 percent improvement up to a 100 percent improvement, depending on both the quality of the original radio equipment and area environmental conditions. Simon believes that a typical range improvement would be about 25 to 50 percent.

Leese de Escobar notes that software programmable radios in particular are built around reconfigurability and simultaneous multiple signal processing. One program in the Office of Naval Research involves a low-temperature superconducting digital correlating receiver for this type of radio. This concept, which aims to exclude co-site interference by correlating signals faster and better, includes high-temperature superconducting filters.

Research efforts for submarine communications focus on several issues, especially in exploiting the small physical footprint of the high-temperature superconductor chips. Space, weight and power all are major concerns aboard a submarine, Leese de Escobar observes.

Another advantage, which can have both commercial and military applications, is the ability to provide steep rejection skirts. Simon notes that many broadband commercial wireless systems abut other frequencies, and their users want as much frequency rejection as possible just outside of their own band. This looms as an issue especially with third-generation, or 3G, cellular systems. In Japan, one of the primary 3G frequency bands is adjacent to that country’s personal handyphone system, or PHS. Users of PHS telephones are broadcasting right on the edge of the new 3G band, and this poses the possibility of significant interference for the multimedia wireless band. Maintaining signal separation will require filters that let in signals across the entire width of the 3G band while simultaneously excluding all adjacent PHS signals, and Simon states that superconductor filters are well-suited for the task.

Simon allows that superconductor filters are not needed for all filter applications. Most filter applications that fall in between the narrowband use and the broadband steep skirt operation, and that do not require large amounts of signal rejection, would not benefit appreciably from the superconductor technology.

One of the biggest technological hurdles involved taming the exotic materials, Simon observes. Following the materials’ discovery in the late 1980s, it took about five years before their properties were understood and they were reproducible in reliable constructs. Subsequent challenges included packaging the technology so that it would work effectively and reliably on the system level, he continues.

One ongoing development aims to increase the types of receiver system elements that can incorporate these filters, Simon notes. These items eventually might include the mixers, the splitters and other passive radio frequency elements.

Tunable superconductor filters are the next evolutionary step. The operating frequency of these devices would be adjustable, which would be especially useful in scanning applications such as SIGINT.

Simon allows that his firm is able to produce devices that shift filter frequencies by about 20 to 30 percent. That figure is increasing, and he predicts the ability to shift by a factor of two or more. Conventional technology does not permit high-performance tunable filters, he says. Any tunability comes at the cost of large losses in the filters. A key goal of tunable superconductor filters is to maintain the advantages inherent in the technology, he adds.

The capabilities of tunable filter technology should gradually increase in several different dimensions over time, Simon predicts. These dimensions include the extent of tunability—the width of the tunable range—the rapidity at which the filters can be tuned, constraints such as complexity and narrowness on the sorts of filters, and eventually the shape of the filters.

Leese de Escobar foresees a future with the devices branching into a number of related applications. Tunable filters can be combined with low-noise amplifiers, with both sharing the same cooling unit, for a system that combines sensitivity with selectivity. The next step will be analog-to-digital filters. Eventually, superconducting antennas may emerge when complex cooling challenges are overcome.