Communications, signals intelligence, radar, and electronic warfare systems designers use the latest digital technology to get the most out of one of the last vestiges of analog technology: antennas.
Gathering and sharing knowledge always have been key to victory on the battlefield. With advances in information technology, today's warfighter has access to more information than ever before. Though more information is available, the means to distribute it needs to keep up. To meet the ever-expanding flow of information in military networks, smart antenna technology, along with digital signal processing and software reprogrammability, have advanced.
Offering increased security and flexibility of communication and driving cost down are some of the primary drives behind smart antennas and digital signal processing.
|Smart antenna technology enables reliable communication for radios and other devices.|
Back when radios were new, antennas that gathered their signals were purely omnidirectional devices. Being omnidirectional is great, but not when the receiver bombarded with noise from all possible angles.
Each different signal an antenna receives interferes with the other signals, meaning a single, omnidirectional antenna can only deal with so much traffic. In response, antennas became sectorized to reduce interference with other antennas and more possible traffic.
Today, smart antennas are the norm for communications. Smart antennas are constructed as receiver arrays with smart signal processing algorithms that can identify spatial signal signature. These antennas can identify the direction of arrival of a signal and use it for beamforming, which enables the array to locate and track the antenna beam on a target. This makes them highly directional, allowing multiple signals to be handled by the array without them interfering with one another.
There are several different types of smart antennas, including switched lobe, dynamically phased array, and adaptive array. Each type of smart antenna array has its benefits and drawbacks.
The switched-lobe antenna consists of a collection of directional antennas, each of which covers a different area. The antenna that is facing a transmitting device communications with that device, and when the device moves out of the beam it begins communicating with a different antenna in the array. Its simplicity, however, leaves this antenna type vulnerable to interference.
A dynamically phased array uses several antennas to beam a signal device digitally. The array activates certain omnidirectional antennas in the array, which has a multiplying effect that forms a beam. The beam then can track a device by adjusting the gain on each antenna element and phasing the transmission of the signal in the elements.
Because the beams are formed digitally, the same array of elements can target beams at several devices and even on several frequencies. As long as two beams aren't targeting the two devices, they can make use of the same frequency. The primary drawback is the processing power required to make use of a dynamically phased array, but as processing power increases, this has become less and less of a problem.
An adaptive array is an advanced version of a dynamically phased array. An adaptive array takes many factors into account that a dynamically phased array does not, such as interfering devices and several signal paths. Adaptive arrays are capable of blocking interfering devices by reducing the signal received from antenna elements in its direction and increasing the signal in others. This enables adaptive arrays to have a high signal-to-noise ratio, which results in superior communication. It requires a large amount of processing power as well, making adaptive arrays difficult to use in large systems due to power, thermal, and size constraints.
Beamforming, or the process by which a smart antenna array creates a beam and then tracks devices, can be done in three ways. Digital beamforming, which involves digitizing the signal of each antenna element in an array; analog beamforming, where signals are not digitized until later; and hybrid beamforming, which uses analog and digital beamforming in a given array.
|The XMC-1151 by Spectrum Signal Processing is a high-speed digitizer and processing solution in a VITA 42.3 XMC form factor.|
"Digital beamforming theoretically provides optimum flexibility in antenna beam creation," says Jim West, principal engineering manager at Rockwell Collins in Cedar Rapids, Iowa. "Specific advantages include independently steered simultaneous beams and adaptive nulling of jammers."
Digital beamforming is widely seen as the sort of ideal way to perform beamforming. By converting the analog signals into digital signals early, it is easy to manipulate them and perform tasks such as nullify interfering signals and track desired signals. Still, the processing power required for pure digital beamforming is a hurdle. "Analog to digital converters with wide bandwidth and high dynamic range are required at each element," explains West.
"This results in a significant thermal and power consumption problem. There are complex calibration and alignment issues related to the large number of distributed receivers," West adds. "Moving massive amounts of data from the antenna elements to the digital beamforming processor is a major issue. As an example, a single element with a 10-bit analog-to-digital converter (ADC) running at 1000 megasamples per second would create 10 gigasamples per second for each element. Multiplying it by 1000 elements is 10 terabits per second. The beam steering computer must handle this aggregate computational load in real time."
While digital beamforming is not yet possible for the largest of smart antenna arrays, the computation problem has become less and less thanks to advances in hardware and digital signal processing. Due to the size, thermal, and sheer computational issues caused by digital beamforming, hybrid beamforming has become popular for certain systems.
"Hybrid beamforming uses analog and digital beamforming in a given antenna array," explains West. "Analog beamforming is applied to antenna elements of subarrays, such as a column of elements. The resulting subarray signals are digitized and combined using digital beamforming. This reduces the number of signals that are digitized, which eases the processing required, while maintaining some of the benefits of digital beamforming, such as multiple beam capability, and the ability to null interferers."
Hybrid beamforming is widely used in systems where digital beamforming is not a sensible solution due to the high processing requirements. By combining the relatively low processing requirements of analog beamforming with the precision and flexibility of digital beamforming, hybrid offers a compromise between the two methods.
Analog beamforming is using beamforming before converting the signal from analog to digital. This results in a signal that is not easily manipulated, making it difficult to null interference or even track a signal. The benefit of analog beamforming is the quality of the wave, as analog provides the most accurate representation. However, analog has largely fallen away to be replaced by more robust digital and hybrid systems.
Multiple input, multiple output
Multiple-input and multiple-output (MIMO) is the use of several antennas at the transmitter and receiver to improve communication. MIMO smart antenna technology offers increases in data throughput and link range without requiring increased transmit power or bandwidth.
It does this by spreading the same total transmit power that a single antenna solution would provide over a series of antennas. It does this to obtain an array gain that improves spectral efficiency (bits per second per hertz of bandwidth) or to achieve diversity gain that reduces fading, improving the link's reliability.
Digital signal processing (DSP) is the mathematical manipulation of an information signal. DSP, for military and aerospace applications, is often performed on field-programmable gate arrays (FPGAs), microprocessors, and digital signal controllers. The primary purpose of DSP is to change an analog signal into a digital one, enabling compression and transmission in digital mobile devices.
DSP is also used for radar, sonar, and any other task where a sensor collects analog information and a computer is needed to understand or transmit the information. This means DSP is required for smart antenna applications.
The challenge of signal processing for military and aerospace applications revolves around transforming raw RF signals at an antenna into useful information for whatever system they are operating in. To be effective, this sequence must be fast and efficient in converting signals from analog to digital, and in filtering out noise and other irrelevant RF energy, while processing data quickly to pull out the most important information from a flood of data.
Digital signal processing has been helped largely by the proliferation of commercial smartphones and other wireless communications devices. "In the old days, military tech would trickle down into the commercial world," says Tudor Davies, director of product management at Spectrum Signal Processing in Burnaby, British Columbia. "Nowadays it's more often the other way around."
Digital signal processing has been helped along by commercial standards enabling better communication between boards and algorithms developed in the commercial world. "One of those technologies that we're seeing is PCI Express," says Davies.
"Basically a high-speed serial protocol that has replaced the old PCI bus. PCI Express allows you to use a standard spaced approach," Davies continues. "The old method wasn't fast enough to carry the high data rates for digital signal processing, so we used other protocols for high-speed serial communications between boards. The benefit that we're seeing with these commercial standards is you can have more interoperability."
With commercial standards, it has become easier for digital signal processing on processing-intensive applications, such as digital beamforming, by splitting the burden amongst several boards. "If you want to do digital beamforming and you want to distribute that load across multiple processors, you need a high-speed data communication fabric between those processors," Davies explains.
DSP is especially important for smart antenna applications, as today's battlefield has a wide variety of devices transmitting large amounts of data at once. Good DSP is required to enable effective communication between the many information gathering and sending nodes in the field.
DSP advances have largely been in the form of hardware, enabling system designers to create systems that can handle more information in whatever fashion they desire.
"Digital downconverters have gotten faster, along with FPGAs getting faster, and A-D converters," Davies says. "It's now possible with the faster A-D converters and FPGAs to digitize at RF, depending on the frequency, to digitize signals up to 3 GHz where you used to have to downconvert that into a lower IF frequency. Then you would use a digital downconverter to FPGA to convert that into baseband."
The benefit of these improved digital downconverters is clear. "Now you can directly digitize an RF frequency up to a fairly high frequency range; that gives you a lot more bandwidth," explains Davies. "You can analyze a large swath of bandwidth in that FPGA. If you're using a smart antenna application, that means you can do more of your processing digitally, which enables you to implement adaptive algorithms in the digital domain, where it would be much harder to do with analog."
FPGAs have also seen improvements. "An FPGA is often accompanied by a microprocessor that handles interface and control aspects for the FPGA. In addition to this, FPGA resources are often be consumed to create a soft processor core, for processing tasks accomplished more easily with a high-level language," explains Rockwell Collins' West.
"These soft processors are less efficient than a microprocessor. Recently, some FPGAs have a microprocessor packaged together with the FPGA in a single device, sometimes referred to as a system on a chip (SoC)," West says. "Use of an SoC can reduce the chip count of a design, and offer benefits in size, power, and cost. It also facilitates FPGA initialization at power up and during re-programming, with the integrated microprocessor handling FPGA programming and memory access."
|The ANT-7000 C-Band Antenna by Rockwell Collins is a small, lightweight, airborne antenna designed to support high data rates.|
Software reprogrammability or the ability to change software quickly, even in the field, is becoming more sought after for many applications. Reprogrammability means different benefits for different systems, such as the ability for a radar system to switch from air-to-air to air-to-surface to the ability of a radio to adapt to different communication situations on the fly. Commercial off-the-shelf (COTS) products also have a role for several different mission types. For smart antenna technology and digital signal processing, software reprogrammability means adaptability and flexibility.
"A reprogrammable system might send updates to their users, and that can be a rapid update" says Davies. "All you have to do is change system code in flash or on a hard disk drive to do a system change or some other hard update. If all your processing is done in reprogrammable FPGAs and DSPs and general-purpose processors that's fairly easy to do."
Software reprogrammability has become more popular due to rapid advancements in processors and other computer components. While in the past special components were required to perform certain tasks, it's now possible for a COTS board with a general-purpose graphical processing unit (GPGPU) to handle digital signal processing.
Rather than requiring that same board to do one task for its entire lifespan, the board can be reprogrammed to perform a different task or to specialize in a specific aspect of DSP. "When you have a highly reprogrammable system, the initial cost might be higher than one that can only do one thing," explains Davies. "But if you can adapt that same piece of hardware to different missions by reprogramming it your total cost of ownership can be lower."
Because of the adaptive way DSP and smart antenna technology functions, software reprogramability is a natural fit with these technologies.
The benefit of software reprogram-mability is clear to scientists in the U.S. Office of Naval Research (ONR) in Arlington, Va., who launched the software reprogrammable payload (SRP) in 2012 to improve software-defined radio (SDR). SRP enables SDR to use a modular design and simplifies communication. Once on station, the operator can select options based on the current situation or reprogram the communications channel from a preselected frequency plan. The program opens up the platform to other developers, enabling a wider community to produce applications for it.
The main intersection of smart antennas, digital signal processing, and software reprogrammability is the SDR. "Software-defined radios can accommodate numerous waveform types within the basic capability of the RF components of the radio," says West. "This reduces the number of federated radios in a vehicle or aircraft as long as simultaneous operation of all the waveforms is not required. Antenna technology can be a key limitation to the full use of software defined radios. The range of operational waveforms requires antennas with conflicting requirements of small size, wide bandwidth, polarization agility, and variable gain."
SDR relies on smart antenna technology to tap into its full potential, and makes use of software reprogrammability to provide updates in the field. The future of SDR is largely dependent on smart antennas, DSP, and software reprogrammability.
Curtiss-Wright Controls Defense Solutions
Spectrum Signal Processing
Technical Communications Corp.