Modern devices get the much needed flexibility through software implementations.
A version of this article appeared in the November 2012 print edition entitled: Just a better way to radio.
Today’s terrestrial mobile radios are largely software-oriented, the result of a quiet 25-year revolution in radio design. Software-defined radio (SDR) allows the designer to provide multiple air connectivity standards on a single hardware platform, accelerating product development and upgrading quickly and easily.
The concept of SDR
The IEEE defines software-defined radio as “radio in which some or all of the physical layer functions are software-defined.” Physical layer functions include, but are not limited to, tuning, antenna diversity, down-conversion, synchronization, modulation, and demodulation. The Holy Grail for software radio designers is a device with an antenna connected directly to an analog-to-digital (A / D) converter, as shown in Figure 1 (a). In this idealized embodiment, the radio signal is immediately digitized and there are no hardware-specific components. In other words, a hardware platform handles all frequencies and all air communication standards.
The receiver of Figure 1 (a) cannot be realized in practice. Using today’s A / D converters, the displayed setting would result in an incredibly poor receiver. The A / D converter has no infinite frequency band or infinite dynamic range and all A / D converters have low noise values. For these reasons, modern terrestrial mobile radios use a bandpass filter and a low noise amplifier (LNA) in front of a conventional superheterodyne converter, as shown in Figure 1 (b). This approach provides good sensitivity and dynamic range, using components available today.
Until recently, cell phones and many terrestrial mobile radios used application-specific integrated circuits (ASICs) to implement the air link standard (also known as waveform). ASICs use many software-radio concepts, but still rely on hardware-specific implementations. Today, it is common to apply the air link standard in a programmable field array from the port (FPGA), which is a universal programmable hardware platform.
Two practical limitations of the SDR are CPU speed and memory. Like computers, new radios always need more speed and storage; for this reason, no hardware platform has endless life. The software-defined radio allows you to extend the life of the hardware platform, complete the design after the hardware is developed, and maintain the same hardware architecture while upgrading the processors and memory. Another advantage of SDR is that the software exists independently of the hardware; therefore, it can be easily transferred to new hardware designs.
Software-defined radio is rooted in military and early research funded by the US Department of Defense. A team from E-Systems, Inc. (now Raytheon) coined the term “software radio” in 1984 while developing a concept proof unit. This prototype provided programmable interference suppression using adaptive digital filters that were new at the time. The military saw the usefulness of software radios early on, mainly due to the need for interoperability between its branches and between US and NATO troops. The military is natural for software radio, as its need for interoperability, encryption and anti-jamming lends itself to software implementations.
Joseph Mitola coined the term “software-defined radio” and published the first article on the subject in 1991. In 1992, the US military launched a software radio project called SpeakEasy, which aimed to develop a radio that provided AM, FM and Forms of SINCGARS wave, from 2 MHz to 2 GHz.
According to SDR Forum (www.wirelessinnovation.org), more than 93% of cellular base stations use SDR technology, and almost 1 billion software-defined cellular phones were delivered in 2011. Almost all tactical radio stations sold to military and terrestrial mobile radios for public safety use SDR technology. The striking transition from hardware to software radios over the last 10 years is largely unknown to the public, as the functionality remains the same. In general, the user is most interested in whether his phone is 3G or 4G, which according to him is equivalent to a higher speed. But basically, software radio is just the better way to radio.