EE Times
04/11/2001

Soft radio runs into hard standards

By Brad Brannon, Senior Systems Engineer,
Chris Cloninger, Systems Applications Engineer
Analog Devices Inc.
 
Software-defined radio (SDR) makes it very easy to rapidly evolve communication services. And it enables new technologies such as adaptive antenna arrays, which can significantly increase caller capacity within a cell site by utilizing spectral diversity.

Software radio is not limited to cellular applications. Already other wireless applications are being explored, including wireless LAN, Bluetooth-based personal-area networks and even standard commercial FM broadcasts as they move to new digital formats. In general, any areas where wireless standards are evolving and hardware may quickly be rendered obsolete are potential adopters of SDR techniques. However, the way is not at all clear yet.

Probably the biggest roadblock is the rigorous expectations placed on these systems. Most transmission standards were written with a traditional, fixed-air interface in mind. These standards evolved to avoid the weaknesses and enhance the strengths of traditional architectures. If it is possible to architect a system that exploits the strengths of software radio, then full deployment could occur much sooner.

Market gaps
Early on, Analog Devices' engineers recognized that in order for SDR developments to begin, various gaps that existed in the marketplace must be filled. These gaps included not only the inadequate performance levels in mixed-signal ICs, but also the lack of digital equivalents of mixers, synthesizers and filters used in a transceiver signal chain. Thus, Analog Devices added digital receive and transmit signal processing to its development efforts. Today, the company's SoftCell chip set consists of high-speed, wide dynamic-range data converters, receive signal processors and transmit signal processors, all of which grew out of this focused effort. These components, coupled with a high-speed digital signal processor (DSP), not only address demanding analog performance requirements, but also provide the necessary versatility to be reprogrammed in order to meet the filtering and modulation requirements of any air standard.

The receiver front end of a software-defined radio consists of an analog RF down-converter that converts the desired signal band to a convenient intermediate frequency (IF) for digitization. The down conversion is followed by a high-performance analog-to-digital (A/D) converter such as the AD6644, which digitizes the IF spectrum. The A/D converter output is processed with the AD6624, which is a quad receive signal processor (RSP) that is responsible for tuning and channel filtering. The output of the RSP consists of a channel-filtered digital IF signal requiring only demodulation that the DSP provides. In the transmit direction, the DSP sends modulated digital data to a transmit signal processor (TSP) like the AD6622, which modulates the digital carrier. Data is then converted to the analog domain using a high-performance digital-to-analog (D/A) converter such as the AD9772A, where it is mixed up to the desired RF frequency.

There are two key specifications for an A/D converter used in wireless applications. The first specification is the signal-to-noise ratio (S/N), which is useful in computing the overall sensitivity of a receiver. Although A/D converters are voltage devices and NF is a power measurement, an equivalent noise figure (NF) can be calculated based on a full-scale referenced S/N.

The noise in the signal-to-noise ratio consists of thermal noise and quantization noise (the noise generated when a signal is digitized). Other factors that affect A/D converter performance include nonideal quantization (imperfect A/D conversion) and aperture uncertainty (a wideband phase noise on the clock) in the sample clock.

The second specification of interest for an A/D converter is spurious-free dynamic range (SFDR). Spurious response consists of the second, third and higher-order harmonics. SFDR primarily determines how large an in-band or out-of-band interferer can be. If an interferer is sufficiently large, the A/D converter may generate a harmonic that appears as a cochannel interferer of the desired signal. Spurious performance is very closely related to the air interface standard and is often the limiting factor in a receiver. For SDR, the distortion (noise and/or spurious) of the A/D converter must not impact overall system performance. Therefore, the A/D converter must be carefully chosen to meet performance expectations.

The receive signal processor is a numeric preprocessor for the DSP. The purpose of the RSP is to replace a local oscillator, quadrature mixer, channel-select filter and data decimation. In a multicarrier application, the RSP replaces the analog selectivity and tuning functions with digital equivalents. An RSP sets the receiver apart from traditional receivers because all channel characteristics are now programmable. This includes data rate, channel bandwidth and channel shape. In addition to the unlimited selection of channel characteristics, a digital filter will perform exactly alike across all boards, unlike analog solutions that always have tolerances.

There are several important specifications to consider when selecting an RSP. First, the device must be capable of handling the high data rates required by the interface (A/D converter). Since the A/D converter's sample rate determines the bandwidth that can be processed according to the Nyquist theory, the RSP must be capable of handling the same data rate. The Nyquist theory states that the sample rate must be at least twice the bandwidth of signals being processed to recover the information contained therein. In practical systems, the sample rate is frequently run three times faster than the bandwidth of the signals being received to allow for antialiasing filter response.

The next specification of interest is the internal and external bus widths. They must be wide enough to preserve the signal integrity. Although the A/D converter may only be 14 bits wide, oversampling followed by narrow-band filtering improves the effective S/N (processing gain) of the A/D converter by up to 30 dB for some air interface standards. This is the equivalent of 5 more bits. Therefore, internal bus widths must have the equivalent of at least 19 bits to preserve signal integrity.

Since a large portion of any air interface is the channel bandwidth and shape, it is important that the RSP include flexible decimation and filtering configurations that allow for a wide variety of data rates and filter bandwidths. Fixed filter widths and shapes should be avoided since they limit channel bandwidths and often preclude raised root cosine filtering.

Finally, one of the most unique features of an RSP is the ability to select the desired analog frequency precisely and rapidly. Most RSPs have a 32-bit NCO that provides a frequency resolution of about 1 in 4 billion. This is usually much more than adequate, and gives great flexibility in frequency selection. In addition to the flexibility, frequency hopping is greatly simplified. Since no phase-locked loop (PLL) is used, changing frequencies is instantaneous. This can be beneficial in applications where hopping must occur within the guard band of a few symbols.

The transmit signal processor is a numeric post-processor for the DSP. The purpose of the TSP is to replace the first local oscillator, quadrature modulator, channel filtering and data interpolation. Like the RSP, the TSP sets the transmitter apart from traditional designs because all channel characteristics are now programmable. This includes data rate, channel bandwidth and channel shape. Since modulation, channel filtering and other aspects of the modulation are done digitally, the filters will always perform exactly alike across all boards, unlike analog solutions that always have tolerances.

There are several specifications that are important when selecting a TSP. First, the device must be capable of generating data at the rates required to preserve the Nyquist bandwidth over the spectrum of interest. As with the A/D converter's sample rate, the sample rate of the D/A converter determines how much spectrum can be faithfully generated by the D/A converter. Therefore, the TSP must be capable of generating data at least twice as fast as the band of interest and preferably three times faster as reasoned earlier for antialiasing filter response.

Similar to RSPs, the bus widths are also important, yet for different reasons. In the transmit direction, there are two different issues. If the TSP is used in a single-channel mode, then the issue is simply quantization and thermal noise. It is usually not desirable to transmit excess in-band or out-of-band noise, since this wastes valuable transmitter efficiency. In a multicarrier application, the concern is slightly different. Here, many channels would be digitally summed before reconstruction with a D/A converter. Therefore, each time the number of channels is doubled, an additional bit should be added so that the dynamic range is not taken from one channel when another is added.

Finally, the ability to frequency hop is vital. Since a TSP implements frequency control with an NCO and a mixer, frequency hopping can be very fast, allowing the implementation of the most demanding hopping applications as found in the GSM specification.

Basically, a D/A converter is similar to an A/D converter when considering performance requirements. Therefore, the first specification of interest is the signal-to-noise ratio. As with an A/D converter, S/N is primarily determined by quantization and thermal noise. If either is too large, then the noise figure of the D/A converter will begin to contribute to the overall signal chain noise. While noise is not necessarily a concern spectrally, the issue does become important when the D/A converter is used to reconstruct multiple signals. In this case, the D/A converter output signal swing ("power") is shared among the carriers. The theoretical signal-to-noise ratio of a D/A converter is determined by the same set of equations that govern an A/D converter, and the noise figure can be derived given a specific S/N.

All material on this site Copyright © 2001 CMP Media Inc.