ARRLWeb: First Report of Software Defined Radio Working Group

To: ARRL Technology Task Force
From: Software Defined Radio Working Group
Date: 1 July 2002

The Software Defined Radio Working Group (SDRWG) was created by the ARRL Board to report to the Technology Task Force (TTF) with an initial assignment^[*] of

Leif Asbrink, SM5BSZ
Gary Barbour, AC4DL
Bob Larkin, W7PUA, Chairman
Mike Marcus, N3JMN/7J1AKO
Doug Smith, KF6DX
Gerald Youngblood, AC5OG

a) radios, or portions of radios, having some capability to externally alter the functionality, by digital program modification,

b) but excluding those radios that have only control functions, such as frequency, gain control or filter switching.

Radios for the Amateur Radio Service (ARS) have often had the control functions available that allow for alternate program control. This group has focussed on the implementation of RF, IF and audio functions, for receive and transmit, that have been moved into Digital Signal Processing (DSP). In this area, the opportunities expand to include modification of parameters or algorithms. In turn, this produces a different radio that is capable of new modes or alternative implementation of existing modes.

a) encourage SDR use, experimentation and the exchange of related ideas and information among radio amateurs;

b) organize, document and disseminate interesting approaches to SDR design and implementation;

c) enhance awareness among radio amateurs of what SDRs are and what they can do; and

d) advocate Amateur Radio designs offering access to digital signals and algorithms.

Two topics, Digital Voice and Wide-Band Data Modes are closely associated with SDRs, but not dealt with explicitly by the SDRWG. This is because of other working groups under the TTF are applied to these areas.

For 10 years or more, DSP devices have been used by amateurs for filtering and adaptive noise reduction^[†]. The performance available could be better than that available in analog circuitry. Amateur experimenters could build more complex filters with less effort. Manufacturers could also lower production costs. As DSP devices came down in cost and performance increased, it was natural to start moving more functions into DSP. These functions can typically include the last IF filtering, frequency shifting to audio for SSB or CW, detection functions for FM or data, audio filtering, noise blanking, AGC and the like.

All these functions, implemented by DSP, exist as computer programs. As such, their function is alterable by changes to the program. This is the basis for an SDR.

Many radios allow control of the existing functions, but that is too limited for an SDR. However, several experimenter SDR designs have been published:

Appendix B contains portions of the Rinaldo Background Paper referenced above. This has an excellent SDR and Amateur Radio DSP bibliography. In addition, an updated Bibliography is in Appendix A of this report.

SDRs have significant advantages: That is why the vast majority of new designs are jumping on the SDR bandwagon. A few disadvantages must be acknowledged.

Communicators across the globe are now enjoying the advantages of SDRs. Desirable characteristics include, but are not limited to:

a) the ability to receive and transmit various modulation methods using a common set of hardware;

b) the ability to alter functionality by downloading and running new software at will.

c) the possibility of adaptively choosing an operating frequency and a mode best suited for prevailing conditions;

d) the opportunity to recognize and avoid interference with other communications channels;

e) elimination of analog hardware and its cost, resulting in simplification of radio architectures and improved performance; and

While SDRs offer benefits as outlined above, a few obstacles remain to their universal acceptance. Those include:

Dynamic range is the ratio of the largest tolerable signal to the smallest usable signal. Receiver dynamic range is a central issue for software-defined radio (SDR) designers today and a major obstacle to advancement in digital receiver designs. The challenge is to find an effective digital signal processing (DSP) implementation of receiver functions that achieves dynamic range sufficient for the frequency range of interest., whether it be HF, VHF or above.

The dynamic ranges of the current crop of DSP receivers are determined partly by their analog stages. The idea is to remove such barriers so that performance is extended by digital technology.

DSP tends to minimize cost and maximize flexibility in receivers. Thus, another goal of SDR research is to eliminate as much analog hardware as possible. It may be possible some day to digitally sample signals directly from the antenna while still maintaining high dynamic range.

Data-conversion hardware capabilities have advanced to the point where direct digital conversion (DDC) radios are performing at levels equal to or better than their analog counterparts, provided that an analog filter restricts the bandwidth, to keep the peak voltage in the range of the A/D converter. Digital transceivers can achieve superior performance in every way. One of our goals shall be to dispel perceptions to the contrary. We acknowledge that SDRs involve a tradeoff between simplicity and performance.

Software may also affect dynamic range in digital transceivers. By example, we want to offer good information about how to avoid software limitations.

SDRs offer an excellent opportunity to examine new and better ways of defining and measuring dynamic range. Traditional definitions of nonlinear effects in analog circuits, such as intercept point (IP), tend to lose their meaning when applied to digital circuits. Therefore, we wish to explore the adoption of more meaningful measures of comparison among transceivers, such as spurious-free dynamic range (SFDR). We want to investigate the consistency of other conventional figures of merit.

We shall strive for our goals through consultation with Amateur Radio experimenters and by getting out information about relevant techniques. We postulate that acceptance of SDRs will rise in proportion to the general ham population's knowledge of what they can do. Therefore, another of our goals is to communicate good information about dynamic range in SDR design.

QST and QEX have already laid a lot of groundwork on that. For example, look at the Jul/Aug and Sep/Oct 2002 issues of QEX. We shall continue to press for publication of significant results and ideas for the future. In addition, our TIS Web page makes an excellent clearinghouse for knowledge, to which a widening readership has access.

The SDRWG strongly supports a continuation of QST articles, QEX papers and ARRL books covering SDRs, their design and application. Four specific areas were identified:

a) An article for QST should be solicited that would address SDRs from the user's point of view. We thought this would explain what an SDR is without a lot of technical depth, and explain how this could be used. The emphasis would be on the application of SDR to new modes, education and public service.

b) We advocate experimentation and publishing of material on all aspects of SDR design and use. Section 5 of this report is a collection of ideas that might serve as a starting point for potential SDR experimenters and developers.

c) A call needs to be made to the general membership for suggestions, ideas and volunteers. This might be in electronic form, supplemented by a notice in QST.

d) SDRWG members will review the TIS SDR web page that Ed Hare has created. The group will pool their collective knowledge of reference material for addition to the site. A new menu organization will also be proposed for easy access to the information.

This section is a summary of SDR ideas that have been discussed by the Working Group. Experimenters or manufacturers looking for SDR areas to explore might find something here. Generally, these ideas have not been fully developed, and no claim is made that they won't lead to major obstacles! Hopefully, these ideas could be made available to the broadest possible group of experimenters and developers to encourage them onward.

No attempt was made to arrange the following items in any particular order. Some of these have straightforward SDR implementations, whereas others will challenge the creative minds of Amateur Radio to provide new and exciting capabilities for the SDR of the future. In many cases, additional information is found in the references in Appendix A or in the Rinaldo paper referenced above.

A - High-speed digital I-F/audio interfaces to radios - Many application programs for the ARS use Sound Cards in a PC. For receivers using DSP, this means the signal path has gone from digital to analog, and was then sent to the Sound Card to go from Analog to Digital again. This leaves the signal subject to unnecessary distortion and additionally, it is exposed to hum and noise when going from the radio to the PC. If the digital data stream could be made available from inside the radio the path would be much cleaner.

B - User programmable elements of signal processing. - Functions such as a DSP filter are determined by a set of numbers (coefficients). Changing these numbers changes the filter. The user has the capability of programming these coefficients, either as a setup procedure, or as a real time change, altering the personality of the radio. One can carry this concept further. Gains can be changed, as can compression characteristics, equalizer settings, AGC constants, noise blanker constants, and speech processor characteristics.

All of these items are usually carried as constant numbers and altering these does not require access to the executable program. Yet they allow a good beginning for the experimenter and student to work with the internal algorithms.

Care is required that these constants not be set to values that can cause unwanted transmissions. But this is no different than the potentiometer adjustments available in a conventional analog transceiver.

C - API's - The ideal world allows control of any transceiver, including its inner workings, with a common language. For a programmer on the users side this presents a universal interface. Such an interface is not likely to exist for amateur radios, SDR or otherwise. A compromise is to insert a computer program function between the user and the radio, that translates the radio's language to standard function calls (the user's language). This is called an Application-Programmer Interface (API) and is a common concept for today's computers (see ref: Dobranski, below). The APIs need be written only once for each device and if all has been done carefully, all application programs will be ready to run.

Much coordination and effort would be required before a set of Amateur Radio APIs were in place. The benefits could be great.

D - Test methods that address characteristics of digital processing in SDR's. - Test methods tend to evolve around known weak areas in equipment or programs. If intermodulation distortion is a limiting factor for amplifiers, tests are developed to measure this. The digital portions of radios, including the data converters are often highly linear, right up to the point that they overload in a dramatic manner. Characteristics of this sort do not tend to produce conventional intermodulation distortion. This contrasts with an amplifier that often has a gradual compression characteristic with strong low-order intermodulation products. This is but one example of new technology asking for new test methods.

E - External programming language(s) for radios.- The DSP portions of an SDR are usually programmed in assembly language or a high-level language such as C. Learning to program requires more time investment than many experimenters are willing to invest. In the case of commercially produced radios, the programs that are used are normally considered proprietary, reflecting their considerable investment. These issues can be impediments for experimenters and students.

One possible alternative would be a programming language that would be interpreted by the SDR, and control the arrangement of the various DSP algorithms, as if they were building blocks. In addition, various parameters and characteristics would be programmable, but never the details of the algorithm. This allows a new mode to be programmed into a SDR. Or a new idea for a noise blanker or a speech processor could be tried.

SDR implementations on a PC might take advantage of DLLs to provide high performance DSP algorithms. It is even possible to call these from within Microsoft Excel using Visual Basic for Applications (VBA) code. Another useful approach could be to provide ActiveX Components that encapsulate many of the functions required so that they can be easily incorporated into any language that works with ActiveX components. An example would be to provide the sound card interface in one of these controls. You would tell the control the sampling rate and the block size you want and it would give you the data through an event.

This would not require learning a detailed language such as assembly language nor, in the case of a commercial SDR, having access to the manufacturers programs.

F - Front-end hardware supportive of high performance SDR - Typically the final data conversion frequency for SDRs is below a few hundred kHz. As noted in section 3, it is possible to attain high dynamic range with A/D and D/A converters and the digital processing that follows. To support this, down-conversion and up-conversion hardware designs are needed that can handle very strong signals, have low noise and use conversion oscillators that do not degrade the performance in the presence of strong signals.

This would be a straightforward hardware package to encourage the development of a

PC/sound card package for schools. It could be used with simple receivers via an audio

connection---less desirable than a digital path to transceivers but simpler and more achievable.

H - Transmitter distortion reduction - A number of techniques have been developed to reduce the intermodulation distortion in transmitters. These have not been applied in the ARS and situations remain where the potential QRM reduction could be valuable. Cartesian feedback, polar feedback, feed-forward corrections and pre-distortion are all candidate methods. Of these, pre-distortion may lend itself best to a DSP/SDR implementation. For this method, the low-level transmitter signal is distorted in amplitude and phase, but in the reverse way from the following transmitter distortion. Normally the pre-distortion characteristics are determined by comparing the low-level signal with a scaled down sample of the transmitter output. By successive approximation techniques, these two signals can be adjusted to coincide.

These techniques do not solve all problems of bad transmitter design. The transmitter amplifiers must be well-behaved, in the sense of having predictable input/output relationships. Oscillations are not allowed, either! Signals and noise must not be incidentally introduced far from the desired signals. This requires careful testing of the transmitter.

I - Antenna Beamforming - Two antennas are commonly used to remove a single interfering source by adjusting the relative amplitude and phase between the two antennas and summing their signals. With luck there is not a null in the direction of the desired signal, and the overall copy has improved. This concept can be extended to more antennas to remove more interfering signals and to sharpen the pattern towards the desired signal. Next, this can be extended to automatically maximize the received signal-to-noise ratio, by assuming some identifying characteristic of the signal (see ref below, Smith, QEX Nov/Dec 2000). If suitably designed, multiple SDR or a single SDR with multiple front ends would be ideal for this application.

The same principle can be applied for transmission, with multiple transmitters operating in concert. If the interfering signal referred to above is coming from another amateur station, the transmitter could place a null on that station to minimize interference back. Alternatively, the transmitter pattern can be set to have maximum gain in the direction of the received signal. All transmitter beamforming occurs at low level with DSP.

This approach allows interesting "growth" for an amateur station. One might start with a single antenna element and radio. As the desire and/or budget allow, more elements and radios (or radio front ends) can be added. This increases the receiving capability and transmitter power with each added element. City lots and 160 meters would not do well with this, but one might imagine small 10 meter arrays on a city lot. Two-meter Yagis as elements would be well adapted to this approach.

J - Interference Rejection - Any interference that is repetitive can be recognized and characterized by an SDR. The precise waveform of the interference may be deduced from the long time average. This way it is possible to subtract the correct waveform from the incoming signal despite the fact that many other signals are present at the same time. The subtracting of interference waveforms becomes particularly efficient if the SDR uses antenna beamforming and wide bandwidth so the wave form characteristics of the interference can be done with a bandwidth and a combination of input channels (antenna lobe) that maximizes the S/N of the interference. The contribution of the interference in the bandwidth and antenna direction used to receive the desired signal can then be deduced with good precision and canceled.

K - Better Decoding of Conventional Modes - An SDR makes it possible to improve reception of conventional modes. For CW an AFC makes it possible to use narrower filters without loosing the signal due to the normal frequency drift of conventional CW signals. Coherent CW is easily implemented and it improves readability of really weak signals. For SSB, an SDR can automatically find the correct BFO frequency from an analysis of the spectrum. Since the waveform of the human voice is repetitive, an SDR can find the time spans where the signal is repetitive and replace the original signal by a repetition of it's average. This way S/N for weak SSB signals can be greatly improved. For FM, a similar procedure is probably even more efficient.

In case two SSB signals occupy nearly the same frequency, an SDR could separate the mess into two clear voices by use of this procedure.

When the repetitive elements of the voice transmissions are found, the SDR can make reasonable assumptions about the influence of the speech processor, at the transmitter side, and present an audio signal that takes that into account.

a) experimenters and developers be encouraged to explore and report on SDR methods (sections 4 and 5, above),

b) a call be made to ARRL members to provide input to the SDRWG for ideas and direction (section 4),

c) all opportunities be used to make ARS SDR experimentation and programming a way to encourage new, technically inclined individuals into the hobby,

d) an article be solicited for general QST readership explaining the role of SDR in Amateur Radio (section 4), and

e) this report be made widely available to experimenters and manufacturers to encourage their SDR activities.

The following bibliography supplements that of the Rinaldo report, referenced above. Not specifically covered here are papers relating to Digital Voice or High Speed Data. The notes following the entries are to give an indication of the nature of the material.

Anderson, Peter T., KC1HR, "A Simple SSB Receiver Using a Digital Down Converter," QEX, Mar. 1994, pp3-7.

Anderson, Peter T., KC1HR, "A Better A/D and Software for the DDC-Based Receiver," QEX, Nov. 1994, pp11-15.

Anderson, Peter T., KC1HR, "A Simple Synchronous-AM Demodulator and Complete Schematics for the DDC-Based Receiver," QEX Sept. 1997, pp3-14.

Anderson, Peter T., KC1HR, "A Better and Simpler A/D for the DDC-Based Receiver," QEX, Aug. 1996, pp21-24.

Material on the LinRad SDR by SM5BSZ, associated hardware and use of the Delta44 data conversion board. The Lin Rad project is advanced and extensive. Executable software (Linux) is available for download.and

Their comments: This article focuses on the filtering techniques used in Software-Defined Radio (SDR), an emerging technology in which digital filters are critical components.

Bloom, Jon, KE3Z, "Negative Frequencies and Complex Signals," QEX, Sept. 1994, pp22-27.

De Carle, Bill, VE2IQ, A DSP Version of Coherent-CW (CCW)," QEX, Feb. 1994, pp25-30.

Dobranski, Lawrence G., VA3LGD/VE3TVV, "The Need for Standard Application-Programming Interfaces (API's)in Amateur Radio," QEX, Jan./Feb. 1999, pp19-21.

Emerson, Darrel, AA7FV/G3SYS, "Digital Processing of Weak Signals Buried in Noise," QEX, Jan. 1994, pp 17-25.

Their comments: HSA Analog I/O Module was designed as a development platform for a wide variety of wireless/digital radio applications. Features are high speed, high resolution I/O and a 1 million gate Virtex FPGA. The two input channels employ Analog Devices AD6644 A/D converters for an unprecedented 14 bit digitization out to 65 MHz

Freeth, Bob, G4HFQ, 'Making a Sound Card Work for You," QEX, May/June 2001, pp 15-22.

Frohne, Rob, KL7NA, "A High Performance, Single Signal Direct Conversion Receiver with DSP Filtering," QST, April, 1998, pp40-45.

Forrer, Johan, KC7WW, "Programming a DSP Sound Card for Amateur Radio," QEX, August 1994, pp 9-15.

Forrer, Johan, KC7WW, "An Adaptive HF Modem for 100 and 200 Baud," QEX Nov. 1994, pp3-10.

Forrer, Johan, KC7WW, "Using the Motorola DSP56002EVM for Amateur Radio DSP Projects," QEX, Aug. 1995, pp14-20

Forrer, Johan, KC7WW, "A Low-Cost HF Channel Simulator for Digital Systems," QEX, May/June 2000, pp 13-22

Hale, Bruce, KB1MW, "An introduction to Digital Signal Processing," QST, July 1991, pp35-37

Hall, Doug, KF4KL, "Spectral Subtraction for Eliminating Noise from Speech," QEX, Apr. 1996, pp 17-19..

Hershberger, Dave, W9GR, "Low Cost Digital Signal Processing for the Radio Amateur," QST, September, 1992, pp43-51. TMS320C10 design that included multiple audio filters and LMS notch and peak. Source code was available.

Puig, Carlos M., "A Weaver Method SSB Modulator using DSP," QEX, Sept. 1993, pp8-13.

Qiuting Huang, Clemens Hammerschmied and Thomas Burger, "Meeting the Challenge of High Dynamic Range, High Speed A/D Conversion for Software-Defined Radio (SDR),"

This is an infant commercial system with AD664 and AD6620, sampling at 65MHz. Interesting examples of capability are shown.

Their comments: The SDR Forum is an open, non-profit corporation dedicated to supporting the development, deployment, and use of open architectures for advanced wireless systems. The Forum membership is international, and growing. The Software Defined Radio Forum will conduct a Technical Conference and Product Exposition in San Diego on 11, 12 November 2002. This conference will focus on technology, standards and business activity related to software radios and will provide an international perspective of the current state of the art.

Scarlett, James, KD7O, "A high Performance Digital-Transceiver Design, Part 1," QEX, Jul/Aug 2002, pp35-44.

Smith, Doug, KF6DX, "Improved Dynamic Range Testing," QEX, Jul/Aug 2002, pp46-52.

Smith, Doug, KF6DX, "Introduction to Adaptive Beamforming," QEX Nov./Dec. 2000, pp50-55.

Smith, Doug, KF6DX, "Deconvolution in Communication Systems," QEX Sept./Oct. 2001, pp45-51.

Smith, Doug, KF6DX, Digital Signal Processing Technology: Essentials of the Communications Revolution, ARRL, 2001.

The Scientist and Engineer's Guide to Digital Signal Processing California Technical Publishing. Can be purchased, or downloaded in .PDF.

Their notes: FPGAs - An FPGA is an ideal platform for a SDR since it has a re-configurable nature that allows system architects to customize hardware to their specifications without committing to an architecture for the lifetime of a product.

Youngblood, Gerald, "A Software Defined Radio for the Masses: Part 1, " QEX, July/Aug 2002, pp13-21, with three more follow-on articles.

by Paul Rinaldo, 3 July 2001.. The following material is included for reference. To avoid duplication this is not the complete paper. Note that the bibliography of this appendix compliments that of Appendix A.

In its basic form, a software defined radio (SDR) is a digital computer connected to an antenna and controlled by software. It is also known as software radio or soft radio, although there are some subtle differences. A related term is cognitive radio.

Most software receivers have an analog front end consisting of band-pass filtering, a low-noise RF amplifier to set a low system noise level, a local oscillator and mixer to heterodyne the signal to an intermediate frequency (IF) where analog-to-digital (A/D) conversion, digital filtering and demodulation takes place. Recently, however, there are some software receivers that perform A/D conversion immediately after the antenna.

Federal Communications Commission ET Docket No. 00-47 Notice of Inquiry (NOI) is an initiative of Office of Engineering and Technology then-Chief Dale N. Hatfield, WØIFO, to solicit comments from the public on software radios. He previously requested advice from the FCC Technological Advisory Council (TAC).

The NOI asked a number of detailed questions about software radio technology capabilities, commercial deployment, international developments, facilitation of interoperability, movement toward standards, improvement of spectrum efficiency and impact on FCC rules. The Commission is understandably concerned about privacy of communication, prevention of using unauthorized frequencies and equipment certification issues.

The National Telecommunications and Information Administration has established a WICI Committee under the Interdepartment Radio Advisory Committee (IRAC) to identify new wireless technologies, including software radio, that could be applied to federal radio users. The WICI Committee is seeking input from industry through ongoing meetings.

The JTRS Joint Program Office describes their mission as "Acquire a family of affordable, high-capacity tactical radios to provide Interoperable LOS/BLOS C4I capabilities to the warfighters." (LOS is line of sight; BLOS is beyond LOS; C4I is command, control, communications, computers and intelligence.) JTRS is issuing Version 1.0 of their Software Communications Architecture on April 28, 2000 and a later version in November.

SDR Forum is a nonprofit group dedicated to the development of advanced wireless systems. The chairman is Stephen Blust, W4SMB. A membership fee of $2000 (for nonprofit organizations) is required to participate or obtain documents. See http://www.sdrforum.org.

At the initiative of Eric Schimmel, Telecommunications Industry Association, ITU Radiocommunication Working Party 8A (land mobile and amateur services) drafted a new Question to mandate study of software radios. (A Question invites contributions to the ITU.)

On motion of Mr. Harrison, seconded by Mr. Bodson, it was VOTED unanimously that ARRL proceed with the development of Software Defined Radios (SDR) for the Amateur Service in accordance with the Technology Working Group report. The President shall appoint a group of individuals knowledgeable in the field from the international Amateur community and industry. The group shall report to the Technology Task Force and submit an initial report at the 2001 Second Board Meeting.

There are a number of computer-controlled radios used by amateurs. These have some features that could be considered early forms of SDRs. Amateurs interested in computer control of additional features of these radios have been somewhat frustrated by the lack of access to some functions.

Probably the primary core technology applicable to SDRs in the Amateur Radio Service is digital signal processing (DSP). A number of amateurs are skilled in DSP programming and there are numerous DSP implementations in amateur equipment designs. A number of references on DSP in the amateur literature are included in the bibliography.

Field programmable gate arrays (FPGAs) and application specific integrated circuits (ASICs) are also important technologies in SDRs. Designing and having them manufactured for amateur applications is probably unrealistic (although there is precedent of the Curtis ASIC used in keyers). However, ASICs designed for commercial applications are worthy of study.

While the FCC has considered specific rules for SDRs in commercial service, they did not suggest any rule changes applicable to the Amateur Radio Service. There does not appear to be any need to change Part 97 of the FCC rules but there is need to monitor potential modifications to other Parts (e.g., Part 2) for constraints on the Amateur Radio Service.

Belk, Nathan. "Universal radio is on the horizon." Electronic Engineering Times, 19 March 2001, pp98-138

Berruto, E. and Colombo, G., "AC015 - RANBOW: Protocol and Architectural Issues for Software Radio," Software Radio Workshop, Brussels, May 1997

Bindra, Ashok, "Speedier High-Resolution Data Converters Make Software Feasible," Electronic Design, 16 October 2000, pp95-104

Bottiglieri, Joe, AA1GW, "The ClearSpeech Speaker and ClearSpeech Line DSP Audio Filters (Product Review)," QST, April 1999, p62

Brannon, Brad et al, "A Look At Software Radios: Are They Fact Or Fiction?," Electronic Design, 1 December 1998, pp117-122

Brannon, Brad and Cloninger, Chris. "Soft radio runs into hard standards," Electronic Engineering Times, 19 March 2001, p98 and 110

Brock, Darren K. "Superconductor Digital RF Development for Software Radio," IEEE Communications Magazine, February 2001, pp174-179

Buhe, Gerrit. "Software Radio -- genauer hingesehen," CQ DL, November 2000, p802

Buracchini, Enrico, "The Software Radio Concept," IEEE Communications Magazine, September 2000, pp138-143

Büscher, Wolfgang, DL4YHF, email to LF reflectors, May 29, 2001, tests with direct reception of LF signals using only a PC soundcard; info: http://www.qsl.net/dl4yhf/vlf_rcvr.html

Chester, David B., "Digital IF Filter Technology for 3G Systems: An Introduction," IEEE Communications Magazine, February 1999, pp102-107

Cummings, Mark, "FPDGA in the Software Radio," IEEE Communications Magazine, February 1999, pp108-112

Cummings, Mark, "Time for software-defined radio," Electronic Engineering Times, 19 March 2001, p94 and 108

Danzer, Paul, N1II, "Timewave DSP-59Y Audio Noise Reduction Filter," (Product Review), QST, September 1997, p78

Danzer, Paul, N1II, "Timewave DSP-599zx Digital Signal Processor," (Product Review), QST, December 1996, p104

Federal Communications Commission, "FCC Proposes Rule Changes to Facilitate Software Defined Radio Deployment," FCC News release, December 7, 2000

Federal Communications Commission, "In the Matter of Authorization and Use of Software Defined Radios," Notice of Proposed Rule Making, ET Docket No. 00-47, released December 8, 2000

Ford, Steve, WB8IMY, "AEA DSP-232 Multimedia Data Controller," (Product Review), QST, July 1996, p59

Forrer, Johan, KC7WW, "A DSP-Based Audio Signal Processor," QEX, September 1996, p8

Frohne, Rob, KL7NA, "Designing Custom frequency Responses for FIR Filters," QST, April 1998, p40

Glassey, Robert, GØTVQ, "New DSP RTTY Demodulator Software (Blaster TeLetype BTL), QST, February 1997, p104

Gratzek, Tom, Software Radios for Cellular/PCS Base Stations, Fact or Fiction,

Hatfield, Dale N., "Software Defined Radio: A Regulator's Perspective," Keynote address at SDR Forum 19^th General Meeting, Seattle Washington, 20 June 2000, http://www.fcc.gov/oet/speeches/sdrforumsph.htm

Hawker, Pat, G3VA, "More on Digital ('Software') Radios," Technical Topics, RacCom, November 2000, p60

Hedberg, Bo, Technical Challenges in introducing Software Radio for Mobile Telephony Base Stations, Ericsson Radio Systems,

Hershberger, David, W9GR, "DSP--An Intuitive Approach," QST, February 1996, p39

Horzepa, Stan, WA1LOU, "DSP Emulation with Your 486," (Digital Dimension), QST, September 1996, p89

Hosking, Rodger, "Digital Receivers Bring DSPs to Radio Frequencies," Wireless Systems Design, October 2000, pp9-14

International Telecommunication Union, Adaptive radio systems for frequencies below about 30 MHz," Recommendation ITU-R F.1110-2

International Telecommunication Union, Software defined radios, Question ITU-R 230/8, 2000

Klopfer, Ron, "DSP-Based Digital Radio Design," Communication Systems Design, October 1998, pp37-40

Larkin, Bob, W7PUA, "The DSP-10: An All-Mode 2-Meter Transceiver Using a DSP IF and PC-Controlled Front Panel," QST, Part 1 September 1999, p33; Part 2 October 1999, p34; Part 3 November 1999, p42

Lindquist, Rick, N1RL, "Japan Radio Company NRD-545 DSP Receiver," (Product Review), QST, February 1999, p62

Lindquist, Rick, N1RL, "MFJ-784B Tunable DSP Filter," (Product Review), QST, January 1996, p72

Maia, Fred, "Comments Close on Software-Defined Radio Rules," W5YI Report 1 May 2001

McKinney, J. David and Von Dolteren George, "PLDs critical to flexible wireless," Electronic Engineering Times, 9 March 2001, p100 and 112

Meyer, Ronald, "Software-Defined-Radio Technology Targets 3G Designs," Wireless Systems Design, February 2000, pp16-20

Mitola, Joseph and Zvonar, Zoran. " Software and DSP in Radio." IEEE Communications Magazine, November 1999, p68; February 2000, p138; August 2000, p140-150

Mitola, Joseph, et al, "Globalization of Software Radio," IEEE Communications Magazine, February 1999, pp82-83

Mitola, Joseph, "Technical Challenges in the Globalization of Software Radio." IEEE Communications Magazine, February 1999, pp84-89

Munro, Alistair, "Mobile Middleware for the Reconfigurable Software Radio," IEEE Communications Magazine, August 2000, p152

Perez-Neira, Ana, et al, "Smart Antennas in Software Radio Base Stations," IEEE Communications Magazine, February 2001, p166

Pucker, Lee, "Paving Paths to Software Radio Design," Communication Systems Design, June 2001, p19

Ralston, John and Riordan, Ken, "Software radio making headway," Electronic Engineering Times, 19 March 2001, p84 and 106

Ralston, John D, "SDR Emerges to Address Wireless Industry Needs," Wireless Systems Design, October 2000, p42

Reed, Jeff , et al, " New Educational Program on Software Radios," The Propagator, June 1999, p15

Rohde, Detlef, DL7IY, "A Low-Distortion Receiver Front End for Direct-Converstion and DSP Receivers," QEX, Mar 1999, p30

Shepherd, Rob, "Engineering the Embedded Software Radio," IEEE Communications Magazine November 1999, p70

Smith, Doug, KF6DX, "Signals, Samples and Stuff: A DSP Tutorial," QEX, Part 1, March 1998, p3; Part 2, May 1998, p22; Part 3, July 1998, p13; and Part 4, September 1998, p19

Srikanteswara, Srikathyayani, et al., "A Soft Radio Architecture for Reconfigurable Platforms," IEEE Communications Magazine, February 2000, p140

Thryft, Ann R, "Soft radio key to universal applications," Electronic Engineering Times, 19 March 2001, p81

Tsurumi, Hiroshi and Suzuki, Yasuo, "Broadband RF Stage Architecture for Software-Defined Radio in Handhelp Terminal Aplications," IEEE Communications Magazine, February 1999, p90

Tuttlebee, Walter H.W., "Software Radio Technology: A European Perspective," IEEE Communications Magazine, February 1999, p118

Wiseman, John, KE3QG, "A Complete DSP Design Example Using FIR Filters," QEX, July 1996, p3

Wong, Bennet, "Net growth is taking SDR to task," Electronic Engineering Times, 19 March 2001, p102 and 122

Zahnd, Hans, "Software-Radio -- Technologie der Zukunft," CQ DL, August 2000, p580

^[*] See Paul Rinaldo, W4RI, "Background Paper on Software Defined Radio," Draft 3 July 2001, extracts of which are included in Appendix B..

^[†] See Hershberger reference in appendix A for an early example of DSP audio processing for Amateur experimenter use. Although not a complete radio, by SDRWG definition this is a (partial) SDR.