Content area
Real and virtual instruments for data acquisition and signal processing are widely used in industry, entertainment and education. In this paper I compared different hardware and software solutions in instrumentation, and I analyzed these technics in a particular case: generation and processing of audio signals using additive and FM synthesis. The comparative study reveals the advantages and the disadvantages of the real and virtual instrumentation especially in academic education.
Abstract – Real and virtual instruments for data acquisition and signal processing are widely used in industry, entertainment and education. In this paper I compared different hardware and software solutions in instrumentation, and I analyzed these technics in a particular case: generation and processing of audio signals using additive and FM synthesis. The comparative study reveals the advantages and the disadvantages of the real and virtual instrumentation especially in academic education.
Keywords: data acquisition; signal processing; FM synthesis; real and virtual Instrumentation.
(ProQuest: ... denotes formulae omitted.)
I. INTRODUCTION
Data acquisition and signal processing are of particular importance nowadays. We encounter data acquisition and signal processing in many fields of science and everyday life, such as telecommunications, entertainment electronics, military technology, video and audio signal processing, television, music and speech recognition. In this paper I would like to present the generation and processing of audio signals, because this research area is very important today, it is very actual, and an entire industry is based on today's multimedia technologies. It has a huge market, user base and large financial background. We encounter audio signal processing in the entertainment industry, music, synthesizers, studio technology, mobile phones, speech recognition and artificial intelligence [1, 2].
I'd like to study and compare the generation, acquisition and processing of audio signals with real (hardware) and virtual (software) devices. Each solution has advantages and disadvantages, and I compare the use of the two methods in university education and distance education [11, 13, 16]. Particularly, I want to investigate the production and processing of audio signals with different hardware devices (signal generators, synthesizers) and the possibility of producing these signals with software tools, such as virtual instruments, virtual synthesizers or LabView applications created by students [3].
After the initial analog electronic era, with the development of computer technology, new possibilities opened in the direction of signal and sound processing. With the advent of modern digital synthesizers, it became possible to replace traditional, classical instruments with electronic instruments and reproduce the sound of entire orchestras. These devices have become more advanced over the years, they can reproduce traditional instruments completely, and they have also managed to produce new, never-heard, synthetic sounds.
The first synthesizers used subtractive synthesis to produce modified sounds. The over harmonics of a given, real signal were cut out with the help of filter circuits, changed and thus obtained a duller, more modified tone. After the subtractive synthesis-based generators, the next class of synthesizers appeared, the analog additive synthesizers, where different, simple signals were combined and added together, thus generating more complex signals [4, 16].
In 1973 John Chowning described the basics of the FM synthesis [5], and this method became a well-known and developed method to produce various complex sound samples since the 80s. The first synthesizer that used frequency modulation synthesis was the Yamaha DX7. This synthesizer had six operators and 32 algorithms and was able to create new, never-before-heard sounds [10].
Towards the end of the 80s, different instrument manufacturers began to use sampling-based tone generation (PCM or AWM) (Advanced Wave Memory). Modern instruments use multiple technologies, such as the Yamaha Montage or MODX series which includes both AWM and FM synthesis [7]. The Montage M series, released in 2024, use three different technologies: AWM, FM and analog synthesis (AN-X engine). Of course, other major manufacturers also use these basic technologies.
In this comparative study, I used as hardware equipment a National Instruments myDAQ data acquisition device running NI ELVIS Software Instruments and a Yamaha Modx8+ synthesizer. For virtual instrumentation and programming, I used the Steinberg FM Lab virtual synthesizer and Dexed, the free VST FM synth based on the Yamaha DX7, and the LabView graphical programming environment.
II. AUDIO SIGNALS AND FM SYNTHESIS
A. Fourier series
A periodic signal, as a mathematical function, can be described with Fourier series as a sum of different trigonometric functions [6, 9]. The Fourier series of a given function over [-, ] can be represented as in (1):
... (1)
... (2)
... (3)
... (4)
This definition applies to all naturally occurring periodic signals, vibrations, waves, sounds, and lights. Generally, in electronics and computer technology, we find high-frequency, complex signals, in many cases with complex waveforms. From the point of view of education, the easiest signal form to present, understand and test is the audio signal, which has a much lower frequency, it can be interpreted and examined with simple laboratory equipment. In the case of distance education, we must assume that students who want to do laboratory exercises remotely, have limited hardware devices at home, and that is why the experiments and tests must be presented in a simple, understandable form (VI's).
B. FM synthesis
In widely used FM synthesis, the basic process involves two components: the carrier signal and the modulator signal. The carrier signal, typically a sine wave, is modified during the FM synthesis with the modulator signal, who influences the frequency of the carrier wave. By changing the modulator's frequency and amplitude, different harmonics and tones can be obtained [5, 7, 8, 13].
The mathematical equation of a frequency modulated sine wave can be described as follows:
f(t) = ð´A ∙ sin (αt + ð¼I ∙ sin (βt)), (5)
where α is the carrier frequency, β is the modulating frequency, and I is the modulation index.
The equation can be expressed as:
f(t) = ð´A ∙ sin(2ðœ‹πfc + 1 ∙ sin (2ðœ‹πfmt)) (6)
where fc is the carrier frequency, fm is the modulator frequency and I is the modulation index. By modifying the modulator frequency and the modulation index various complex signal forms can be obtained, using a small number of oscillators.
III. HARDWARE BASED WAVEFORM GENERATION
Expensive, complex hardware devices can be found in a school or university laboratory, such as signal or waveform generators in a microprocessor or data acquisition laboratory, or synthesizers in multimedia labs. However, it would be good to reproduce these tools using software tools that students can access remotely or test at home. By comparing the hardware signal generation with the software signal processing, I would like to show the possibility that students can perform certain laboratory exercises even from home.
In classrooms, with suitable hardware devices, like waveform generators, oscilloscopes and logic analyzers, the generation and processing of different signals can be easily solved. As an alternative to dedicated hardware devices, it can be used modern data acquisition devices too, which can successfully replace the real laboratory instruments [3, 11]. In this case the hardware (e.g. NI myDAQ) is accompanied by appropriate software tools (NI ELVIS Software Instruments, LabView), as can be observed in Fig.1.
Among various hardware solutions, I'd like to examine the generation of different audio signals using a Yamaha Modx8+ synthesizer, to present the basics of the FM synthesis [7, 10]. In the FM engine there are 8 operators that can be interconnected by different algorithms. To create a new waveform, the first step is to create a new performance named "Init Normal (FM-X)", based on a simple sine wave. Next step is the selection of the generator algorithm, which describes the interconnection mode of the oscillators. Some oscillators are configured as carriers, the others are modulators. In every algorithm exists only one feedback loop, by the carrier or by the modulator. The value of the feedback can be modified obtaining a significant alteration of the signal (Fig.2.).
There are 88 algorithms, with multiple levels of modulation. Starting from simple sinewaves and algorithms containing only a few oscillators (e.g. 1 carrier and 1 modulator), students can understand the principles of FM synthesis, and then they can choose more complicated algorithms, with multiple oscillators (Fig.3. and Fig.4.).
The obtained signal can be adjusted with the envelope curve (attack, decay, release, sustain) (Fig.5.), other processing (Fig.6., Fig.7), and finally can be applied various effects from the effect processor.
IV. VIRTUAL INSTRUMENTATION IN SOUND GENERATION
Expensive hardware laboratory devices can be replaced by various software solutions, such as Steinberg FM Lab or Dexed (Fig.8.). Of course, a complex hardware or software device is capable of much more than a simple signal generation, capable of producing a wide variety of tones, with a high degree of polyphony, playing several instruments at the same time and even playing orchestral sounds. These devices are capable of MIDI editing, audio conversion and control of other devices, but in this paper, we focus only on the signal generation and synthesis function. The above-mentioned software instruments can reproduce the functionality of an FM based hardware synth, having some advantages and disadvantages.
Students can familiarize themselves with the basics of sound synthesis using this kind of complex software, many of them being free. The next important step is the design and the implementation of their own signal processing application. They can use for example the LabView graphical programming environment, and this is suitable even for online education [1, 2, 3, 11]. With powerful mathematical and signal processing libraries this environment allows signal generation and analysis in time and frequency domain. Students can experiment different operations applied to the signal such as filtering, resonance, effects (delay, white or pink noise), resulting in a highly processed, modified signal.
In the following examples I created two applications in LabView to generate and acquire different waveforms and to study the additive, and the FM synthesis. I used two oscillators, both can reproduce sine, triangle, square and sawtooth signals. Combining them using the FM Synthesis (Fig.9, Fig.10), or by superposition (Fig.11), we can obtain more complex waveforms.
The obtained signal is sent to the computer's soundcard, configured to function as a data acquisition board, so the signal can be heard as interesting, pulsating notes. Every parameter (carrier and modulator signal's shape, amplitude, frequency, offset, phase) can be modified in real time, and the software displays the waveform in time and frequency domain.
Using these tools, students can implement different, more complex signal generation and processing algorithms, and they can work at the classroom, or even from distance using Virtual Instrumentation.
V. CONCLUSIONS
Real and virtual instruments are widely used in industry, entertainment and education. In this paper I studied these instruments in a particular case: audio signal generation and processing using FM synthesis. Each solution has its own advantages and disadvantages. Real instruments usually are more expensive, more rigid than virtual instruments, but they offer a real feel and can be used easily. Virtual instruments are flexible, less expensive than real laboratory instruments, in many cases can be used from a distance (e.g. online education) but they need carefully configuration and don't offer the real feel in some experiments.
In the special case of FM synthesis, both solutions can be used. There are complex real devices that can produce audio signals using FM synthesis (waveform generators, synthesizers), but they are usually more expensive than virtual instruments. For hardware signal generation I used a NI myDAQ data acquisition board with NI Elvis Software Instruments and a Yamaha Modx8+ synthesizer. For software-based signal generation we can use some FM synth software like Steinberg FM Lab, Dexed, or others, but in education is highly recommended for students to create their own applications. In this case the LabView graphical programming environment can be successfully used, and I presented some examples for additive and FM signal generation and processing.
REFERENCES
[1] Emmanuel Hagawusia, Moli Watson, Sharma Avines, Charan Kaushal, "Measurment And Control Of Physical", Project Title "A Labview Based On Fm/Am Modulation", University of the South Pacific, April 2020
[2] Željko Obrenović, Dušan Starčević, Emil Jovanov, "Virtual Instrumentation", 14 April 2006, https://doi.org/10.1002/9780471740360.ebs1265
[3] National Instruments Virtual Instrumentation, 2024, https://www.ni.com/en/shop/labview/virtualinstrumentation.html?srsltid=AfmBOopwWt9e5dUbiXxeE P_GLoNSPg88PSy0nGMRLvu0xXfkEkqi1Bwx
[4] Yamaha Corporation, Substractive Synthesis, 2024, https://yamahasynth.com/learn/synth-programming/subtractive-synthesis-101-part-one-the-basics/
[5] Chowning, John M. 1973. "The Synthesis of Complex Audio Spectra by Means of Frequency Modulation". Journal of the Audio Engineering Society 7, no. 21:526–34. ISSN 0004-7554
[6] Wolfram Mathworld, Wolfram Mathematica, Wolfram Alpha, "Fourier Series", 2024, https://mathworld.wolfram.com/FourierSeries.html
[7] Yamaha Corp., Mastering MODX: An FM-X Exploration, 2024, https://yamahasynth.com/learn/modx-seriessynthesizers/ mastering-modx-fm-x-exploration-part-i/
[8] Victor Lazzarini, Joseph Timoney, "Theory and Practice of Modified Frequency Modulation Synthesis", Sound and Music Technology Group, National University of Ireland, Maynooth, Ireland, 2010, J. Audio Eng. Soc., Vol. 58, No. 6, 2010 June, pp 459-471
[9] Teodorescu Rodica-Mihaela, Anghel Gheorghe, Drăgușin Sebastian-Alexandru, "Fourier Series Using Virtual Instrumentation", Department of Electronics, Communications and Computers, University of Pitesti, Romania, June 2017, ISSN – 2344 – 2166
[10] Franco Caspe, Andrew McPherson, Mark Sandler, "DDX7: Differentiable Fm Synthesis Of Musical Instrument Sounds", Queen Mary University of London, ISMIR 2022
[11] Soumya Ranjan Panda, Avinash Mohanty, "Application of Virtual Instrumentation & LabView in communication systems", Department of Electronics and Instrumentation Engineering National Institute Of Technology, Rourkela, India, 2007
[12] Tae-Hwan Kim, Young-Joo Lee, In-Cheol Park, "A Scalable and Programmable Sound Synthesizer", Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Republic of Korea, Conference Paper, May 2009
[13] Li Luo, "Implementation and Optimization of the Synthesis of Musical Instrument Tones using Frequency Modulation", Phd Thesis, Fakultät für Ingenieurwissenschaften, Elektrotechnik und Informationstechnik, Universität Duisburg-Essen, 2017, https://nbnresolving.org/urn:nbn:de:hbz:464-20170622-133355-9.
[14] Hilko Tondock, Henrik von Coler, "A Virtual Instrument For Ifft-Based Additive Synthesis In The Ambisonics Domain", Audio Communication Group TU Berlin, Proceedings of the 26th International Conference on Digital Audio Effects (DAFx23), Copenhagen, Denmark, 2023
[15] H. von Coler, "A System for Expressive Spectro-spatial Sound Synthesis", PhD Thesis, Technische Universität Berlin, 2021
[16] Ed Doering, "Musical Signal Processing with LabVIEW – Subtractive Synthesis", Rice University, Houston, Texas, Publisher OpenStax CNX, 2022, https://ftpmirror.your.org/ pub/misc/cd3wd/1006/Musical_Signal_Processing_with_L abVIEW_labview_music_cnx_x17.pdf
[17] Victor Lazzarini, Joseph Timoney, "Higher-Order Frequency Modulation Synthesis", Maynooth University, Ireland, 2023, https://doi.org/10.48550/arXiv.2305.0790
Copyright University of Oradea Oct 2024