This proposal was approved by the IEA Wind ExCo at ExCo 78 in Brussels on December 1st 2016. As of December 2016, the new task has several confirmed participants from Ireland, Denmark and Sweden, with many others in discussions to join. The aim is to start work in early 2017. You can find more information here.
This document has been prepared by the Distributed Wind Energy Group in the Department of Mechanical & Manufacturing Engineering at Trinity College Dublin, in conjunction with the Sustainable Energy Authority of Ireland.
Table Of ContentsMore renewable power generation systems were deployed than thermal power generators for the first time in 2013. The rapid rise to the dominant position of these developing technologies in new power sector investments reflects the policy priority of decarbonising global electricity systems.
Societal acceptance of new technologies can be key to their ultimate successful adoption, the utility of a technology may not automatically guarantee this if its deployment affects those in society who do not accrue direct benefits from it. Emergent societal acceptance issues are commonplace for new technologies as they reach the mass deployment stage, particularly where this is rapid and these issues must be satisfactorily resolved for the technologies that are ultimately to be successful.
Particular public concerns have been raised on the impacts of wind turbine noise in some jurisdictions. While the sources of excessive noise for the majority of society will be from other man-made technologies, it is important to ensure that those who live in the proximity of wind turbines will not be subjected to excessive noise from these.
Regulation has an important role in the technology development process. Good regulation will anticipate progressive and societally beneficial technology advances and facilitate their adoption
It is important that regulations are based upon objective analysis and that guidelines are put in place to allow policy makers focus the debate on to the right topics.
A survey of national regulations dealing with noise control for wind turbines reveals that, while there are specific international standards for noise measurement, there are a diversity of national noise limits and defined approaches to the overall wind turbine noise assessment process. This diversity indicates that there may be room for convergence towards a common framework underpinning national guidance.
From an industry perspective, the wind turbine supply market is international and not local. The diversity of local noise regulations may bring unwarranted compliance costs and confound progressive technical advances to develop wind turbines with lower noise impact.
This proposal is for a new IEA Wind RD&D research collaboration to accelerate the development and deployment of quiet wind turbine technology. The collaboration will carry out its work in a series of focussed work packages.
The objective of the proposed task, the Task, is to convene an international expert panel to: (i) exchange learning, (ii) identify and report best practice in the measurement and assessment of and noise; (iii) develop an IEA Recommended Practice contributing to the ongoing development of IEC standards for wind turbine noise.
Of course, there is a considerable body of work already completed; the broader work on wind turbine noise is considerable and growing [25]. On amplitude modulation specifically, work has been completed by organisations in several countries and across many disciplines. These organisations include NREL, Risø DTU, the Institute of Acoustics in the UK and several efforts in Canada, e.g. [5] among many others.
This task will build consensus through interdisciplinary cooperation, involving insofar as possible engineers and scientists, environmental scientists and acousticians, psychologists, physiologists, audiologists, sociologists and policy makers. Initial work will structure discussion between disciplines, groups and countries. An aim is to understand and recognise the value in the work completed across disciplines. Conflicting or incomplete ideas can be identified for further research work as part of the task. Such a process will lead to robust, scientific, widely accepted and transparent metrics for wind turbine noise. It is in this spirit that we make this proposal.
It is worth highlighting that noise is not only or even primarily an engineering problem. The engineer quantifies noise source mechanisms and characteristics and develops instrumentation, data analysis tools, models and procedures for measuring and predicting wind turbine noise. Effective environmental noise metrics for wind turbines, however, require psycho-acoustic testing and an understanding of the physiological response of the human ear, among other things. Individual technologies have individual peculiarities to their noise characteristics and methods and simplifying assumptions developed for existing technologies may have to be reassessed and adapted for the noise characteristics of new technologies. For example, consideration of amplitude modulation is not commonplace for existing technologies but specific consideration of this may be required for wind turbines.
For this reason, two technical work streams are proposed, understanding the problem of wind turbine noise from the perspective of both the physics of the noise generation and the effect of turbine noise on human receptors.
In the end, the aim is well-informed agreement. This will happen through engaged participation.
In addition to a management work package, an interdisciplinary education/guidance work package (WP 1) is proposed; this will help set the direction of the investigative work (WP2 2) that will in turn feed into a knowledge base and an evolving recommended practice (WP 3). The evolving recommended practice will form the basis of discussions during WP1. The Distributed Wind Energy Group (DWEG) at Trinity College Dublin will lead all work packages, guided by the expertise of the task participants. Priorities will be finalised after consultation with potential participants.
The Task will be managed within this work package. The coordinator will ensure that the technical objectives and deliverables are achieved and will facilitate ongoing technical communication among the Work Package Leaders and the IEA Wind ExCo.
One of the first jobs of this WP will be to establish a web site with both public and private access for the Task. This website will facilitate discussion and collaboration between the Task partners as well as provide public information, e.g. papers. The WP will help define standard data and reporting formats for the Task. The WP will use appropriate project management procedures. Dissemination and regular liaison with the IEA Wind ExCo will be an integral part of this work package.
How wind turbine noise affects people and how best to monitor exposure; these questions are not yet fully answered. Much work has been conducted at international level across disciplines including engineering, physiology, psychology and sociology. This work package will support interdisciplinary discussion, e.g. [18][15][25]. The aim is for a consensus with robust, scientific, widely accepted and transparent metrics for wind turbine noise.
At the outset, each partner will present their interpretation of best practice and summarise the outcomes of the research work in their field and in their country. These discussions will take place at six-monthly workshops. The individual contributions to these workshops will be collated into a state of the art report. The Task 34 approach of topical white papers will be considered. The value that each discipline brings to solving the principal problem will be reinforced by these discussions.
This work package will provide guidance for the two technical work packages, WP2 and WP3. The group will identify direction and work to find consensus on areas where further research is required.
Engagement with policy makers is an important function of this WP. The location of the workshops will rotate through partner countries to facilitate dissemination to local policy makers and others. The normal practice in Tasks is to make meetings open to selected local experts and informed officials with a tandem local event for wider local dissemination.
As a starting point, we propose that WP2 (and WP3) begin with amplitude modulation, see the Appendix.
WP 1 deliverables will be 6 x meetings / workshops, i.e. every 6 months for 36 months.
As mentioned, it is proposed that WP2 will begin with amplitude modulation. The work package objectives are gated, i.e. a decision to change the focus of the research can change depending on progress. One year of work is planned into amplitude modulation to begin. After one year, following agreement among participants and work package leaders, WP 1 may recommend a change of focus. The focus will be refined during the six-monthly meetings. Normal wind turbine noise measurement processes use ten-minute averages with A-weighted sound pressure levels. See Zajamsek et al. [26] for more as well as the IEC standard and local guidelines [9][6]. This averaging does not capture, for example, amplitude modulation of wind turbine noise. An open question is how best to capture data with sufficient fidelity to allow later processing for effects peculiar to wind turbines. Several methodologies for quantifying amplitude modulated noise from turbines have been proposed. See for example papers from Sgurr Energy [23] and RES [2]. A standard approach does not exist. This is in part because how such sounds are perceived by the listener is not well understood; understanding what it is means to have an acceptable level of AM is important. These ideas are tackled in WP 3 and will inform this technical work. Building on recent work completed at Trinity College Dublin, the KTH Royal Institute of Technology and new work reported in the literature, WP 2 will work on standardising sound source, propagation and receptor models. This work will be informed by and validated against outputs from WP 3. Broadband aerodynamic noise is the most influential noise source at modern wind turbines [17]. Standard industry tools, such as NREL's FAST code, can be used to predict aerodynamic noise emission from wind turbines [10]. Much work remains, not least conducting rigorous reviews of the underlying theory that informs these codes and the addition to the standard codes of routines to capture additional phenomena such as amplitude modulation, if this is determined to be necessary. In addition to sound source models, sound propagation models must also be considered. Again, in addition to reviewing underlying theory, this work will help us understand if and how existing models are to be modified to account for additional phemomena. Aside from amplitude modulation, other possible topics for consideration include indoor sound propagation, sound propagation in complex terrain and peri-urban areas, where obstacles can shield sound or cause back-scatter [22] as well as masking effects [8][4][3]. Finally, we will work on improving and standardising receptor models using outputs from WP 3. There are variations in the sound generated by different types of wind turbine; for this reason the acoustic performance of blades is important \cite{moller:2011}. in addition, recent developments, e.g. receptor noise warranties from manufacturers, require robust understanding and an ability to tune wind turbines and wind farms where possible. This work will look to make recommendations for future development of wind turbine technology based on outputs from the other parts of WP 2 and WP 3. Again, the focus of this work package will be initially on amplitude modulation. WP3 also takes its lead from the deliberations that form part of WP1. In order to resolve the on-going debates on the sensitivity of human listeners to characteristics of wind turbine noise such as amplitude modulation and low frequency sound, definitive answers are required on the physiological response of the ear to wind farm noise. Potential mechanisms for a physiological response of the ear to low frequency sound have been proposed, e.g. in particular by A.N. Salt and others [19][21][20]. There are several important relevant studies on health and wind turbines completed and in progress [1][24]. More work is required in this field. Agreed thresholds for physiological response are needed for a robust noise metric for wind farms. This work package will include audiologists and medical experts specialising in otolaryngology. Even in the event that noise from a wind farm exceeds a threshold for a physiological response this is no guarantee that this will affect the individual in a negative manner. A consensus must also be reached on the psychological impact of wind farm noise exposure. In the past, psychological research has suffered from an incomplete understanding of the noise source mechanisms and propagation problems for wind farm noise. This work package will produce a standard database of suitable noise sources for use in participant testing. It will also aim to reach consensus on the psychological impact of wind turbine noise and threshold levels for noise annoyance. It may be interesting to investigate the use of new technology, e.g. virtual and augmented reality techniques. This work package will answer the questions of what type of noise information, presented in what format, is required to positively influence public response to wind turbines. We will combine focus group, laboratory and real world studies to assess the performance of a variety of presentation types over time. An advantage with this approach is that it gives better access to the people's holistic experiences of a phenomenon, such as noise. The conclusions drawn must be valid across social and cultural demographics. To this end studies will be conducted across partner countries and demographics. Much work has been completed in this space, including the WISEPower project [7]. Cooperation with Task 28 may be useful. The efforts of the wind turbine noise community would be greatly helped by the creation of the reliable and open dataset of wind turbine noise and other data and information. Many groups such as SMEs and academic researchers do not have the resources to develop their own datasets. In the case of noise measurement, this hinders the development of data analysis and prediction tools and also prevents correct investigation of the human response to wind turbine noise. This work package aims to collated and deploy an open and extensive data and information sets that will be available to the community. This website will also serve as a public dissemination tool for the Task and help to engage the wider wind noise community. A flexible database structure capable of incorporating data from sensors, user surveys and historic datasets is required. The objective is to provide validation data for research on wind farm noise modeling and prediction as well as a dataset suitable for use in auralisation for psychological studies. Constructing a rich, environmentally and geographically-aware database structure will allow the efficient retrieval and mining of collected data of wind farm noise which will be to great benefit to the wind farm noise community. The information in the dataset will be provided through different interfaces for different user groups (e.g. acousticians, environmental noise specialists, policy makers). The Task will reconcile an open access approach for data and results generated with the business interests of the participating industries and partners. This open dataset will be a core output of the Task for the wind farm noise community. It will facilitate further national and international research programmes as well as providing vital validation data to industry. The Task Co-ordinator and Work Packager Leaders will form a working group to provide a recommended practice document. The outputs of WP 1, WP 2 and WP 3 will be integrated and summarised over the course of the Task to provide a continuous update to the recommended practice document. This working group will engage with policy makers to facilitate the implementation of integrated policies in different regions by translating the output of the task into practical policy applications. The guideline will be the core output of the Task for policy makers and manufacturers. There may be more than one recommended practice, e.g. there may be one on the framework for wind turbine noise regulations and another for specific phenomena including amplitude modulation. At the 2015 INCE Wind Turbine Noise conference in Glasgow, a workshop about the implementation of noise standards, there was agreement that work needs to be carried out to (i) build consensus around the measurement and assessment of amplitude modulation; (ii) establish a ``dose-response'' relationship for the setting of guidelines. The Task will liaise with the IEC to contribute to the development of new standards. The task will be centrally managed by the Trinity College Dublin with work package leaders that will be identified during this drafting phase and during the scoping work package.
The task operating agent is proposed to be: The Task will be managed by TCD, representing the task to the IEA Wind ExCo. TCD's DWE team, which includes Prof. Henry Rice, has broad experience running successful technical projects including FP5 - FP7 and SEAI projects. Niall McMahon has led several industrial efforts and has directed an academic research centre. John Kennedy is an experienced scientist and research manager. See biographies for more. Overall management of the Task will be handled by the Task Coordinator who will hold responsibility for:
In addition to on-going communication about developments in the task, the official progress meetings of the consortium will be a key tool for the Task Coordinator in terms of ensuring that the IEA Wind ExCo is informed of progress. Risks primarily include time slippage and budget overspend. This will be mitigated by formation of an oversight committee that includes members of the IEA Wind ExCo. Additionally, TCD's DWE team intends to continue development of these efforts beyond this two month period as they align with our strategy. John Kennedy will act as Quality & Risk Manager to establish and validate quality control procedures; risk management procedures and risk register management. The Quality and Risk Manager will take care of the monitoring of quality control and risk register follow-up. The risk assessment and mitigation process used within the Task has been utilised within past projects coordinated by TCD. There will be two phases of the risk management process, namely risk identification and risk assessment. The probability of occurrence as well as the potential impact will be used to assess the severity of a risk. A standard tool for risk assessment, utilised by TCD within European H2020 and FP7 projects will be used. The project management review carried out in WP 0 will be responsible for the continuous monitoring and identification of risks to the task. This assessment will be led by the project coordinator in conjunction with the work package leaders. Reports will be made on a regular basis via memo. to the IEA Wind ExCo. We are defining milestones but these will include having up four meetings per annum, at least two face-to-face, with two annual reports and a final report. In addition, the aim is to produce guidelines around wind turbine noise. More detail in next draft following partner contribution. The main responsibilities of the operating agent include management, coordination and dissemination and reporting to the ExCo, in addition to technical work undertaken by the lead organisation. All of the partners are responsible for: This draft proposal was circulated to potential participants following the 77th ExCo. We will meet with potential participants at Wind Turbine Sound 2016 [25] and present at ExCo 78 seeking approval to initiate the Task with a view to a kick-off meeting by January 2017. If the ExCo accepts this proposal, the Task will begin immediately and will continue for 3 years. A kick-off meeting will be organised in the near future, ideally to coincide with a relevant conference or event. Assuming an immediate start in November, the outline timeframe will be: Further detail around timelines are in development and a draft may be ready for presentation before ExCo 78. We have budgeted for an additional ExCo meeting and Task meeting. The total costs to the operating agents for coordination, management, reporting and so on is 61,977 euro per annum over the three years, and may not exceed this level except by unanimous agreement of the partners. The budget is shared between the operating agent and the work package leaders. WP leaders receive payments once per year. To Trinity College, DWEG:
* Salary for a senior research fellow grade in Ireland is currently 65,032 euro per annum. With national social insurance, obligatory pension contributions and university overheads of 25%, this amounts to 106,286 euro per annum or 53,143 euro for half salary. This includes all coordination, management, reporting and technical work by Trinity College Dublin.
Financed by hosting organisation, i.e. not included in budget:
** It's very probable that this website can be run from a university partner at no appreciable cost. Additionally, it's probable that the website can be built for little cost by students/staff at a university partner.
As Task co-ordinator TCD will establish a web site for the task early on in the project. This will allow for both public and restricted access, the former to ensure information on the progress of the work is in the public eye and the latter to provide for an efficient means of data and information exchange between the partners and the IEA Wind ExCo.
The main results of the task will also be published in the scientific journals and presented at conferences. The consortium has an excellent track record for publication in the premium journals relating to noise, aero-acoustics, aerodynamics and human behaviour.
In addition to annual reports and end guidelines, each work package will produce reports/memos on useful findings. A proposed list of publications and conferences will be agreed following kick-off. We aim to engage closely with industry throughout the process.
The outputs of this task will integrate with world leading graduate level education at the university partners. This will enable graduate level researchers to develop skills targeted at the needs of the Wind Energy community. In addition, the project will serve to sustain the momentum of a number of significant national and international programmes which have been focused on the study of the wind turbine noise problem.
The means to reach the target audiences will be through:
Solicitation for partners and contributions is in progress and we encourage all interested to take part if possible. We hope to have participants from across disciplines and industries.
Niall is a senior research fellow at Trinity College Dublin and lead of the DWEG. He has worked building small wind turbines with Ampair in England and others and was previously director a small wind research centre in Ireland. He holds a Ph.D. in computational flow modelling from Dublin City University and a degree in mechanical engineering from Trinity College Dublin.
Prof. Henry Rice is an Associate Professor of Mechanical Engineering and has worked in the field of vibro-acoustic analysis for over thirty years. His expertise is in efficient modelling of source and propagation of sound, non-linear system identification and environmental acoustics. He has published over 80 refereed papers and has been an invited keynote lecturer at a number of conferences.
John McCann is the Vice Chair of IEA Wind and Programme Manager for Electricity and Wind with the Sustainable Energy Authority of Ireland.
[1]
R. H. Bakker, E. Pedersen, G. P. van den Berg, R. E. Stewart, W. Lok,
and J. Bouma, Impact of Wind Turbine Sound on Annoyance, Self-reported Sleep
Disturbance and Psychological Distress, Science of the Total Environment, 425
(2012), pp. 42 – 51.
[2]
J. Bass, Wind Turbine Noise - from Source to Receiver Far-field Noise Issues -
AM, Tonality and Impulses, (RES) EWEA Noise Workshop, (2013).
[3]
K. Bolin, A. Kedhammar, and M. E. Nilsson, The influence of background
sounds on loudness and annoyance of wind turbine noise, Acta Acustica united with
Acustica, 98 (2012), pp. 741–748.
[4]
K. Bolin, M. E. Nilsson, and S. Khan, The potential of natural sounds to mask
wind turbine noise, Acta Acustica united with Acustica, 96 (2010), pp. 131–137.
[5]
Council of Canadian Academies, Understanding the Evidence: Wind Tur-
bine Noise, 2015. http://www.scienceadvice.ca/en/assessments/completed/wind-
turbine-noise.aspx.
[6]
EPA (Ireland), 2011. Guidance Note on Noise Assessment of Wind Turbine
Operations at EPA Licensed Sites (NGS3) .
[7] EWEA/Wind Europe, WISE Power, Fostering Social Acceptance for Wind
Power, 2016. http://wisepower-project.eu/.
[8] B. R. Glasberg and B. C. J. Moore, Development and evaluation of a model
for predicting the audibility of time-varying sounds in the presence of background
sounds, Journal of the Audio Engineering Society, (2005), pp. 906 – 918.
[9] IEC, 2012. IEC 61400-11 2002-12 Wind Turbine Generator Systems Part 11:
Acoustic Noise Measurement Techniques.
[10] J. Jonkman, Fast. an aeroelastic computer-aided engineering (cae) tool for
horizontal axis wind turbines, National Wind Technology Center, (2015).
https://nwtc.nrel.gov/FAST.
[11] M. V. Lowson, Applications of Aero-Acoustic Analysis to Wind Turbine Control,
Wind Engineering, 16 (1992), pp. 126 – 140.
[12]M.V. Lowson, Assessment and Prediction of Wind Turbine Noise, Flow Solutions Report
19/92. ETSU W/13/00284/REP, (1992), pp. 1 – 59.
[13] H. Møller, and C. S. Pedersen, Low-frequency noise from large wind turbine,
J Acoust Soc Am, (2011), pp. 3727 – 3744.
[14] A. Moorhouse, M. Hayes, S. von Hunerbein, N. Piper, and M. Adams, Re-
search Into Aerodynamic Modulation of Wind Turbine Noise, Report URN 07/1235
(Uni. Salford), (2007).
[15] S. Oerlemans, Wind turbine noise: primary noise sources, National Aerospace
Laboratory NLR, (2011).
[16], Wind Turbine Amplitude Modulation: Research to Improve Understanding as
to its Cause and Effects, RenewableUK, (2013).
[17] S. Oerlemans, P. Sijtsma, and B. Mendez Lopez, Location and quantification
of noise sources on a wind turbine, National Aerospace Laboratory NLR, (2007).
[18] A. L. Rogers and J. F. Manwell, Wind Turbine Noise Issues, 2004. Renewable
Energy Research Laboratory, U. Mass.
[19] A. Salt and J. DeMott, Longitudinal endolymph movements and endocochlear
potential changes induced by stimulation at infrasonic frequencies, J Acoust Soc
Am, (1999), pp. 847 – 856.
[20] A. Salt and J. Lichtenhan, How does wind turbine noise affect people?, Acous-
tics Today, Winter (2014).
[21] A. Salt, J. Lichtenhan, R. Gill, and J. Hartsock, Large endolymphatic
potentials from low-frequency and infrasonic tones in the guinea pig, J Acoust Soc
Am, (2013), pp. 1561 – 71.
[22] M. Schiff, M. Hornikx, and J. Forssen, Excess attenuation for sound propa-
gation over an urban canyon, Applied Acoustics, 71 (2010), pp. 510–517.
[23] Sgurr Energy, Measurement of Amplitude Modulation Frequency Spectrum, 2012.
[24] M. Sørensen, O. Raaschou-Nielsen, R. Nordsborg, and A. Pena, Noise
from Wind Turbines and Risk of Cardiovascular Disease, 2013. Danish Cancer
Society and DTU.
[25] Wind Europe, Wind Turbine Sound 2016. Gdańsk, November 17 - 18, 2016.
https://windeurope.org/workshops/wind-turbine-sound-2016/.
[26] B. Zajamsek, D. J. Moreau, and C. J. C.J. Doolan, Characterising Noise
and Annoyance in Homes Near a Wind Farm, Acoustics Australia, 42 (2014).
The amplitude of wind turbine noise is modulated, i.e. its loudness varies with time in a regular way. This is perceived as a characteristic swooshing sound. State of the art has it that this modulation results from competing directional aerodynamic effects: stall, produced by varying angles of attack especially during the upper part of blade revolution, and inflow turbulence [12][11][16]. A recent report defines amplitude modulation as, ``a greater than normal degree of fluctuation of sound level at about once per second which makes it more noticeable'' [14]. The report goes on to note that complaints about the ``low frequency'' and ``infrasound'' (sound with a frequency of less than 20 Hz) noise from wind turbines were often about amplitude modulation, i.e. a signal that was changing more frequently than usual. The rate of amplitude modulation for wind turbine noise is relatively low but higher than for typical environmental noise sources. Amplitude modulation does not change the frequency or pitch of the sound, only its loudness, i.e. amplitude modulation is not concerned with the frequency or pitch of the source noise. In general, amplitude modulation is perceived close to the wind turbine (in the near field) and generally not perceived further away (in the far field). Far-field amplitude modulation (AM) is termed ``other'' amplitude modulation (OAM). The reasons for OAM probably include atmospheric disturbances and/or terrain effects; this requires clarification. Normal wind turbine noise measurement processes use ten-minute averages with A-weighted sound pressure levels. This averaging does not capture the modulation of the wind turbine noise. Capturing noise data with sufficient fidelity to allow later processing to identify AM/OAM at low frequency noise requires non-standard data acquisition. See Zajamsek et al. [26] for more as well as the IEC standard and local guidelines [9][16]. Several methodologies for quantifying amplitude modulated noise from turbines have been proposed. See for example papers from Sgurr Energy [23] and RES [2]. A standard approach does not exist. How such sounds are perceived by the listener is not well understood. Understanding what it is means to have an acceptable level of AM is important. Recent developments, e.g. receptor noise warranties from manufacturers, require robust understanding.3.2 WP 2: Technical Work
WP 2A: Physics of Noise
WP 2B: Human Response to Wind Turbine Noise
WP 2C: Shared Resources
3.3 WP 3: Recommended Practice
Overview
A Note on Amplitude Modulation
4 Organisation
Niall McMahon,
Distributed Wind Energy Group,
Dept. Mechanical & Manufacturing Engineering,
Trinity College Dublin,
Dublin 2
Ireland
Email: mcmahon2@tcd.ie
4.0 Management
4.1 Risk Management
4.2 Meetings, Oversight and Review
4.3 Obligations and Responsibilities
4.4 First Steps
4.5 Timelines and Dates
5 Funding and Budget
5.0 Budget
Coordination
Meetings
Website and Communications
5.1 Funding
6 Outputs
6.0 Information and IP
It shall be the responsibility of each Participant supplying proprietary information, and of the
Operating Agent for arising proprietary information, to identify the information as such and to
ensure that it is appropriately marked.
6.1 Dissemination of Results
7 Participants and Authors
7.0 Potential Participants
We expect participants from:
7.1 Proposal Authors
Dr. Niall McMahon
Professor Henry Rice
7.2.4 Mr. John McCann
References
8 Appendices
Summary Note on Amplitude Modulation