Login 
Noise and its impact on people has become a prominent sustainability issue, affecting a large proportion of the population, and a broad spectrum of society. Increasingly, evidence is emerging from research confirming associations between environmental noise and health outcomes such as hypertension, and more severe forms of cardio-vascular disease.
Consequently, management of noise features strongly in government policy and EU Directives, their effectiveness relying on a co-ordinated and strategic approach. Noise management often begins with measurement or prediction to characterise the extent of the problem, and estimate the likelihood and severity of any ensuing nuisance. Thus, it is commonplace to see noise measuring equipment around airports, on construction sites, at music festivals and in a host of other everyday situations.
New challenges in noise measurement
Modern requirements for noise measurement are not always fully catered for, however. The propagation of noise is pervasive, and its characteristics vary from place to place and with time, yet noise measurements usually have to be limited to one or a few selected locations, and are often of limited duration. The impact of noise exposure is also subjective, depending on factors such as individual susceptibility and tolerance, context, time of day, necessity and many other physical and non-physical parameters.
These subjective factors are the focus of current academic research, but no ready solution is available at present to characterise these aspects systematically. The emergence of MEMS (Micro-Electro-Mechanical Systems) microphone technology, however, and the proliferation of wireless communication offer new opportunities for widespread and long term (or permanent) distributed noise monitoring, which has not been cost-effective in the past. There is now a myriad of MEMS devices to be found in mainstream commercial products and advanced technological applications alike; for example, in mobile phones, games consoles, car air bags, and in medical robotics and energy harvesting systems.
As a consequence, and amidst a strongly competitive environment, MEMS microphones developed rapidly, leading to a range of commercially available microphone products on the market today. MEMS microphones have two significant features: they are very low cost (at £1-£2 per unit), and they have a very small form factor. The technology is also in its early stages of development, and further improvements in performance and miniaturisation are already evident. Amidst these developments, the NPL saw a role for such devices in measurement, an area it’s thought no manufacturer of MEMS microphones appeared to be considering.
When combined with appropriate signal processing electronics, readily available wireless technologies such as Wi-Fi or GSM, and web-based data visualisation applications, the potential for a powerful new set of tools for monitoring noise becomes apparent.
MEMS microphone system for remote monitoring
Systems for long term noise monitoring typically consist of a common set of components including: • A calibrated microphone with a measurement grade performance • Signal processing instrumentation for the calculation of the required acoustic parameters and indicators • Accessories for weatherproofing the microphone and electronic equipment • Hardware for mounting the equipment in a secure and unobtrusive configuration • A power supply, e.g. batteries, and/or energy harvesting systems such as solar panels • Local storage of the acquired noise data and/or telemetry for the transmission of data • Analysis software and data visualisation tools
At the heart of the system is the measurement microphone. The benefits offered by MEMS devices are set to disrupt the instrumentation technology, but MEMS microphones themselves do not have the necessary performance without modification. NPL has, however, recently been successful in adapting commercially available MEMS microphones to conform to industry standard specifications of performance (IEC 61672-1 Class 1), thus releasing their potential. While many of the remaining tools in the list above are common to a wide range of environmental monitoring applications, the process of developing an autonomous distributed noise measurement system, capable of long term deployment, reveals limitations with off-the-shelf components.
The greatest of these is the electrical power supply. Despite widespread demand for a super-capacity, ultra-compact battery power supply, the ideal component for noise measurement applications does not yet exist. All available choices involve a trade-off between capacity, cost, size, weight, and usage constraints. Latest R&D activities do, however, now indicate that the correct choice of battery coupled with energy harvesting appears to offer the scope for virtually unlimited periods of operation.
Proof of concept
MEMS measurement microphones were used for the first time in a series of trials designed to investigate the role of measurement in noise mapping exercises. A collaborative project, known as DREAMSys, included the design and manufacture of equipment that could be replicated and deployed over a wide area, to enable a measurement-based noise map to be produced.
While capable of producing these noise maps, the greatest value of the system was found to be in augmenting and improving noise predictions. More importantly, the project provided the ‘proof of concept’ that MEMS microphones could be used reliably in outdoor noise monitoring, and had an equivalent performance to conventional noise measuring systems.
Virtually unlimited possibilities
The possibility to monitor noise for prolonged periods with a distributed system of sensors, and with minimal human intervention is a significant achievement. When this can be achieved cost effectively, numerous potential applications arise across the whole spectrum of environmental, industrial, commercial and social activity. A selection of these are described as illustration. Transportation noise, especially traffic noise, is recognised as a serious public health problem.
Affordable autonomous destributed noise measurement systems provide solutions for reducing the burden associated with effective mitigation. When such systems become cost effective, both in terms of equipment costs and system maintenance, there are opportunities for increasingly widespresad deployment of monitoring systems with proportionally increased benefits to quality of life, health and economy. The challenges in delivering true cost effectiveness lie not only in the equipment costs and making systems accessible to users, but also in the costs of operating and maintaining systems, and making use of the data it yeilds.
The development of expert systems that can be accessible to non-specialist users and the general public has the capacity for significant impact and public appreciation of complex noise issues. Tools to present data visually are very effective in achieving this. While most existing noise measurement systems produce metrics relating to the energy content (e.g. the noise level in decibels), there is growing interest in using measurements to predict the perceived impact of noise, in terms of annoyance for example. This brings us back to the challenge of determining the subjective response to noise using an objective approach. The type of distributed noise measurement system considered here is well suited to being adapted for studies on noise perception and ‘soundscapes’.
Such studies enable target areas to be assessed more comprehensively to improve quality of life where the noise levels are higher than recommended by the World Health Organization, but not sufficiently high to trigger action under the requirements of the EU Environmental Noise Directive. This is a field where significantly more research is needed to understand the relationship between noise characteristics and features, and the subjective response they create. Some argue that progress towards this understanding has been constrained by an inability to measure. MEMS based measurement systems now provide the tools to enable authoritative studies to be carried out. This type of soundscape analysis can yield benefits from reduced annoyance due to noise, but also from availability to design acoustic environments to match expectations arising from the use of the area.
The concept of ‘positive sounds’ is already being used to design the appropriate soundscapes. This concept also extends to buildings designed for a specific function and enables the acoustic environment to enhance the function of the building, e.g. better recovery in hospitals, optimal learning environment in schools. The final example is not concerned with noise, e.g. unwanted sound, but with the subject many people associate most closely with the study of acoustics – the performance of auditoria. Modern design practises make good use of ray tracing software to model the acoustic performance of different auditorium shapes, designs and acoustic treatments. Often scale models are built and the sound field painstakingly probed point-by-point to build up a picture of the performance expectation.
Once the auditorium is built, however, the level of measurements undertaken to validate the performance is relatively conservative. Measurements are typically carried out sequentially at selected key locations, usually for a limited duration and most critically, when the auditorium is empty. A distributed measurement system can overcome the constraints on spatial and temporal sampling and provide a means to comprehensively validate the performance expectations, and take any remedial action to optimise the acoustic features. Indeed, the small size of the sensors provides an opportunity for an unobtrusive permanently installed system. Validation of the acoustic performance of the auditorium then becomes possible on an on going basis, and not just a matter for when the auditorium was first built. This provides exciting possibilities to tailor the acoustic characteristics for different kinds of performance (voice and different musical styles), and even to adjust the setting during performances.
While the examples above illustrate the potential of MEMS based noise measurement, everyone discovering this technology quickly identifies their own potential uses for it, which is perhaps the strongest illustration of its potential impact.
Published: 01st Sep 2012 in AWE International
