Signal processing is a science that has unlimited applications. Everyone from the U.S. military to audio sound engineers to the healthcare industry deal in the signal processing arena each and every day. So vast is the power of signal processing to completely change an industry, or change the way we live our lives, it is not quantifiable. And yet, the potential of this process is continually underestimated. To fully utilize the power of this digital transformation, it is important to understand how it works.
If you were to mention Joseph Fourier to an average passerby in a crowd, you would likely get very little reaction. Doing the same with the engineers at Rock West will immediately pique their interest and likely prompt a discussion about how the Fourier transform (FT) is at the heart of signal processing today. The modern world is bursting with signals that are virtually meaningless until they can be processed or transformed. This game-changing mathematical method long predates a smartphone however. Though applications for this transform were few during its inception in the 19th century, it became the foundation upon which modern signal processing was built. In the 1960’s the Fast Fourier Transform (FFT) was developed, enabling efficient solutions of the discrete FT for practical problems. Limited to critical applications initially, the increased access/affordability of computers necessitated the broad and frequent use of this algorithm .
Understanding Signal Processing
Signals come in a variety of formats, from sound to vibrations to moving images. The human brain is an excellent processor for our world of analog data, which comes to us through our sensory organs. A sensation is converted into electrical signals that are transmitted via our nervous system for the brain to interpret and respond accordingly. This process, which occurs so effortlessly via our analog signal processing brain, has been the focus of decades of research and development in order to replicate and exceed a human’s capabilities on digital computers. To achieve this, an analog signal needs to go through a few more steps of processing. First a transducer converts the analog input into an electrical signal. Next it is digitized, at which point it can be processed further depending on the application . These applications are both numerous and highly varied. See Figure 1 from for some examples.
Figure 1 Examples of applications for Digital Signal Processing (DSP) 
Gleaning useful information from signals often requires the use of filtering and/or spectral analysis functions. Filtering is optimized based on the processing end goal, while spectral analysis is concerned with the frequency makeup. See Figure 2 for a visual representation of the difference between the two . The Fast Fourier Transform (FFT) is an essential tool used to take a signal from its original domain (most often the time domain) and convert it into or out of the frequency domain [Figure 4]. Breaking a signal down into its frequency components can make it easier to filter, manipulate, and interpret. Think of the signal like a smoothie. All the individual components (yogurt, fruit, ice) are blended together and it is very difficult to know what they are and how much of each is present. If you could process it with a single tool that could identify every ingredient and the corresponding quantity, you would be able to easily record the recipe for recreation, alter it, or compare it to others. FFT’s are an important aspect of signal processing, but they are only one aspect of a complex and growing field of signal processing tools. These tools can be utilized by industry in a variety of ways.
Figure 2 Differentiating between filtering and spectral analysis processing 
Figure 3 Visualization of a Fourier Transform 
The Power of Signal Processing
Extolling the potential uses for signal processing among all applications would be next to impossible because they are simply too vast. As an example of this power, consider the effect of this field on ‘personal auditory devices’ (PAD), the modern generation of hearing aids. A typical PAD recommend by an audiologist can create a customized hearing environment that addresses an individual’s hearing difficulties in a much more dynamic and functional way.
Let’s break down some improvements signal processing can has contributed to PAD technology:
- Amplification – The need for amplification has not changed, but now it can be applied strategically so that specific frequencies are amplified rather than just all local sound.
- Compression – PADs must process information as quickly as possible to be used in real-time. Signal compression simplifies the data so that it can be signal processed as efficiently as possible.
- Digital Noise Reduction (DNR) – This form of signal processing identifies and reduces background noise. It can greatly improve hearing for those people who have trouble distinguishing between various sounds in a room.
- Directional Gain – We can now combine tiny directional microphones with signal processing to restore a person’s ability to focus on sounds coming from a specific direction.
- Feedback Reduction – Feedback has long been a problem with hearing aids. New signal processing methodologies are using digital filtering to make feedback a thing of the past.
- Noise Cancellation – It is possible to completely cancel out certain sounds by creating identical signals with opposite phase. Canceling out certain signals can improve hearing drastically.
- Frequency Control – We can use both frequency transposition and compression to change the frequencies of certain signals. For example, transposition can take a high frequency signal that cannot be heard and lower its frequency to a level that enables a person to hear it.
If this melding of electrical engineering, mathematics, computer science, and product design can have such an impact on hearing assistance, one can only imagine the potential is has on a larger scale. Our world implements this technology in everything from Snapchat filters to gastrointestinal tracking devices. Whether your project falls in the realm of medicine, defense, or commercial products, maximizing your implementation of signal processing is essential for optimal results. Our team at Rock West Solutions has decades of experience helping our clients work with a broad range of signal varieties and applications. Contact us today to discuss how we can make signal processing work for you.
- 1 Smith, Steven W. “The scientist and engineer’s guide to digital signal processing.” (1997): 35.
- Unversity of Hertfordshire http://www.yildiz.edu.tr/~kunal/dsp1.pdf
- Dr. Michael Piovoso https://www.swarthmore.edu/NatSci/mzucker1/e15_f2014/E15Week11_DSP.pdf