“Smell is chemistry in motion. When IoT sensors learn to read the invisible signatures in the air, odour becomes data, and data becomes early warning.” – MJ Martin
In a past work life I participated on teams that recreated the five human senses – taste, hearing, seeing, feeling, and smelling.
Two significant projects come to mind to explain the use cases – one was to detect mould odours and another was to measure smells permeating from a major wastewater treatment plant.
At a recent work luncheon, we discussed these projects, but one of the respected experts was unclear how we could possibly recreate the sense of smell. At that time and place, I failed to explain it all very well. So, here is a better explanation that I hope makes sense – pun intended. lol
Introduction
Yes. An IoT sensor system can detect smells, but it does not “smell” the way a human nose does. It detects airborne chemicals associated with odours, then uses software to interpret the chemical pattern. The more accurate term is usually an electronic nose, or e-nose.
A smell is normally caused by a mixture of volatile compounds in air. One sensor may detect one gas, but a realistic odour is usually a chemical fingerprint made from many compounds. That is why serious odour detection often uses a sensor array plus AI, pattern recognition, or statistical classification. E-nose systems commonly combine a sampling system, a detection system, and a computing system that classifies the odour signature.
Common Sensors Used to Detect Smells
Metal oxide semiconductor sensors, or MOS sensors, are probably the most common low-cost option. They detect gases and volatile organic compounds by measuring how the electrical resistance of a heated metal oxide surface changes when exposed to chemicals. They are compact, sensitive, and affordable, which makes them attractive for IoT devices, although they can drift over time and are affected by humidity and temperature.
Electrochemical sensors are used when the target gas is more specific. These are common for gases such as hydrogen sulfide, ammonia, carbon monoxide, chlorine, nitrogen dioxide, and ozone. For example, rotten egg odours are often associated with hydrogen sulfide, while agricultural, wastewater, or decomposition odours may involve ammonia or sulfur compounds. Electrochemical sensors are often more selective than MOS sensors, but they usually target a narrower range of gases.
Photoionization detectors, or PID sensors, are used to detect many volatile organic compounds, often called VOCs. These are useful where smells come from solvents, fuels, chemical processes, coatings, adhesives, or industrial emissions. A PID does not usually identify the exact compound by itself, but it can detect the presence and relative concentration of VOCs.
Non-dispersive infrared sensors, or NDIR sensors, detect gases that absorb infrared light, such as carbon dioxide and methane. These are not “smell sensors” in the strict sense, because methane and carbon dioxide are odourless, but they are useful in applications where odour is associated with gas leakage, fermentation, wastewater processes, landfill gas, or indoor air quality.
MEMS gas sensors are miniaturized sensors designed for compact IoT devices. They are often based on MOS or related sensing principles. They are useful where size, power consumption, and wireless deployment matter.
Acoustic wave sensors, conducting polymer sensors, field-effect gas sensors, and optical sensors are also used in more advanced e-nose systems. These are less common in basic IoT deployments but appear in research, laboratory, food quality, medical, and industrial odour-classification systems.
Typical Sensor Combinations
The Best Approach: Sensor Fusion
The most capable system is not one sensor. It is sensor fusion. You install several partially selective sensors, collect readings over time, and train software to recognize patterns. For example, a wastewater lift station odour profile may show elevated hydrogen sulfide, humidity, temperature, and VOC activity. A garbage odour may show a different VOC and ammonia pattern. A natural gas event may require methane detection, while the human-recognized “gas smell” may come from added odorants rather than methane itself.
This is where AI or machine learning becomes useful. The system learns that “smell A” produces one multi-sensor signature and “smell B” produces another. Current e-nose research increasingly combines sensor arrays with AI methods such as principal component analysis, neural networks, and classification models.
Practical Applications
IoT smell detection can be used for wastewater treatment plants, sewer odour complaints, landfill monitoring, natural gas infrastructure, food spoilage detection, indoor air quality, industrial emissions, agricultural operations, cold-chain monitoring, and leak detection. In municipal utility work, the strongest applications would likely be sewer odour detection, hydrogen sulfide monitoring, wastewater lift stations, landfill gas, gas leak correlation, and indoor air quality monitoring in facilities.
Important Limitation
An IoT smell sensor does not usually tell you, “this smells bad.” It tells you that certain gases or VOC patterns are present. Human smell is subjective and extremely sensitive to some compounds at very low concentrations. Sensor systems need calibration, drift correction, placement discipline, and usually field training against known odour events. EPA air sensor guidance emphasizes planning, siting, data quality, and fit-for-purpose use when deploying air sensors.
So the direct answer is: yes, IoT systems can detect smells, but they do it by measuring chemical signatures. The best sensor suite would usually include MOS VOC sensors, electrochemical H₂S and NH₃ sensors, possibly a PID VOC sensor, NDIR methane or CO₂ sensors where relevant, and environmental compensation sensors for temperature and humidity.
About the Author:
Michael Martin is the Vice President of Technology with Metercor Inc., a Smart Meter, IoT, and Smart City systems integrator based in Canada. He has more than 40 years of experience in systems design for applications that use broadband networks, optical fibre, wireless, and digital communications technologies. He is a business and technology consultant. He was a senior executive consultant for 15 years with IBM, where he worked in the GBS Global Center of Competency for Energy and Utilities and the GTS Global Center of Excellence for Energy and Utilities. He is a founding partner and President of MICAN Communications and before that was President of Comlink Systems Limited and Ensat Broadcast Services, Inc., both divisions of Cygnal Technologies Corporation (CYN: TSX).
Martin served on the Board of Directors for TeraGo Inc (TGO: TSX) and on the Board of Directors for Avante Logixx Inc. (XX: TSX.V). He has served as a Member, SCC ISO-IEC JTC 1/SC-41 – Internet of Things and related technologies, ISO – International Organization for Standardization, and as a member of the NIST SP 500-325 Fog Computing Conceptual Model, National Institute of Standards and Technology. He served on the Board of Governors of the University of Ontario Institute of Technology (UOIT) [now Ontario Tech University] and on the Board of Advisers of five different Colleges in Ontario – Centennial College, Humber College, George Brown College, Durham College, Ryerson Polytechnic University [now Toronto Metropolitan University]. For 16 years he served on the Board of the Society of Motion Picture and Television Engineers (SMPTE), Toronto Section.
He holds three master’s degrees – in business (MBA), communication (MA), and education (MEd). As well, he has three undergraduate diplomas and seven major certifications in business, computer programming, internetworking, project management, media, photography, and communication technology. He has completed over 80 next generation MOOC (Massive Open Online Courses) [aka Micro Learning] continuous education programs in a wide variety of topics, including: Economics, Python Programming, Internet of Things, Cloud, Artificial Intelligence and Cognitive systems, Blockchain, Agile, Power BI, Big Data, Design Thinking, Security, Indigenous Canada awareness, and more.
Martin in a volunteer, a photographer, a learner, a technologist, a philosophizer, and a romantic optimist.