Reading Time: 7 minutes

“I could be the ceaseless mist that fogs your colourless eyes when you’re lost in your universes.”
Moonshine Noire

Most readers are well aware of Cloud Computing, some may have even heard of Fog Computing already, but have you considered Mist Computing yet? In the information technology industry, we liberally borrow terminology from other industries to help us to explain our complex technical concepts and ideas. This use of analogous language permits us to convert abstract ethereal ideas into concrete and tangible thoughts that most visual learners can easily comprehend. Since over 80% of us are visual learners, it is import to use analogies derived from the physical world to help explain these technological approaches.


First, let us consider the source of these terms from the world of meteorology and weather forecasting.

  • Cloud – a visible mass of condensed water vapor floating in the atmosphere, typically high above the ground
  • Fog – a thick cloud of tiny water droplets suspended in the atmosphere at or near the earth’s surface that obscures or restricts visibility (to a greater extent than mist; strictly, reducing visibility to below 1 km)
  • Mist – a cloud of tiny water droplets suspended in the atmosphere at or near the earth’s surface limiting visibility, but to a lesser extent than fog; strictly, with visibility remaining above 1.5 miles (1 km)


Now, if we consider these weather definitions from an internetworking perspective, what we can appreciate are differences based upon density, proximity, visibility, mixing, time, and location.

  • Cloud Computing – the practice of using a network of remote servers hosted on the Internet to store, manage, and process data, rather than a local server or a personal computer, Cloud Computing can be a heavyweight and dense form of computing power
  • Fog Computing – Fog computing is a term created by Cisco that refers to extending cloud computing to the edge of an enterprise’s network. Also known as Edge Computing or fogging, fog computing facilitates the operation of compute, storage, and networking services between end devices and cloud computing data centers. It is a medium weight and intermediate level of computing power
  • Mist Computing – a lightweight and rudimentary form of computing power that resides directly within the network fabric at the extreme edge of the network fabric using microcomputers and microcontrollers to feed into Fog Computing nodes and potentially onward towards the Cloud Computing platforms

While we can use Cloud Computing for any applications, Fog Computing, and Mist Computing define topics primarily related to the Internet of Things (IoT) terminology due to the nature and ballistics of the traffic flows. This multiple staged network is a federated network since it has different functionality at each of the discrete, yet disparate stages. A federation is a group of networks agreeing upon standards of operation in a collective fashion. The term may be used when describing the inter-operation of two or more distinct, formally disconnected, telecommunications networks that may have different internal structures. The term may also be used when groups attempt to delegate collective authority of control and traffic flows to prevent fragmentation.

In general, IoT traffic is of a very low data rate, typically measured from as low as 1 kbps to as high as 300 kbps. Often, the datagram traffic is sized for a single frame-by-frame design to meet the desired real-time aspects, thus avoiding delays due to datagram reconstruction from multiple packets. The Internet of Things is a Layer 3 technology, as it uses IPv6, which is the de facto standard for IoT, so it is packet based. The latency can vary depending upon the topology used, but it runs from a low of 10 ms to a high of 1,000 ms per hop, with 100 ms considered average latency. For the access tier of the IoT topology, it is considered to be wireless, in the form of a star, mesh, or cluster tree. However, it can make use of a wired solution too, albeit not normally, but again with low data rates and low to ultra low latency.

The Internet of Things supports several architectural models, unlike other networks that typically support just one model. The IoT can use a centralized, distributed, or federated model. The fundamental difference between these three models is based upon where the intelligence is located, in the data center for the centralized model, at the network edge for the distributed model and at both for the federated model.

The IoT architecture therefore supports the client-server design and the peer-to-peer design simultaneously, which greatly enhances its capabilities.

In classic Machine-to-Machine (M2M) networks, all of the data is sent to the center of the network in the data center. However, in the IoT network, some data goes to the center and other data just resides at the edge or on the network fabric. Other data may come from or go to unrelated data centers operated by third party users. Therefore, the edge data may or may not go to the center of the network. There may never be any need for it to go to any data center at all. Yet, some edge data like smart meter reads for billing will most certainly journey to the data center.


Some of the edge data is not actually gathered from devices, such as smart meters, but is derived and provided from device to device just when it is necessary in an “on-demand” approach. In a smart grid, this data may be from a device that is informing nearby associated devices of its status. An example might be a recloser indicating if it is open or closed to the next recloser on the feeder or to a controller located on the relevant feeder line.

Some extreme edge data may only traverse the network as far as the local substation and communicate with a substation controller managing voltage, control, or other parameters on a feeder or orchestrating a group of associated feeders from the same or neighbouring substations. The controller may then aggregate data for forwarding or derive new data that is forwarded to the master controller, which is centrally located at the data center.  It may not share any data upstream as well.  In this case, we are describing Mist Computing connecting only as far as the Fog Computing.  The Fog Computing may or may not communicate to the Cloud Computing.

Therefore, the architecture of the network plays a key role in meeting the needs of the applications that operate over it. The architecture maps to the applications and is configured to operate in a local, regional, or network-wide level.

Node and Connections

As with all networks, the IoT networks are composed of nodes and connections. The nodes are the devices, which transmit or receive the datagrams. The connections are the means for joining the nodes into a holistic network fabric.

There are different kinds of nodes that perform different work within the smart IoT network. Many nodes, such as smart metering devices in an electrical utility IoT network simply provide power consumption data at some prescribed interval, often at the top of each hour for 24 reads per day, but it can be as granular as every 5 minutes for 288 reads per day. There are different types of smart meters and we are beginning to see an emergence of smart meters that provide comprehensive power quality readings. These meters continuously measure and monitor frequency, voltage variation, dips and swells, voltage outages, voltage unbalance, total harmonic distortion (THD), and power factor. Therefore, the data payload per transmission is much greater than the initial consumption smart meters or time-of-use smart meters. Other devices provide alarms and status monitoring conditions. These can be from reclosers, segmentation switches, tie switches, and other similar devices. Smart grid devices like cap banks, power line monitors, and transformer monitors all exchange data too.

Nodes can be simple sensors, smart sensors with compute power built into the sensor itself, actuators that respond to command and control communications, or microcomputers and microccontrollers that aggregate and manage data from the local sensors and actuators.

As stated, these connections are normally wireless connections, but some wired connections can also be a part of the IoT model. Wireless uses different frequency bands in different countries. In North America, the 902-928 MHz band is common for electrical solutions. Many water metering and gas metering nodes use spectrum in the 220 MHz and 450 MHz bands.  In Europe, the new 860-867 MHz is considered the ideal spectrum for smart grid use. Some European companies are using the 1 MHz channel at 868 MHz too. In some South America countries, the same 902-928 MHz band as is used in North America is applied, but with band power level and time duration restrictions and partial band exclusions for a middle block reserved for other purposes. In Japan, the 920-928 MHz band and the 950-958 MHz band are used, while in China, the 220 MHz band, 470-510 MHz band and the 779-787 MHz band are popular. Korea used the 917-923.5 MHz band. Worldwide the 2400-2483.5 MHz band is used too. Other bands are used in various countries that may require special approvals, but these are the most popular blocks of spectrum available for mass deployments.

The industrialized flavour of the Internet of Things is still evolving and being defined by its uses and users. There is still much work ahead that needs to be done to complete the final version, including work efforts focused on security, standards, protocols, and ever changing definition of IoT. It is still very much a technology that is in a state of fluctuation. So, consider this summary to be a draft of the current reality as it will change to reflect the future applications. Stay tuned.


About the Author:

Michael Martin has more than 35 years of experience in systems design for broadband networks, optical fibre, wireless and digital communications technologies.

He is a Senior Executive with IBM Canada’s GTS Network Services Group. Over the past 13 years with IBM, he has worked in the GBS Global Center of Competency for Energy and Utilities and the GTS Global Center of Excellence for Energy and Utilities. He was previously 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 currently serves on the Board of Directors for TeraGo Inc (TGO: TSX) and previously served on the Board of Directors for Avante Logixx Inc. (XX: TSX.V). 

He serves 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) and on the Board of Advisers of five different Colleges in Ontario.  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 diplomas and certifications in business, computer programming, internetworking, project management, media, photography, and communication technology.