Connectivity in IoT Environment

AIoT ,Artificial Intelligence of Things is the combination of artificial intelligence technologies with the infrastructure of the internet of things to achieve more efficient IoT operations, improved human-machine interactions and improved data management and analysis.

The IoT is based on data captured by sensors, analyzed in edge, fog or cloud computing, respectively local, distributed or in the cloud, connected to an active internet, virtual or private network according to the latency determined by the application. 

 Smart devices monitor, store, communicate and, in some cases, automatically adapt to the data, taking actions. They can be found in production lines, financial equipment, cars, electronics, digital devices, commercial devices, agribusiness, healthcare, homes, cities, buildings, government, environmental control, logistics systems and countless areas frequented by consumers.

The IoT environment has its fascination, from the “thing” analysed, which is an operating system that, with integrated sensors, generates a massive amount of data; the “edge”, which is the processing place, outside datacenters and clouds, close to “things”; and connectivity, which is communication for network reach with the user. The varieties occur according to the specifics of the environment.

The IOT architecture can be classified into 3 layers: the first is the sensor/actuator device, the second is the interconnection/connectivity and the third is the application/broker, accessed by the user.

Connectivity, the subject of this introductory article, brings a lot of information: mobility, connection technology, addressing, standardization of protocols and a wide universe of network parameters to better determine the shortest possible time and the best quality between the “thing” and its connection destination.

Each solution that uses IoT has its own set of network requirements. Having a basic understanding of these requirements helps in understanding the deployment environment, determining the most appropriate connectivity design.

An intelligent IoT project usually has in its structure the following information gathered for contextualization: (1) scope and use; (2) mission critical and reliability; (3) security: authentication, confidentiality, encryption; (4) type of network, whether public or private; (5) network capacity; (6) range, speed and range; (7) implementation site coverage; (8) possibility of using multiple technologies; (9) data processing power and model, whether on-premises, distributed or in the cloud; (10) energy and battery optimization and (11) budget for available services and infrastructure.

Each solution that uses IoT has its own set of network requirements for connectivity. The main ones are: 

Network is a set of entities interconnected to each other. This interconnection allows for scalability and the aggregation of information value. The greater the number of entities, the greater the value of the network. It is found in it: client servers, connection support, data and shared devices. The internet of things requests connected devices. They may use licensed or unlicensed frequencies, and may provide prior authorization to the regulatory body, Anatel. It is expected, with the evolution of the network, the generation of a massive ecosystem for the IoT.

Broadband is high-speed internet with uninterrupted availability. It is significant connectivity, with speed according to service requests, this is said because yesterday's broadband will most likely not be tomorrow's broadband, it needs to be constantly evolving to support the desired content transmission quality. The wider the bandwidth, the more information can be transmitted. Some examples of broadband are: xDSLs, cable modem, mobile networks, radio, fiber optics, wi-fi, satellite internet. The increase in connection speed allows consumption of more complex services with less difficulty.

Latency is the network connection response time. Measures data transmission time from one point to another. It is essential for applications that demand real time. It is possible to identify latency by evaluating: propagation, throughput, delay, jitter, loss, router and ping. Decreased connection response time provides real-time interaction for services. 

Range is how far a signal can propagate without blocking. Determines scope. It is one of the main cost drivers of network infrastructure. Determines the number of network and repeat points. The higher the frequency, the shorter the range, the more network and repeat points are needed. IoT business verticals require local and wide area networks, LAN and WAN, with virtual network configurations to align network costs with application needs.

power consumption is critical for low-power devices in active mode, but also in sleep modes and startup and shutdown phases. It depends on the operational behavior of the application, on the interaction with the network. Longer battery life is required in most IoT devices, applications, network operators and service providers.

Security deals with protecting the vulnerabilities of network elements and devices from cyber attacks. Protection from ransomware, worms, DDoS (Distributed Denial of Service) and others. Fundamental to guarantee the functioning of the IoT ecosystem and avoid interruption of operations due to unavailability.

The evolution of cellular networks has accompanied IoT solutions. There have been points of attention: the high cost of internet plans, high battery consumption and coverage gaps in underground and remote locations. 5G brings the promise of more connected devices, higher speed, lower latency, security, battery savings and more. 

IoT needs one IP address per device. IPv6 comes to increase the addressing capacity and thus promote the connectivity of more things on the internet. 

It is essential to seek timely connectivity adapted to each need:

And it is in this universe that the I2AI AIoT committee focuses on its studies, meetings and discussions, bringing experiences from the professional and academic day to day of the group, invited colleagues, the literature and AIoT industry, with the objective of generating content that is easy to understand for the group and interested public of Brazilian society.



Glossary

Denise Nobre is a member of the I2AI AIoT Robotics committee. Telecommunications Engineer from UFF- Universidade Federal Fluminense, with an MBA in Business Management from IBMEC-RJ, certified as a DA Senior Scrum Master by PMI – Project Management Institute, specializing in information technology and telecommunications engineering project portfolio management, finishing TCC in postgraduate studies in New Technologies, Digital Transformation and Agility at FIA-SP. He worked for 20 years with leadership and team building in project management, programs and product portfolios and mobile, fixed and satellite services.