Most IoT devices are often powered by an energy storage system (e.g., a battery or a capacitor/supercapacitor/ultracapacity) with limited energy capacity. When all the energy stored in the energy storage system is depleted or consumed, the device will shut down if the energy storage system is not changed or recharged. Changing or recharging the energy storage systems of IoT devices in an IoT infrastructure with hundreds, thousands, or tens of thousands of IoT devices may be tedious and costly. The limited available energy to power the IoT devices constrain the design choices for IoT devices. That is, the primary goal of IoT designers is to minimise the energy consumption (or to maximise the energy efficiency) of IoT devices. The following IoT choices are often made to minimise the energy consumption (or to maximise the energy efficiency) of IoT devices and deal with other device limitations:
When designing IoT systems, the design choices should be made in such a way as to maintain a balanced trade-off between energy consumption, quality of service, and security. Since most of the hardware limitations result from the need to minimise energy consumption, incorporating energy harvesting systems to harvest energy from the environment to supply the IoT devices and recharge the energy storage systems in IoT devices is essential. It will relax the energy consumption constraint and increase the functional capabilities of IoT devices.
Data is the lifeblood of IoT systems, and robust connectivity is required to facilitate data transfer between IoT devices, networking nodes, computing devices (fog nodes and cloud data centres) and applications. Some IoT applications require real-time monitoring and control of systems and processes to improve efficiency and productivity and facilitate decision-making processes. Most IoT devices are connected to the Internet through wireless access networks, which are highly complex, especially in urban areas with dense deployments of wireless networks. Ensuring reliable and efficient connectivity between the various systems in an IoT infrastructure (from the IoT perception layer to the application layer) is a challenging problem for IoT designers and developers.
In order to minimise the energy consumption of IoT devices to prolong (maximise) their lifetime, low-power communication and networking protocols and technologies have been adopted. However, the energy-efficiency or energy-saving goal is achieved at the expense of performance (networking reliability, throughput, packet delays, packet losses, packet collisions in shared wireless channels). Achieving a reasonable balance or trade-off between energy consumption and network performance is challenging but crucial when designing IoT networks.
Apart from the technical issues of network reliability and quality of services, the cost of connectivity is also essential, especially for small and medium-sized businesses. The higher cost of Internet connectivity and maintenance of the local IoT network will make the cost of operating the IoT infrastructure very high, reducing the return on investments and discouraging the adoption of IoT solutions. Therefore, the networking solution or technologies chosen when designing IoT networks should be reliable and relatively cheap to provide the users with reliable connectivity at a reasonable cost.
Most IoT devices are powered by cost-effective, small-sized batteries with limited energy capacity. Recent advances in low-cost and low-power IoT technologies have enabled the cost-effective, energy-efficient, data-driven, and flexible automation of cyber-physical systems. However, the energy required to power these systems and related infrastructures will be enormous when hundreds of billions of IoT devices are connected to the Internet, increasing the carbon footprint of the IoT industry and sustainability concerns.
In order to ensure that IoT devices are small and cheap for commercial deployment in large numbers, they are generally designed to have limited battery capacity, low computational power, limited memory, and use low-power networking and communication protocols and technologies. The essence of minimising energy consumption is to prolong the battery life and the lifetime of IoT devices (the time required to deplete all the energy stored in the battery of the IoT device), minimising downtimes and the cost of recharging or replacing IoT batteries.
There are increasing concerns about the resource sustainability and environmental impact of deploying billions or trillions of IoT devices in various sectors (e.g., intelligent transport systems, smart health care, smart manufacturing, smart homes, smart cities, smart agriculture, and smart energy) of the society or economy. There is a growing demand for energy to power the small IoT devices, transport the IoT traffic through the Internet, and analyse the massive amounts of data generated by the IoT devices at the fog nodes or cloud data centres, increasing the carbon footprint from IoT-based services.
One way to reduce the carbon footprint of the IoT industry and increase its sustainability is to incorporate energy harvesting in IoT systems and infrastructure. Although energy harvesting reduces some of the energy and sustainability challenges in the IoT industry, it has its challenges. One of the challenges with energy harvesting is the intermittent nature of the energy source, as energy availability depends largely on environmental factors that fluctuate significantly. Because of the need to keep the IoT devices or systems small and light and the low energy density of some energy sources, energy harvesting is sometimes minimal and barely satisfies the needs of IoT systems. Thus, one design challenge is to size the energy harvesting systems, the energy storage systems, and the energy demand of IoT systems in such a way as to ensure reliability (reduce related downtimes or outages).
Initial research in IoT was focused on developing proprietary solutions that were vendor-specific [124]. However, some efforts have been made by the Internet Engineering Force (IETF) to standardise some IoT protocols, such as IPv6 over Low-Power Wireless Personal Area Networks (6LowPAN) [125], Routing over Lower power and Lossy networks (ROLL) [126] and Contraint RESTful environment (CoRE) [127] by its 6LowPAN, ROLL and CoRE working groups. Using heterogeneous IoT devices, protocols, and platforms developed by different vendors could result in inadequate regulations, standardisation and governance in the IoT domain [110]. These result in multiple security vulnerabilities and difficulties scaling up IoT application deployments. It will also slow down the rate of adoption of IoT in the food and agriculture industry, but adequate regulations, standardisation, and governance will enable interoperability, scalability and development of IoT solutions, taking into consideration security requirements.
The debate about the ownership of data in the context of IoT is still an open issue, and it is very controversial. This is because there are different stakeholders in the value chain of the IoT ecosystem. The stakeholders in the IoT ecosystem and in the various sectors where IoT systems are being deployed to solve specific problems or create value are very concerned about privacy, ownership, the possibility of losing control over their data and the fear that their data could be misused. The question is whether the data belongs to the users (individuals or businesses using IoT services), IoT network provider, cloud platform provider or cloud infrastructure provider.
Resolving data ownership issues and creating data protection plans or strategies is challenging for IoT designers as it involves technical, legal, and ethical issues that are sometimes very complex to resolve. The IoT data management plan and strategy should be developed during the design and deployment stage, covering aspects such as data types, data lifecycle, data governance, and data security. Ownership of measurement data collected from the environment, data about the IoT devices, the data about the users of the devices, and the metadata (e.g., location data, timestamps, and labels) is critical, as well as how the data is stored and used.
Another challenge is to develop a data governance plan or strategy. That is, to develop a set of policies, procedures, regulations, and standards that should be followed when addressing issues regarding data ownership, data access, and data sharing within an organisation or across organisations. The objective is to ensure that IoT data governance policies, procedures, regulations, and standards are clear, fair, and simple to guarantee consistency, transparency, and accountability among all the stakeholders (e.g., data owners, users, service providers, and regulatory agencies).
It is also challenging to ensure data privacy (protection of the IoT data) throughout all the stages of the data life cycle, from the data collection, transmissions, processing, analysis, storage, and deletion (if necessary) stages. At every stage of the data life cycle, it is important to ensure data security. That is, to ensure that the data is kept confidential, to maintain data integrity, and to ensure that it is readily accessible or made available to relevant stakeholders. Data protection plans should be designed to ensure the timely detection and prevention of data security breaches. However, it is challenging to guarantee long-term data security due to the rapid changes in the cybersecurity ecosystem and attack landscape.
Cost is one of the factors that influence the adoption of IoT systems or solutions and should be considered when designing IoT systems or infrastructure. Satisfying all the IoT design goals such as performance (QoS and reliability), security, and energy and sustainability increases design, deployment, operation, and maintenance costs, which may influence the perception and adaption behaviour of the users. Users are often very sensitive to prices such that a slight increase in price could influence their decision to adopt and pay for the IoT production or solution.
In order to ensure relatively lower prices for IoT devices, IoT manufacturers tend to sacrifice some important design goals. For example, Some IoT device manufacturers tend to use weak security mechanisms or ignore the implementation of IoT securities altogether, making such IoT devices easy targets for hackers to exploit and conduct cyber attacks on the IT infrastructure of organisations and the Internet infrastructures.
Also, using cheaper batteries with smaller energy capacities and shorter life cycles reduces the cost of IoT devices. Still, it increases the cost of operations and maintenance as batteries need to be changed regularly. Cheaper and low-quality IoT devices also increase operations and maintenance costs. Therefore, IoT designers, manufacturers, and developers should not only prioritise the cost of devices and a shorter time to market but should ensure high-quality devices to avoid the high cost of operations and maintenance. It is very challenging to satisfy all the design goals while keeping the cost of the devices and the time to market at a reasonable level.
When designing and developing IoT applications, it is important to take into consideration those factors that could affect the acceptance of these solutions by the stakeholders. Some of these factors include perceived usefulness, perceived ease of use, perceived cost, perceived cost, perceived compatibility, perceived value, [5], perceived behavioural control, social influence, trust [129], and the challenges discussed above. If these issues are not addressed in the course of development of the solutions, then adoption may be slow. Stakeholders may need to understand the capital expenditure (CAPEX) and the operational expenditure (OPEX) that IoT will be added to their business, its value and the return on their investments.