consists of a variety of a smart sensors like smart lock for keyless entry in car and homes, photo-director for turning garden light on at night and off at the day, sensor for detecting movement in room and triggering alarm when an intrusion happens, wireless camera with internet access to remotely monitor home and other premises, smart thermostat for controlling temperature according to human need, and smart fire alarm system for triggering alarm when a thing catches fire; IoT sensors can also be used in smart traffic management system for monitoring vehicle tracking on roads and switching traffic light according to traffic need. IoT sensors can also be used in warehouses for tracking workers activities. The increasing demand of smart devices in various fields creates huge amount of data and these data falls under various risk and challenges, if does not handled properly (especially health-related data). Patient data generated by smart heath sensors are very sensitive and critical, and management of these data is quite challenging Alihamidi et al. [32]. Every healthcare system must be provided security mechanism to handle five major factors in data handling: availability, integrity, confidentiality, authentication, and non-repudiation.
Figure 2.1 Healthcare using IoT.
2.2 Related Works
A paper proposed by Sun et al. [1] named “Security and Privacy in the Medical Internet of Things: A Review” was published in March 2018. In this paper, various security issues are discussed about health-related data which are traveling over the internet. Their main focus was data usability, integrity, and auditing.
They also talked about different existing techniques available like encryption and access control.
A survey paper by Zhang et al. [2] on “Security-Aware Measurement in Software-Defined Networking (SDN)” consists of SDN basic architecture, issue in security, performance analysis, bandwidth analysis, topologies, and future scope.
A paper named “Privacy-Preserving and Multifunctional Health Data Aggregation With Fault Tolerance for Cloud Assisted WBANs” written by Han et al. [3] discusses privacy preserving technique in wireless body area networks.
Abuwardih et al. [5] presented a paper on privacy preserving in healthcare data; in this paper, they discussed various types of attacks and privacy issues related to patient data. They also proposed some architectures and procedures to handle different type of attacks related to patient data. The paper was named “Privacy Preserving Data Mining on Published Data in Healthcare”, and it was proposed in 2016.
Anwar et al. [6] proposed a paper in 2015 named “Anytime Anywhere Access to Secure Privacy-Aware Healthcare Services: Issues Approaches and Challenges”; this paper consists of various approaches and challenges arises in healthcare industry for providing anytime and anywhere access of health-related resources. In this paper, they have shown approaches that are currently available and also discussed different policies made by government for information and technologies–related and international data–related security issues. The issues are generated by human, machine, and some other factors. All the security-related concern is discussed in detailed manner.
Rahman et al. [18] published a paper in ICOST (International Conference on Smart Homes and Health Telematics) naming “Inclusive Society: Health and Wellbeing in the Community and Care at Home”. In this paper, they have proposed a generic model “PriGen” for securing patient health-related data with the help of cloud storage. PriGen facilitates access of patient health-related data without involvement of any other party from the cloud as well as hiding highly sensitive data. PriGen uses homomorphic function for encryption of data that needs very high security and which is critical. This algorithm preserves the highly sensitive health-related personal data in public cloud environment and maintains confidentiality of the patient data.
Bindahman et al. [19] proposed a paper in 2011 ICIEIS (Informatics Engineering and Information Science: International Conference) in which they talked about general concept of privacy problem related to patient health data. They discussed various available security measures and its performance comparison related to healthcare data. Based on that, they also suggested some techniques for those security-related problems.
Next is the paper written by Dubovitskaya et al. [20] entitled “ICT Systems Security and Privacy Protection: 30th IFIP TC 11 International Conference SEC 2015”. In this paper, they have discussed various problems in building of health-related database for heterogeneous environment, where data are coming from various sources of different network environment of different hospitals. The integration of data comes from different locations. They introduced scaling and securing techniques for patient e-health data. They have used an algorithm called RSDB (Representative Protein Sequence Database) for collecting patient data efficiently and securely that are coming from various sources.
A paper by Idoga et al. [22] has discussed different issues related to privacy in the application e-healthcare environment.
2.3 Architecture
In this section, we will discuss three-layer architecture of next-generation healthcare industry with smart sensors, fog node, and cloud computing. The new healthcare industry will change the way hospital staff or doctors treat the patient. This IoT-driven healthcare industry will get highly efficient environment at very low cost, which decreases the workload and increases the throughput. The layered architecture will do different tasks at different layer. The three layers are device layer, fog layer, and cloud layer as shown in Figure 2.2.
2.3.1 Device Layer
At this layer, a very large number of smart sensors are involved, which are gathering tons of health-related data in near real-time. Any patient or healthcare specialist can access these data using any web-enabled device like phone, tablet, or computers. What they require is a secure and stable communication protocol to next layer in this case (fog layer). There is various communication are available for the wireless sensor nodes for communication within each other or propagating information to next layer. But selection of best protocol from the pool of various protocols is a tedious task. There are some protocols which are widely adopted for some general data transfer tasks are Low-Energy Bluetooth and High-Fidelity Wi-Fi.
Figure 2.2 Three-layer architecture.
Low-Energy Bluetooth is good for beacon type signal and where power constrains matter too much or battery is irreplaceable (like in heart pacemaker). On the other hand, Wi-Fi is used where long-range and high-data rate required (like transferring raw ECG data).
2.3.2 Fog Layer
This layer consists of high-end processors, large and high-speed data storage, and Network Interface Card for communication over internet. The patient health data are very critical data a normal considerable amount of latency can cost lives, example of such type of data is Myocardial Infarction where latency of seconds can cause serious damage Singh et al. [26]. So, we cannot rely only on cloud for data processing and analyzing critical time sensitive data. In handling these types of data, we require analysis, processing, and storage as close as possible to the devices where data is generated Badidi et al. [23]. Thus, need of fog computing arises here, and it processes, analyzes, and stores (for further use) health-related data which is very time-sensitive in nature Kraemer et al. [28]. This layer also filters, compresses high volume raw data, and passes to next layer (cloud computing layer) for big data analytics purposes.
