Passage C IoT Endpoint Monitoring Systems

A use case of IoT-based endpoint monitoring systems will be discussed in this section.IoT endpoints (sensors and actuators) are of various types and capabilities distributed where Iots of systems are deployed.Many of these endpoints serve single purposes and are located in remote locations.These IoT endpoints are often fragile,low power but serve critical functions in the data or information they gather,transmit or services they provide in the IoT system.IoT endpoints are often too frail to report their own state (power/batterylife,on/off,failure to wake up,performance profile or security profile).This use case proposes the introduction of a capability in the network,local or remote whose sole purpose is monitor or gather state information of sensors/endpoints on a network segment and transmits this information to a central location to enable manage,prevent failures or malfunctioning of critical IoT endpoint,as well as life cycle management of IoT endpoints.The typical node hardware structure is shown as Figure 1.7.

Figure 1.7 The typical node hardware structure

With billions of IoT endpoints and sensors going alive,it will get extremely complicated to monitor and manage the health of an end point sensor,especially if it is constrained,remote and critical to other systems relying on it for information,action.Today,most sensor malfunctions and failures are detected only after the fact in a reactive manner.This might not be acceptable going forward for critical infrastructure and IoT systems that have an impact on human lives.Use case actors and description is shown table 1.1.

Table 1.1 Use Case Actors and Description

Availability of constrained endpoints is a security requirement that is important to critical IoT systems and solutions.IoT endpoints vary widely in capability and they typically work in concert with wider network of IoT endpoints and Infrastructure that can be placed locally,remote and sometimes in inhospitable,not easily accessible remote placements.Status monitoring,life cycle management,upgrades and patch management status can be tedious and difficult.Some embedded systems in sensors are designed to last for up-to 10 years or more and keeping track of their upgrade,patch or battery life status or being aware if they have failed,become defective or are compromised is challenging.When managing thousands of endpoints in the field,if handled manually,can be cumbersome and in some cases impossible.IoT sensor monitoring systems can help alleviate this.Placing a security function in sensor networks whose sole purpose is to monitor sensors within its reach and transmit data continuously,intermittently or periodically subject to the requirement of the IoT solution can help significantly with endpoint lifecycle management.It can help prevent critical safety and security failures,which would otherwise be detected only after the fact and allows for pro-active remediation,repairs,replacement or decommissioning of endpoints.

The health of critical IoT systems will become progressively important,its availability,reliability and security becoming vital,as society progressively moves towards relying on them for automation and critical functions,be it in factories,smart cities,remote field sites or in medical devices.The most fragile component of the IoT system is generally the endpoint or sensor,that is often unprotected,low cost,but where the infrastructure relies on the consistency and availability of the end point and its unfailing performance.This introduces a need for monitoring systems towards the health of the IoT endpoint,sensor or device.Architecture of the endpoint monitoring systems is shown as Figure 1.8.

Figure 1.8 Architecture of the endpoint monitoring systems

A monitoring sensor or device placed amidst other functional sensors to gather and keep track of sensor status can function by either periodically detecting the presence of the set of sensor on the network,polling the sensors or tapping into the data/information being sent by the sensors on its designated communication path.The monitoring sensor could be configured with auto-detect capability or by an initially programmed set of data points that the monitoring sensor could update.In addition,maintenance records,in terms of,replacement,software version or latest patch upgrade status may also be recorded.In instances where a sensor fails to wake up,respond or ceases to transmit critical data,the monitoring sensor can generate alarms to central locations.In instances where average battery life is known and tracked the monitoring endpoint could proactively forecast when various sensors are due for a battery refresh thus allowing for proactive maintenance preventing segments of sensor networks from becoming unavailable.This type of sensor monitoring or surveillance would be a preventative security measure to ensure IoT system availability and reliability.