← Previous lessonFree lesson 2 of 3 · lesson 2 of 17Next lesson →

Unlock all 17 lessons of Medical Embedded SystemsFull course ₹40,000 · paid

Price & syllabus
Free lesson · 19 min · no payment

Embedded Software Architectures and Classification of Embedded Systems

A free lesson from Skill-Lync's Medical Embedded Systems course, on Grow by Skill-Lync · Last updated 16 Nov 2026 · Session: “Module 1 Session 2”

What will you learn in this free Medical Embedded Systems lesson?

The instructor compares software architectures: interrupt-controlled, cooperative multitasking, preemptive multitasking with an RTOS, microkernel and monolithic kernels. It then classifies embedded systems as small, medium or high scale, and as real-time, standalone, networked or mobile, naming medical examples such as CT, MRI, defibrillators and ventilators.

  • Interrupt service routines and cooperative multitasking
  • Preemptive multitasking and why RTOS is used
  • Small, medium and high scale systems
  • Hard versus soft real-time embedded systems

What comes after lesson 2?

The next free lesson is lesson 3, “Embedded System Architecture, Designer Skills and Applications” (20 min). Later in Skill-Lync's Medical Embedded Systems course (paid): lesson 4, Module 2 Session 1; lesson 5, Module 2 Session 1 Part 1; lesson 6, Module 2 Session 1 Part 2. The full course has 17 lessons (8 h 32 min of video); see the course overview or its price and syllabus.

Lesson transcript

Machine transcript of the lesson audio (lightly cleaned). Technical terms may contain errors; the video is the reference.

Now let's see the next embedded system architecture which is used for the software. Now the interrupt control software architecture. So some of the embedded systems are predominantly controlled by interrupts. This means that a task is going to be performed by some system that are going to be triggered by different kinds of events. Basically an interrupt is going to be generated. For example, it can be by means of a timer at some particular predefined interval or it can be an occurrence of an event in a serial port based on some data which is going to be generated.

received in the serial port. This type of an architecture is used. Event handlers need to be of low latency and the event handlers are also going to be very short and very simple. So it can be used in various scenarios. Basically these interrupts, these systems are going to run a simple task in the main loop and the task is going to be not very sensitive and there is going to be some type of a delays also can be there. So sometimes these interrupt handlers will add longer tasks to the queue structure as well.

So once the interrupt is going to be given, there is going to be some kind of an event which is going to be activated or some kind of an activities are going to be happening there and these activities take you to a separate routine that is called as an interrupt service routine and it is going to carry out some kind of a task or a simple task. Once the interrupt service is completed, once the interrupt service is completed, then it is going to come back to the main routine and the activities go on. So basically it works like a multitasking kernel with separate processes.

Now let's see on what is a cooperative multitasking in an embedded software architecture. Basically a cooperative multitasking is similar to a simple control loop. Here the loops are going to be hidden inside the APIs. The programmer is going to assign a set of tasks to it and the common operations like pause where there is going to be a temporary break or wait when there is going to be no operation to be carried out are going to be used here. Basically it operates similar to the control loop where it can be added to a new application or queues.

So the advantages and disadvantages are similar to that of the control loop except that adding new software is easier here and also by writing simple new task is very easy and getting added to the queue is going to be very easy here. Now let's see what a preemptive multitasking or a multi-threading in an embedded software architecture. Basically timer switches between the threads are tasked by means of an interrupt. So this system has the operating system kernel and here multiple tasks are going to run parallely. Basically when you are going to have preemptive multitasking or a multi-threading being carried out, you need to extensively test your software to see its functionalities and then only you implement them.

Basically real-time operating systems are usually used on these devices and when you are going to choose a real-time operating system, it has to be chosen very critically. And also you need to handle interfaces like the various devices like serial, Bluetooth, USB, TCP, IP and wireless, LAN, data and voices all need to be handled very carefully. Basically in the preemptive multitasking or multi-threading environment, these type of systems are low level piece of code is going to switch between task or threads based on some kind of a timer which also can be connected to an interrupt.

This is the level at which the system is generally considered to have an operating system kernel. Depending on how much the functionality is required, it is going to introduce more or less type of complexities of managing the multiple tasks which are going to run conceptually in parallel. You can know that any amount of code can potentially damage the data of another task or in a program and you need to carefully design them and test them. And you need to access to the shared data must be controlled by means of some methods of synchronization such as the messaging queue is semaphores or by means of non-blocking synchronization schemes.

Otherwise, it will lead into some catastrophic damages. Because of these kind of complexities, it is always common for the designers to use RTOS where the application programmers can concentrate on device functionalities rather than on the operating system services. At least for the larger systems, smaller systems, you can not afford for going for an very expensive real-time operating system. Also, when you are going to use the real-time operating system, you need to have a choice of the RTOS so that if any issues are going to come up, the selection has to be prioritized and the applications have to be critically tested and you need to specifically choose your applications based on the RTOS so that you will not lead into an expensive real-time operating system being running on an simple embedded system applications.

Now, let us move on to the next embedded software architecture that is the microkernel. Basically, the kernel manages the operations of the computer. Microkernel uses the services and the kernel services are implemented in different address space. Here, the user services are kept in the user address space and the kernel services are kept under the kernel address space. Now, what is an exos kernel? Exos kernel is a type of an operating system developed at the MIT. Basically, it seeks to provide application level management of the hardware resources.

Exos kernel architecture is designed to separate resource protection from management to facilitate application-specific customizations. Now, let us see what are monolithic kernels. Basically, these monolithic kernels are used in complex environments. Here, the operating system is similar to desktop such as in Microsoft, Windows or in Linux system. Here, more resources are going to be required and it may be expensive also. Basically, the operating systems like Linux and Windows CE use the embedded monolithic kernels. The monolithic kernels are popular because of their portability on the platform and the reuse is very easy.

And reuse for public codes for the device drivers, firewalls, web servers, etc. are very simple and they are reusable. And the codes are very easy to debug. They are highly reliable and simple to enable the development process. And they are more portable and they can be ported onto multiple devices. And also, they have faster response after integrating into the hardware. So, the response time is quite good in this monolithic kernels. Now, moving on to the next topic is the exotic custom operating systems. So, basically, they are integrated system architectures which require to work on a safe, fast and efficient manner to a greater extent.

The system can be divided into mechanism controller and desktop controller with a traditional operating system running on them. The communication scheme transmits two-way information in an exotic operating system. Now, what are the additional software components which are there in the embedded software architecture are? In addition to the core operating system, many embedded systems have additional upper layer software components. So, basically, these components can be network protocol stacks like such as TCPIP, CAN, HTTP, FTP or HTTPS flash memory and its management. Also, they have video and audio features.

And these features are all going to be depending on the what are the intended applications which we are going to use. If the embedded devices has an audio and video capabilities, then the appropriate drivers and codecs will be present in the system. In the case of the monolithic kernels, many of these software layers are included. Then, in the RTOS category, the availability of additional software components depends upon the commercial offering which we are going to consider into the embedded system. Now, let us move on to the next topic that is the embedded systems classification.

Now, let us see how the embedded systems are being classified. Now, considering upon the degree of design complexity and system performance, embedded systems are being classified into various levels. Now, let us cover the classification of embedded systems in detail. Now, the embedded systems are classified based on the performance or functional requirements or based on the complexity and performance of the microcontroller. Now, coming to the classification based on the performance or functional requirements, they are again classified as real-time systems. They are classified as standalone systems or network systems or as a mobile system.

Now, if you classify the embedded system based on the complexity and performance of the microcontroller, they can be classified as a small-scale embedded system, medium-scale embedded system or an high-scale embedded system. Based on the complexity or performance of the microcontroller, it can be classified as small-scale, medium-scale or high- or a sophisticated embedded system. Now, the small-scale embedded systems, what are they? How are the embedded systems classified as a small-scale embedded system? Why do you classify them as a small-scale embedded system? Small-scale embedded systems are basically designed using either an 8-bit or an 16-bit microcontrollers.

They can be consisting of a very simple hardware and the software. The cost is going to be very, very less. The processor is going to use very less or limited resources of memory and processing speed. And also, the processing power is less and the power consumption is going to be very less. And also, the cooling requirements are going to be very less. And also, in this small-scale embedded system, the programming languages is very commonly used as an assembly or it can be a high-level language with a cross-compailer built into them.

And also, the small-scale embedded systems are going to be very less and they are going to be very cheap. And you can see that they can be used in very simple devices like your car radios, ovens, dishwashers and washing machines. Now, the classification of embedded systems as a medium-scale embedded systems are basically these systems uses 8, 16 or 32-bit microcontrollers. Basically, these systems use 8, 16 or 32-bit microcontrollers. The architecture in them will be either a risk-based architecture or it can be a DSP-based processor will be there.

They commonly use the system-on-chip concepts. So there are a number of devices built into them. They basically are going to be used for highly computational intensive purposes. And also they can have some signal processing algorithms running on them. They will have some kind of a processing capability built into them. Also these systems will have real-time constraints and also they are very commonly used in multimedia, speech processing and voice recognition systems. Here the programming tools are slightly high-end programming tools are used like high-level languages like C, C++, Embedded C, Autos.

And they are going to be used on some integrated development environments are going to be used. And then these medium-scale embedded systems will use different types of software tools like compilers, debuggers, simulators, etc. for the development of the systems. When a designer is going to use a medium-scale embedded system, then these skill sets are also accordingly have to be highly skilled so that you will be able to design these kind of an embedded system. Now when you see the next classification, the next classification is the high or sophisticated scale embedded system.

Basically here we are going to use a very high-end microcontrollers. Basically they will be of either 16 or 32-bit microcontrollers. Here again the architecture can be based on RISC and DSP-based architectures can be there. They are highly powerful system on chips. They operate on very high clock speeds. And here these highly sophisticated embedded systems will use power processor for computational purposes. And also they will have real-time signal processing applications built into them. And they are very hard real-time constrained systems. They are used for very critical applications.

They will be used for like aircraft navigation, missile guidance systems, radars, electronic warfare applications. They are very commonly used where the response time has to be in a very, very short time. So these are the high or sophisticated scale embedded system. Now again based on the functional performance, we have classified embedded systems as real-time embedded systems, standalone embedded systems, or networked embedded systems, and mobile embedded systems. Now let us see what are these embedded systems classified based on the functional or performance requirements. A real-time embedded system is strictly time-specific, which means that these embedded systems provide output in a particular or a defined time interval.

This type of embedded systems provide quick response in critical situations, which gives most priority to time-based task performance and in the generation of the output. That is why you can see that the real-time embedded systems are used in very commonly in the defense sector. They are going to be used in medical and healthcare sectors and other some critical industrial applications where the output has to be provided at the right time. So you can see that in this kind of a real-time embedded system, you can see that there are a number of sensors which are going to be interfaced and there are subsystems, systems connected.

There are going to be various IO interfaces. There are going to be a number of activators are also going to be present. So these systems are going to be generally used in physical and biological systems, and also they are going to be life-saving equipments as well as in case you are going to use them in medical systems. Typical examples of a real-time embedded systems are CT, MRI, defibrillator, ventilators, patient monitoring systems in the healthcare sectors. If you see in the commercial sectors, it can be used in the airplanes, or if you take the military applications, you can see the fighter aircraft, submarines, missiles, spacecrafts, or if you see the transportation systems, the railways, ships can use them.

So these type of systems are real-time embedded systems. Again, if you see the real-time embedded systems, they are classified as soft real embedded system or hard real embedded systems. So what is this soft real embedded, soft real-time embedded system and hard real-time embedded system? A soft real-time embedded system is one where the embedded system time or a deadline is not so strictly followed. If the deadline of the task is passed, meaning that the system didn't give any result in a predefined time interval, still the result or output can be accepted.

They are called as soft real-time embedded system. Now, in case of a hard real-time embedded system, in this kind of embedded systems, the time or deadline of any task is strictly followed, meaning that once a task is given, the task must be completed within a time frame or within a defined time interval. Otherwise, the result or an output cannot be accepted. For example, you can take that traffic light control system or if you take a military missiles or you can take an aircraft navigation system or you can take an healthcare systems where the abnormalities detected in a health parameter has to be defined within a critical time interval.

These are some of the hard real-time systems. So, you had studied what a hard and real-time embedded system is. Now, let's see what a standalone embedded systems are. These standalone embedded systems are independent, which can work by themselves. They don't depend on any of a host system. So, they don't have a host system. They are going to be working independently. So, basically, it takes inputs in the digital or in analog form and it can provide an output after processing them or converting the data and then it is going to display the data in the output devices.

Typically, some examples of this standalone embedded systems are like your MP3 players or microwave ovens, audio systems, your mobile phones, calculators, your TVs, smart TVs. They are all standalone embedded systems. The next classification is the networked embedded systems. The networked embedded systems are basically connected to a network which may be wired or wireless and that is going to provide an output to an attached device. Basically, these devices communicate with embedded web server through one network. Some of the examples of networked embedded systems are like your home security systems, ATM machines, your card swipe machines.

So, basically, you can see that the networked embedded systems are connected to a resource with the use of a network. Typically, you can take your ATM machines. You have a lot of ATM machines distributed all over and they are now connected via a network. So, the network can be either wired or wireless network and they are connected again to a server. So, this connectivity is very, very important because it has to be connected so that whatever the access happens at one of the nodes can be updated in the server and the users can benefit from that.

So, the connectivity can be by means of your wireless communication or your mobile connectivity or a LAN or a dedicated lease lines. So, these kind of networked embedded systems are growing up very fast and many of the devices and computers and systems are being networked. And here, the web server embedded controllers, all are integrated devices within the web servers and they are controlled through by means of the web browsers. So, a number of devices which you are practically seeing are the ATM machines which are commonly interconnected where you can have access globally for withdrawing your money from your bank accounts.

Various devices like your IoT devices, mobile networks, they are all typical networked embedded systems. Now, let's see the next classification that is the mobile embedded system. This mobile embedded systems are small and easy to use and they require very less resources and they are constantly on the move. They are highly preferred kind of an embedded systems because of its portability and they can be taken wherever you keep moving on. Typical examples of a mobile embedded systems are your mobile phones, laptops, your notepads and digital cameras.

Because the main limitations of these mobile embedded systems are here, the resources will be limited and there will be a number of memory restrictions and other problems will be there. Even power, battery and also the recharge mechanisms will be a problem for you. So, these are some of the limitations of the mobile embedded systems. So, these used machines will be more widely utilized when possible. I am sorry, it is not very focused and embedded systems are not exactly believed. I am sorry, I am sorry, it is easily left but I am sorry.

But I am sorry but I am sorry, but I may be more obviously light up to the cellphone.

Want the rest of Medical Embedded Systems?

Get the full course syllabus and fees.

The full course is paid (₹40,000). We'll send the syllabus and call you back within one working day.

Talk to a career counsellor

Free 20-minute call. We'll map the right course to where you want to work.

By submitting you agree to our Terms and Privacy Policy.

Full course · paid

Unlock the full Medical Embedded Systems course

The remaining 14 lessons, mentor-reviewed projects and a certificate come with the paid course (₹40,000). Leave your details: we'll send the syllabus and fees and call you back.

Get the syllabus and fees

A counsellor calls you back within one working day.

By submitting you agree to our Terms and Privacy Policy.

Grow by Skill-Lync · free lesson

Start watching

Free, no payment. Leave your details once and every free lesson on Grow plays straight away. A Skill-Lync counsellor may call you about the full Medical Embedded Systems course.

By submitting you agree to our Terms and Privacy Policy.

Talk to a career counsellor

Talk to a career counsellor

Free 20-minute call. We'll map the right course to where you want to work.

By submitting you agree to our Terms and Privacy Policy.