The architecture of IOT

Okay, next some topic we are going

to look at: the architecture of IOT.

This is architecture from the viewpoint of protocol stack. As we will see this basically the interconnection of devices. For devices to be communicate, connected together, we need some rules. This is called “protocol.” Since the protocol has to cover lots of things to do, so we have to partition the protocols into different tiers. This is what to mean a protocol stack. Okay, from this viewpoint, the IOT architecture. It represents is equally, the protocol stack of IOT. They will be 3 different layers. From the bottom to the up, we will see the first layer is a circle, sensing layer or the perception layer.

This layer will be in charge of data gathering, and exchanging data between devices and also collect data send out, upward, and through the internet through the backend for processing. Okay, the second layer, middle layer, is so called networking layer. This layer will be in charge of data processing, formatting, and propagation. To the left you will see we have braces to connect these two layers together. That means, in general, in the IOT devices, networking layer and sensing layer will basically be combined together. That means, there will be devices with capability of sensing and also embedded with networking capabilities at the same time. The upmost layer will be the application layer. The application layer will be the design for the usage.

It will cover the data collected with suite from the sensing nodes, we call it data storage. Since, in most case, the data will be gathered in a very high speed. For example, we do the telemetry, we measure some geographic information, one information, or one data, per second. Okay, there are lots of data has to be propagated through networks back to the application layer for processing. So for IOT, first we need so called “big data.” Since we have lots of big data to process and we need lots of computing powers for processing. This is the second requirement, which is so-called cloud computing. So anytime we talk about IOT we always associated with the big data and cloud computing.

Okay, anyway, back to the application layer. They are also based on the data collected, we do the analysis. Then we do the presentation. So the IOT, in application layer, will be, not only provides the information, but also the intelligence. We have to learn something. We have to provide some kind of suggestions from the data we collected. Then this is purpose of designing the IOT. This is basically the architectural. This is simplest way to look at IOT architecture, sensing, networking, and application. Okay, so in what follows we will look at the sensing, networking, and application. Basically, in the design of these sessions, we combine the application into this session.

Since once we talk about the perception, we have to do the associate connections to the application. So these two will be combined together for introduction. And also we look at the networking layers, see what kind of main technology applies for data propagation. So, look at how the sensing layer gather data. Basically, it will be through various kinds of sensors. Here we categorize the sensors based on the operations. So, we have electrical-related, mechanical-related, and so forth. First we will look at these slides. Electrically related, we have voltage sensor, current sensor, power sensors.

Voltage sensor means we can, not only detect the voltages, the existence of voltages, and also we can set up a threshold for voltage so we can provide certain kind of signaling, such as alarms, to notify the voltage has been over a certain range. Current power will be the same purpose. So this is the electronic or electrical related applications. The second one is called the mechanical related. We have accelerometer, gyroscope, and distance sensors. Look at the figures below. The leftmost one is the three axis accelerometer. Accelerometer means we can measure the change of velocity in three different directions. Okay, so, this is the accelerometer modules.

Basically we make this module smaller into an integrated circuit and we put this modules into certain devices. Here is an example. This is a smart phone. In most smart phones, they all equipped with accelerometer. An accelerometer can measure, as you see the XYZ, three directions. That means if you have cellular phone, when you activate the accelerometer, you can measure the moving speed, the change in the speed, of the cellular phone. So we can measure the acceleration, y and z. That mean we can move, measure the movements in the front, and back, left to right and visa versa. And also up and down. So this is what we mean accelerator.

Gyroscope, or in short we call it gyro, is measured also the, basically measure the directions, orientations, changes in orientations and also in three directions. The next slide we have bits, comparisons for these two devices. The last one is called the distance sensor, this uses infrared or lasers. Then we emit the light and receive the light and then based on the delay times so we can calculate the distance since we know the light speed, we know the infrared or the laser will be propagated in lightspeed you know the times from the light you emitted when you detect the light back. Then you know what time elapsed. Then you can calculate the distance. This is the operation principle of the distance sensors.

As you can see more, this is like as we say it, you can look at the difference between the three axis, three directions, accelerometer and the gyroscope. You can take a look. So first, basically the accelerometer is measure change in velocity in three different directions. You know the change in velocity, that means the acceleration then you can base on the simple physical formulas. You can calculate the moving distance and so forth. Gyroscope measure the orientation. The directions means the angles and also change the rotational velocity. This is the basic principle of the gyro. So this is the difference between these two devices. Again, you look at the bottom, we have the URL.

If you would like to know, to see, more detail on the differences between the accelerometer, gyroscopes, you can look at the web pages. Okay, next we look at magnet-related (sensors). The most popular one is magnetometer. Magnetometer is basically measure the magnetic fields. Chemically related (sensors) can measure some dust or some environmental informations. For example, we have two examples here. One is dust sensor can measure cigarettes. Another one, we can measure the C02 or Carbon Dioxide exists in environments. If we like to know the environmental, the air pollutions. This kind of sensors will be useful. This is what to mean sensors. Basically look sensor, once operating, is sensing the data have transduced to sensing the signal. Sensing the physical changes.

Mechanical changes. It has to convert signals to data then transmitted out. So if we look at these sensors. They are not only the sensing devices, the components. They also include some microchips, processing units to combine those together. Otherwise, there is no data that can be transmitted out. Okay.