IOT system framework learning

 Introduction to IOT architecture 
 
 
 
 Please add groups for problem consultation and Project Exchange ：
 
 Group name ：IT Project communication group 
 
 Group number ：245022761

Internet of things system framework learning

Here are some key issues ：

• How the device connects to the network ？
• How to communicate between devices ？
• The purpose of Internet of things data ?
• How to build an Internet of things system framework ？ What about its technical architecture ？
• Internet of things terminal software system architecture ？
• Internet of things cloud platform system architecture ？

1. How Internet of things devices access the network ？

Only devices are connected to the network , To be an Internet of things device . This involves 2 A key point ： Access mode and network communication mode .

Device access mode ：

The current is 2 Access mode

1. Direct access ： Internet of things terminal equipment itself has networking capability and directly accesses the network , such as Add... At the device end NB-IOT Communication module ,2G Communication module .
2. Gateway access ： Internet of things terminal equipment itself does not have the ability to access the network , After the local networking , It needs to be connected to the network through the gateway . For example, the terminal equipment passes through zigbee Wireless networking , Then the data of each equipment passes through Zigbee The gateway is uniformly connected to the network . Commonly used local wireless networking technologies include Zigbee,Lora,BLE MESH, sub-1GHZ etc. .

In Internet of things devices , Internet of things gateway is a very important role . An intelligent device between local LAN and external access network . The main function is network isolation , Agreement conversion / Adaptation and data transmission inside and outside the network .

A typical Internet of things gateway architecture is as follows ： 

 

Network communication mode

Common communication networks mainly exist 2 Ways of planting ：

1. The mobile network ( For main outdoor equipment ) 
The mobile network 2G/3G/4G/5G/NB-IOT etc.  
2. Broadband ( Mainly used for indoor equipment ) 
WIFI,Ethernet etc.

  How to carry out after the Internet of things equipment is connected to the network M2M,M2C signal communication ？

After the IOT device terminal is connected to the network , Just the beginning of the Internet of things application . After the device is connected to the network , Devices need to communicate with each other , Devices and the cloud need to communicate with each other . Only interworking , The value of the Internet of things is shown . Since we want to communicate , You need a set of Internet of things communication protocol . Only devices that follow the protocol can communicate with each other , Can exchange data .

What are the commonly used Internet of things communication protocols ？ There are mainly the following agreements ：MQTT,COAP etc. , They have one thing in common, which is based on the message model . Between equipment , The communication between the device and the cloud is realized by exchanging messages , The message carries communication data .

 

Between IOT devices , After the device and cloud platform can exchange data , What are we going to do next ？ 

21 century , What's the most important ？ data . With a lot of Internet of things device data , For data , Human beings can dig the rules inside , Tap the commercial value inside , Predict the future state of the equipment, etc .

For IOT data applications , There are several levels  

1. Basic applications ： monitor  

After collecting equipment data through the Internet of things , If the equipment data status exceeds the preset status, it will automatically alarm for the first time , The administrator handles the problem at the first time , It can be operated remotely , give a command . Solve the problem in the bud .

 

2. Advanced applications ： Report statistics

By statistical methods , Make statistical analysis on the historical operation data of the equipment . Different reports can be analyzed according to different dimensions . Then it is displayed in front of the administrator in the form of chart or large screen . Administrators can quickly and intuitively understand the operation status of the whole Internet of things devices .

3. Advanced applications ： data mining / machine learning

This part needs to dig out valuable things from the data . For example, through continuous tracking and analysis of equipment data for a period of time and combined with human past equipment operation and maintenance experience , Predict the probability of equipment failure through machine learning , And possible causes after failure , And give the maintenance plan . The example just given , It is the tip of the iceberg in advanced applications of the Internet of things . By introducing now hot AI technology . The Internet of things can become an intelligent Internet of things . Maybe in the near future , People and devices can talk freely , Devices can also talk to each other and make optimal decisions automatically .

 

To sum up , The value of the Internet of things is to improve productivity , The efficiency of management , Greatly promote the improvement of social productivity .

How to build an Internet of things system framework ？ What about its technical architecture ？

Software system architecture of Internet of things terminal equipment

---

The common system frameworks are summarized as follows 2 Kind of ： belt RTOS Of （ Dealing with complex business scenarios , In the scenario, multiple transactions are required to complete the work in parallel ） And without RTOS Of （ Usually, the business scenario is relatively single ）

No RTOS Equipment terminal system framework ： 

belt RTOS Equipment terminal system framework ： 

RTOS What is it? ？ Real time multitasking operating system , With it , Multiple tasks can be run in parallel in the terminal device . Each task is responsible for one transaction . Run through parallelization , Real time response and efficiency will be improved .RTOS The real-time operation kernel generally contains the following important components ：

1.   Task scheduling

2.   Synchronization and communication between tasks

3.   Memory allocation

4.   Interrupt management

5.   time management

6.   Device drivers

Take task scheduling component as an example ： 
In an embedded operating system , The task is CPU Upper minimum operating unit . Usually a little more complicated IOT APP It is completed by multiple tasks . For example, some tasks are responsible for handling user event input and UI Show , Some tasks deal with data communication , Some tasks are responsible for business logic processing .

Since there are multiple tasks running in a system , and CPU Resources are indeed single , This leads to only one task at a time CPU Run up . So for every task to be able to CPU There is a chance to run , It involves the concept of task scheduling . Task scheduling needs to follow certain rules , What rules do you usually follow ？ Our common scheduling methods are 3 Kind of ： A priority based scheduling , One is based on time slice scheduling , One is scheduling by combining priority and time slice .

Take priority scheduling as an example , When defining tasks , Assign a priority to each task , During operation , High priority tasks will be run first . Until there are no high priority tasks , Low priority tasks will be run . If a low priority task gets CPU After resources , This is what if there are high priority tasks ready ？ Two ways to deal with it ：1. Continue operation 2. Preemptive . High priority preemption CPU The resource is running .

Other components are not listed one by one .

Internet of things cloud platform system architecture

---

The system architecture of Internet of things cloud platform mainly includes four components ：

• Device access
• Equipment management
• Rules engine
• Security authentication and authority management

Device access

What does the device access mainly do ？ 
1. Contains a variety of device access protocols , The mainstream is MQTT agreement . Some cloud computing vendors are also MQTT On the protocol, the simplified protocol becomes a unique access protocol . 
2. Concurrent connection management , Maintain long connection management for potentially billions of devices .

If you deal with billions of device connection management ？ It's open now MQTT Proxy servers are mostly stand-alone , At most, it means connecting more than 100000 devices concurrently . So if you want to manage billions of connections , Load balancing is required , Use distributed architecture . The cloud platform requires the deployment of distributed services MQTT proxy server .

Equipment management

Devices are generally managed in a tree structure , Including device creation management, device status management, etc . The root node starts with the product , Then there is the device group , Then to the specific equipment . 
It mainly includes the following management ： 
Product registration and management  
Add, delete, modify and check the equipment under the product  
Device messaging  
OTA Equipment upgrade management, etc

Rules engine

Internet of things cloud platforms are usually built based on existing cloud computing platforms . A mature business of the Internet of things not only uses the functions provided by the cloud platform of the Internet of things , Generally, the functions provided by the cloud computing platform are also needed , For example, virtual machine , Cloud database, etc . Users can build on virtual machines web Industry application services .

The main function of the rule engine is to forward the data of the Internet of things platform to other cloud computing products through filtering . 
For example, you can forward the data reported by the device to table store Database products .

General usage of rule engine ： class SQL Language , By writing SQL Language , Users can filter data 、 Processing data , And send the data to other cloud computing products , Or other cloud computing services .

Security authentication and authority management

The Internet of things cloud platform issues a unique certificate for each device , You need a certificate to allow the device to access the cloud platform . The minimum authorization granularity of cloud platform is generally at the device level . What we call certificates are generally divided into 2 Kind of ： One is the product level certificate , One is equipment level certificate . The product level certificate has the maximum authority , It can operate all the equipment under the product . Equipment level certificate , You can only operate your own equipment , Cannot operate on other devices .

Therefore, each device connected to the cloud platform stores a certificate locally （ In fact, the form of existence is a KEY, There are multiple strings to form ）. Every time a connection is established with the cloud , Bring your certificate with you . So that the cloud security components can pass the verification .

Cloud computing products

Big data computing products , Cloud infrastructure services （ High performance servers , Cloud database , Cloud network ） etc.

Internet of things applications

Smart city , Smart agriculture , Industry 4.0 etc.

The next one is from Hua Weiyun Ocean Connect Internet of things platform architecture ( End 、 edge 、 tube 、 cloud 、 application )：