AWS IoT SDK with QCS610 SoC

The main goal of this project is to integrate the C++/Python AWS IoT SDK source into the QCS610 SoC target platform and build an application to post the status of camera recordings to the AWS IoT core console.

This project is designed to integrate the AWS IoT SDK into the QCS610 platform and create a small application to notify the current status of the camera to the AWS IoT core console. The source tree of the project starts with the application directory (/home/user/aws-iot)

aws-iot/
- video_record.c
- video_record.h
- basic_pub_sub/
- cJSON.c , cJSON.h , main.cpp, CMakeLists.txt
- publish.py
- certificates/
- xxxx-certificate.pem.crt, xxxx-private.pem.key, root-CA.crt
- publish
- videoaws

1. Setting up the Application SDK on the host system.
2. Camera environment configuration setup on the target device.
3. Setting up aws-iot device on aws console.
4. Compiled the aws-iot-c++ SDK on yocto environment.

1. Setting up the Application SDK on the host system.

• To install the application SDK, download from the given url: https://thundercomm.s3.ap-northeast-1.amazonaws.com/shop/doc/1593776185472315/Turbox-C610_Application-SDK_v1.0.tar.gz
• Unpack the SDK using below command. $ tar -xzvf Turbox-C610_Application-SDK_v1.0.tar.gz
• Execute the below script file, it will ask the default target directory, press enter and input ‘Y’ $ ./oecore-x86_64-armv7ahf-neon-toolchain-nodistro.0.sh

Now the environment setup is complete.

Further reference refers the below document https://www.thundercomm.com/app_en/product/1593776185472315

2. Camera Environment configuration setup on the target device

To setup the camera environment configuration follow the below document ‘Turbox-C610_Open_Kit_Software_User_Manual_LE1.0_v2.0.pdf’ From url below
https://www.thundercomm.com/app_en/product/1593776185472315 and Refer section 2.10.1

3. Setting up AWS IoT connect

   Please refer to the link below to connect with AWS IoT services. Inside the IoT quick start document in step 2, while creating the think object, select device as python instead of node-js.

   https://docs.aws.amazon.com/iot/latest/developerguide/iot-quick-start.html

   Save the credentials to the certificates folder in the application directory.

4. Compiling the aws-iot-c++ sdk on yocto environment

Step-1 : clone the meta-aws repository to your host system

$ git clone https://github.com/aws/meta-aws.git

Step-2 : copy the aws-iot-device-sdk-cpp-v2 to yocto project environment

$ cp meta-aws-master/recipes-sdk/ <working directory>/poky/meta-aws /recipes-sdk/ $ cd <working directory>

Step-3 : Initialize the build environment

$ export SHELL=/bin/bash
$ source poky/qti-conf/set_bb_env.sh

Select “qcs610-odk meta-qti-bsp” from the available machine’s menu. On selecting the machine, the ncurses menu starts. Select “qti-distro-fullstack-perf meta-qti-bsp” from Available Distributions menu.

The shell prompt will now move to the following build directory

<working directory>/build-qti-distro-fullstack-perf $

Step-4 : Add the aws-meta path in bblayers.conf file in ‘<working directory>/build-qti-distro-fullstack-perf/conf/bblayers.conf ’ folder.

Step-5 : run the bitbake command

$ bitbake aws-iot-device-sdk-cpp-v2

After successfully running this command, you can find the required compiled library as well as include folders generated in the path “<build_directory>/tmp-gibc/sysroots-components/armv7ahf-neon/aws-iot-device-sdk-cpp-v2/usr/”, copy the lib/ folders to the application directory.

$ cp <build_directory>/tmp-gibc/sysroots-components/armv7ahf-neon
/aws-iot-device-sdk-cpp-v2/usr/lib /home/user/iot_aws/.

Step-6 : To compile the application, use the bitbake command below

$ bitbake -c compile -f aws-iot-device-sdk-cpp-v2

Step-7 : After successfully running the bitbake command you can find the git version of code available in “<build_directory>/tmp-gibc/work/armv7ahf-neon- oe-linux-gnueabi/aws-iot-device-sdk-cpp-v2/1.6.0-r0/” folder. Replace the current basic_pub_sub folder in path “git/samples/mqtt/basic_pub_sub” with ‘basic_pub_sub’ folder given in our repository. Execute the bitbake command given in step-6. After the successful execution, the target binary will be generated in the “build/samples/mqtt/basic_pub_sub/basic_pub_sub” folder

Step-8 : Rename the binary ‘basic_pub_sub’ to ‘publish’ and copy it to the application directory. Also copy the libcJSON.so file to lib/ folder in the application directory.

$ cp <build_directory>/tmp-gibc/work/armv7ahf-neon-oe-linux-gnueabi/
aws-iot-device-sdk-cpp-v2/1.6.0-r0/build/samples/mqtt/basic_pub_sub/publish
/home/user/iot_aws/. $ cp <build_directory>/tmp-gibc/work/armv7ahf-neon-oe-linux-gnueabi/
aws-iot-device-sdk-cpp-v2/1.6.0-r0/build/samples/mqtt/basic_pub_sub/libcJSON.so
/home/user/iot_aws/lib/.

Example: for playing 4k video

Whenever the camera application starts and stops capturing, the application is designed to establish connection to the aws-iot-core and publish the current status of the camera details in json format to the particular topic. The details include recording format, device details and current action of the camera. The subscriber of this topic can receive the camera details.

To Download video:

Using the adb pull command we can download the video to the host system.

To see the live streaming, open the new terminal in the host system and enter below command. Also make sure you have installed the ‘vlc player’ on the host system in order to view video playback.