Control AC using Raspberry Pi, AWS IoT and LIRC...

Pointers

1. Read data from Temperature sensor (DHT11) and send it to AWS IoT through raspberry pi.
2. Store the data in DynamoDB in AWS and send notification to mobile device.
3. Send signal/message back from internet to raspberry pi to carry on a event.
4. Transmit the IR code through IR transmitter with Raspberry Pi 2 to device compatible like A.C.

Activity 1 and 2: Overview

A simple raspberry pi and AWS IoT application to send notification to user on temperature change. The temperature sensor continuously senses and sends data to Amazon’s DynamoDB (sensor readings data stored) over secure AWS IoT connection and when temperature goes below a certain threshold (a rule) a text message (SMS) will be sent to your mobile device using Amazon SNS notification service. Note: This experiment is specifically for DHT11 temperature and humidity sensor.

Logical design of project

Requirements

Hardware

• Raspberry pi 2 (with Raspbian Jessie OS) + breadboard
• DHT11 temperature and humidity sensor
• Wi-Fi dongle
• Jumper wires
• HDMI cable

Software installations

• Python 3 And here's some code!

sudo apt-get install python3	

Note: Pip is included by default with Python 3.4 libraries.

• Mosquitto MQTT client software (version 1.4 onwards preferred). There is really easy tutorial for installation I found here

wget http://repo.mosquitto.org/debian/mosquitto-repo.gpg.key
sudo apt-key add mosquitto-repo.gpg.key
cd /etc/apt/sources.list.d/
sudo wget http://repo.mosquitto.org/debian/mosquitto-jessie.list
sudo apt-get update
sudo apt-get install mosquitto-clients

• Paho Python client

sudo  pip install paho-mqtt	

• Adafruit Python DHT Sensor Library

1. Library for python to communicate with Raspberry pi gpio pins Tutorial for Adafruit’s guide and library installation can be found here

sudo apt-get update
sudo apt-get install python-dev
sudo apt-get install python-rpi.gpio

2. This is a python library for reading DHT series of temperature and humidity on raspberry pi. For referrence code is available here

git clone https://github.com/adafruit/Adafruit_Python_DHT.git
cd Adafruit_Python_DHT
sudo apt-get update
sudo apt-get install build-essential python-dev python-openssl
sudo python setup.py install

And,

• Amazon Web Services AWS IoT

Register and login through AWS console here. Create a free account (check free tier details for more information). Now you can access all free tier amazon web services. Create a AWS Identity and Access Management (IAM) corresponding account. Note: The ‘region’ selected for this experiment is ‘us-west-2’. Also install and configure AWS CLI, the instructions can be found here

• Connecting Raspberry pi with DHT11 sensor Set DHT11 temperature and humidity sensor on raspberry pi and bread board. Sample connection diagram:

Power up your raspberry pi, take free breadboard, dht11 sensor and some jumper wires. Use of 10K Resistors is preferred.

1. connect negative pin of sensor to pin 6 of pi

2. connect vcc pin of sensor to pin 2 of pi

3. connect signal pin of sensor to pin 7 of pi

Note: For this tutorial, no resistors are used.

Test: To check connection a sample program, Run

sudo python testrun.py

Note: testrun.py file is in test-run folder. The sample output for above file can be found in output-testrun file.

The output should show temperature in degree Celsius and humidity percentage readings every 2 seconds iteratively.

• Amazon Web Service’s AWS IoT connection Let us connect our Raspberry pi to AWS IoT service, for more information read here.

Traditionally we need, “thing” created on AWS IoT console, here a thing is our raspberry pi Generate certificates and connect them the “thing”. Three certificates are needed are root CA certificate, client’s private key and client certificate generated by AWS IoT (these certificates names generated should be similar but not exact to ca.pem.cert and private_key.pem and certificate.pem).

1. The AWS IoT service root CA is provided by Symantec and can be accessed from here, generates ca.pem.crt file

Thing creation and generation of certificates can be through AWS console online and through AWS CLI. Here this will be done using OpenSSL,

2. Creating a private key an RSA Key

openssl genrsa 2048 > private_key.pem

3. Creating client certificate

openssl req -new -key private_key.pem -out csr.pem

This command will need you to answer few questions here, this process is interactive. The fields to answer will be to add Country Name, State or Province, Locality or City, Company, Organizational Unit, Common Name, email, password, company name and few extra attributes. Remember the answers you provide here.

aws iot create-certificate-from-csr --certificate-signing-request file://csr.pem

The client certificate generated will be certificate.pem file.

aws iot describe-certificate --certificate-id <your_certificate_id>

Now, you have ca.pem.crt, private_key.pem, certificate.pem and csr.pem certificates for IoT devices to establish a secure connection to AWS IoT. We need to set these certificates as ACTIVE explicitly.

aws iot update-certificate --new-status ACTIVE --certificate-id <your_certificate_id>

• Create a 'thing'

aws iot create-thing --thing-name temperature-1

here name given to our aws iot thing is 'temperature-1', we can see all things created in AWS IoT console online. A ‘thing’ is a client device connected to AWS IoT and all these devices can be found in AWS IoT a thing registry. Create a thing for experiment, name it anything you like, here (the thing) named ‘temperature-1’. Execute this command and you will see a ‘thingArn’ and ‘thingName’ on screen, this means that a thing is created and you can check it on AWS IoT console.

1. Create AWS IoT policy and attach it to client certificate (generated earlier). The policy document is in json format, allowiotoperations.json

{
	"Version":"2012-10-17",
	"Statement":[
	{
	  "Effect":"Allow",
	  "Action":[
	      "iot:*"
	   ],
	   "Resource":[
		"*"
	   ]
	}
    ]
}

 aws iot create-policy --policy-name "AllowIotOperationsPolicy" --policy-document file://allowiotoperations.json

On screen policyName, policyArn, policyDocument and policyVersionId will be displayed.

 aws iot attach-principal-policy --principal "<certificate ARN>" --policy-name "AllowIotOperationsPolicy"

To attach your certificate with its policy to your thing

 aws iot attach-thing-principal --thing-name "temperature-1" --principal "<your_certificate_ARN>"

A unique AWS account custom endpoint is used for establishing a secure connection to AWS IoT, get your endpoint information using this command,

 aws iot describe-endpoint

aws iot account specific endpoint address will be displayed on screen, something like this e.g. “abc123d4efghi.iot.us-west-2.amazonaws.com”

To test your devices connection to AWS IoT

 mosquitto_pub --cafile ca.pem.crt --cert certificate.pem --key private_key.pem -h abc123d4efghi.iot.us-west-2.amazonaws.com -p 8883 -q 1 -d -t topic1 -i temperature-1 -m "Helloooo"

On successful connection output like this can be seen,

Client temperature-1 sending CONNECT
Client temperature-1 received CONNACK
Client temperature-1 sending PUBLISH (d0, q1, r0, m1, 'topic1', ... (8 bytes))
Client temperature-1 received PUBACK (Mid: 1)
Client temperature-1 sending DISCONNECT 

• Create a Dynamodb table and keys

Using AWS CLI,

aws dynamodb create-table --table-name IOT-TempData --attribute-definitions AttributeName=key, AttributeType=S AttributeName=timestamp,AttributeType=S --key-schema AttributeName=key, KeyType=HASH AttributeName=timestamp,KeyType=RANGE --provisioned-throughput ReadCapacityUnits=1, WriteCapacityUnits=1

On success, the TableDescription will be displayed and make a note of TableArn generated. Now, we need previously created AWS IAM account, AWS IoT will assume your IAM role to process and place messages in DB.

1. Create actions role Make file iot-tempdata-actions-role.txt and add, Using AWS CLI,

{
	"Version":"2012-10-17",
	"Statement":[
	{
		"Sid":"",
		"Effect":"Allow",
		"Principal":{
			"Service":"iot.amazonaws.com"
		 },
		 "Action":"sts:AssumeRole"
	}
    ]
}

 aws iam create-role --role-name "iot-tempdata-actions-role" --assume-role-policy-document file://iot-tempdata-actions-role.txt

On success, you will see Role details, make a note of Arn. Make a policy to attach to this role,

In file iot-tempdata-ddb-insert-policy.txt add,

 {
	"Version":"2012-10-17",
	"Statement":[
		{
		   "Effect":"Allow",
		   "Action":[
			"dynamodb:*"
		    ],
		    "Resource":[
			"arn:aws:dynamodb:us-west-2:123456789101:table/IOT-TempData"
		      ]
		}
	]
}

Note: The arn mentioned in Resource above is the TableArn generated while creating a dynamodb table.

 aws iam create-policy --policy-name "iot-tempdata-ddb-insert-policy" --policy-document file://iot-tempdata-ddb-insert-policy.txt

 aws iam attach-role-policy --role-name "iot-tempdata-actions-role" --policy-arn "arn:aws:iam::123456789101:policy/iot-tempdata-ddb-insert-policy"

2. Create a Topic Rule Amazon web services allow to describe specific actions to be performed based on AWS IoT topic strings received from client devices. These messages are parsed through MQTT topic strings,

Example: humidity topic string may look like this, topic/tempdata/humidity

Create a Topic rule to add new item to dynamodb on receiving a sensor read Make iot-tempdata-topic-ddb-insert-rule.txt file and add,

 {
  "sql":"SELECT * FROM 'topic/tempdata/temperature'",
  "ruleDisabled":false,
  "actions":[
	{
	  "dynamoDB":{
		"tableName":"IOT-TempData",
		"hashKeyField":"key",
		"hashKeyValue":"${clientId()}",
		"rangeKeyField":"timestamp",
		"rangeKeyValue":"${timestamp()}",
		"roleArn":"arn:aws:iam:: 123456789101:role/iot-tempdata-actions-role"
	    }
	}
     ]
}

 aws iot create-topic-rule --rule-name "iot_tempdata_topic_ddb_insert_rule" --topic-rule-payload  file://iot-tempdata-topic-ddb-insert-rule.txt

• Amazon Web Service CloudWatch logs

CloudWatch logs displays performance measure of all operations completed. To get the data passing through IoT to cloudwatch we create a new policy and attach it to rule. Make iot-logging-role.txt file and add,

 {
  "Version":"2012-10-17",
  "Statement":[
	{
	  "Sid":"",
	  "Effect":"Allow",
	   "Principal":{
		"Service":"iot.amazonaws.com"
	    },
	    "Action":"sts:AssumeRole"
	  }
   ]
}

aws iam create-role --role-name "iot-logging-role" --assume-role-policy-document file://iot-logging-role.txt

Adding logging policy to this rule, Make file iot-logging-policy.txt and add,

{
	"Version":"2012-10-17",
		"Statement":[
		   {
	"Effect":"Allow",
	"Action":[
	     "logs:CreateLogGroup",
	     "logs:CreateLogStream",
	     "logs:PutLogEvents",
	     "logs:PutMetricFilter",
	     "logs:PutRetentionPolicy",
	     "logs:GetLogEvents",
	     "logs:DeleteLogStream"
	],
	"Resource":[
		"*"
	]
     }
  ]
}

 aws iam create-policy --policy-name "iot-logging-policy" --policy-document file://iot-logging-policy.txt

 aws iam attach-role-policy --role-name "iot-logging-role" --policy-arn "arn:aws:iam::123456789012:policy/iot-logging-policy"

This logging option can be switched, on and off, for logging option on set ‘ logLevel=”INFO” ’. For logging option off set ‘ logLevel=”DISABLED” ‘.

Test whether sensor readings are getting stored in DynamoDB run your python application. An advantage of AWS IoT is also that we can send data to be stored on two applications simultaneously here I have tried to store sensor data on DynamoDB and to custom-endpoint at real time. You can describe your custom endpoint (optional).

Run

 sudo python egrun.py

On success, the data will be added in table (IOT-TempData) of database and can be viewed on DynamoDB console. Amazon CloudWatch logs will store the performance index of experiment.

• To get a SMS notification on temperature reduce

Here we will create a rule, that will send sms to your mobile device when the temperature of room goes below 24 degrees Celsius.

Create SNS topic

aws sns create-topic --name iot-tempdata-inside-temperature

Create policy to publish this SNS topic and attach to iot-tempdata-actions-role created earlier. Make file iot-hivedata-publish-sns-policy.txt and add,

{
	"Version": "2012-10-17",
	"Statement": [ { "Effect": "Allow",
		       "Action": [ "sns:Publish" ],
		       "Resource": [ "arn:aws:sns:us-west-2:742485087494:iot-tempdata-temperature" ]
		}
	]
}

aws iam create-policy --policy-name "iot-tempdata-publish-sns-policy" --policy-document file://iot-hivedata-publish-sns-policy.txt

aws iam attach-role-policy --role-name "iot-tempdata-actions-role" --policy-arn "arn:aws:iam::123456789012:policy/iot-hivedata-publish-sns-policy"

Create a rule that publishes to this topic when temperature goes below 24 degrees Celsius Make file iot-tempdata-topic-sns-templessthan24-rule.txt and add,

{
   "sql":"SELECT * FROM 'topic/tempdatatemperature' WHERE Temperature < 24",
   "ruleDisabled":false,
   "actions":[ {
   	"sns":{
   		"targetArn":"arn:aws:sns:us-west-2:123456789101:iot-tempdata-temperature",
   		"roleArn":"arn:aws:iam:: 123456789101:role/iot-tempdata-actions-role"
   	 }
   	}
   ]
}

 aws iot create-topic-rule --rule-name "iot_tempdata_topic_sns_templessthan24_rule" --topic-rule-payload file://iot-tempdata-topic-sns-templessthan24-rule.txt

Now we have all generated and attached policies and rules. To publish our SNS topic to endpoint SMS enabled mobile device For detailed steps, check here

Above we have created a topic and have attached policies and rule to it, these can be viewed in AWS SNS console online, on console click on topic created, from Actions tab select 'Subscribe to Topic' add SMS service and your mobile number in this format 91-000-000-0000 and add service.

There is also a way to add all sns actions through AWS SNS console directly (i.e. without commandline), the steps are as follows,

Go to AWS SNS console, you will notice iot-tempdata-temperature topic (created earlier) on console

Select this topic and from the Actions menu, choose Subscribe to topic

From Protocol choose SMS option and add your mobile number in this format 91-000-000-0000

Note: Here 91 is country code for India.

Click on ‘create subscription’

Go to AWS IoT console and select Create a rule, enter a name for your rule and simple description.

In Attribute field enter (*), which specifies to send entire MQTT message that triggered the rule.

In Topic filter, type iotbutton/your-button-DSN. If you are not using an AWS IoT button, type my/topic.

In condition field enter temperature < 24

From the Choose an action drop-down select Send message as a push notification (SNS)

From the SNS target select the Amazon SNS topic you created earlier

Choose the Create a new role link new webpage in IAM console will open, here accept the default values and select Allow,

Select Add action and then create.

Run

sudo python egrun.py

On success, whenever the temperature received is less than 24 degree Celsius you will receive a SMS notification on described mobile number. While simultaneously all sensor reads are being stored in DynamoDB as usual.

Note: In egrun.py file the temperature, humidity and connected raspberry pi’s cpu serial number will be added to database and be sent in sms.

Mobile notification,

Experiment complete 👍

Activity 3

Capturing/recording signal from IR reciever to raspberry pi.

Usecase: To read and emulate remotes IR (infrared) signals to control HVAC (Heating, Ventilation and Air Conditioning) devices or any other electronic devices like TV's and etc.

For this, LIRC-Linux Infrared Remote Control packages will be used, allows to encode-decode, recieve, store and send infrared signals of many commonly used remote controls.

Step 1: Install Lirc on raspberry pi, I found easy way to do so here

sudo apt-get update
sudo apt-get upgrade
sudo rpi-update
sudo reboot

We are trying to capture and emulate remote signals to control Panasonic TV device,

sudo apt-get install lirc
sudo apt-get install liblircclient-dev
sudo apt-get install lirc-x

liblircclient-dev is a infra-red remote control support - client library development files which is necessary for running 'irw' command.

lirc-x, package provides X utilities for LIRC, irxevent: allows controlling X applications with a remote control;

Step 2: Use IR reciever to record remotes ir signals, the recorded signals can be seen in lircd.conf file generated. There few steps to follow, which I found here, follow from step 8. all circuits used here are followed through above tutorial,

update /etc/modules file

lirc_dev
lirc_rpi gpio_in_pin=23 gpio_out_pin=22

Update /etc/lirc/hardware.conf file

LIRCD_ARGS="--uinput"
DRIVER="default"
DEVICE="/dev/lirc0"
MODULES="lirc_rpi"

Reboot, so rpi can fetech latest updates.

update sudo nano /boot/config.txt

dtoverlay=lirc-rpi,gpio_in_pin=23,gpio_out_pin=22,gpio_in_pull=up

Now, for recording IR signals, you need to first stop the LIRC daemon.

sudo /etc/init.d/lirc stop

Or

sudo systemctl stop lirc
sudo systemctl status lirc

Make IR receiver (TSOP1738) circuit for capturing TV, AC or anyother electronic device remote

here, connect ground pin of receiver to (ground) pin 6 of raspberry pi, connect vcc pin of receiver to (power) pin 2 of raspberry pi, connect signal/output pin of receiver to gpio pin 23 (gpio_in_pin) of raspberry pi.

To analyse the IR signal timings.

mode2 -d /dev/lirc0

run above command, point your remote to IR receiver and press buttons, you will see pulse/space values for these buttons.

Now, check list of allowed button names for lirc remotes,

irrecord --list-namespace

To record IR signals for each remote button,

Sudo irrecord -–driver=devinput --driver=/dev/lirc0 ~/lircd.conf

using this command we can give names to buttons e.g. KEY_VOLUMEUP and KEY_VOLUMEDOWN, the generated button names and their associated hexcodes can be viewed in lircd.conf file. Copy this file to /etc/lirc folder.

sudo cp lircd.conf /etc/lirc/lircd.conf

Let's check whether the codes for buttons are recorded correctly, start the LIRC daemon

sudo systemctl start lirc
sudo systemctl status lirc

To get the button's assigned name whenever you press that remote button

irw

now create a file lircrc and store it at /home/pi/.lircrc and /etc/lirc/lirc/lircrc

begin
button=KEY_VOLUMEUP
prog=irexec
repeat=0
config=/etc/lirc/lircd.conf
end

begin
button=KEY_VOLUMEDOWN
prog=irexec
repeat=0
config=/etc/lirc/lircd.conf
end

To check whether the assign names are set in lircd.conf file

irsend LIST panasonicTV “”

the output shows hexcode and keyname assigned fron lircd.conf file.

Once the signals are recorded we use them to control electronic device from our raspberry pi.

Activity 4

Transmit the IR code through IR transmitter with Raspberry Pi 2 to device compatible like TV. To send IR signal for one of the recorded buttons using LIRC, run irsend command, point your transmitter towards TV

irsend SEND_ONCE -d /var/run/lirc/lircd panasonicTV KEY_VOLUMEUP
irsend SEND_ONCE -d /var/run/lirc/lircd panasonicTV KEY_VOLUMEDOWN