<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Jonathon Cronin on Medium]]></title>
        <description><![CDATA[Stories by Jonathon Cronin on Medium]]></description>
        <link>https://medium.com/@jon_83595?source=rss-6b13a6faed29------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*NPDhHxe2fJeeXoVfStXyXw.jpeg</url>
            <title>Stories by Jonathon Cronin on Medium</title>
            <link>https://medium.com/@jon_83595?source=rss-6b13a6faed29------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 06:33:26 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@jon_83595/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Wi-Fi Hacking with Wi-Fi Pineapple]]></title>
            <link>https://medium.com/@jon_83595/wi-fi-hacking-with-wi-fi-pineapple-647f0373b0df?source=rss-6b13a6faed29------2</link>
            <guid isPermaLink="false">https://medium.com/p/647f0373b0df</guid>
            <category><![CDATA[hacking-tools]]></category>
            <category><![CDATA[wifihacking]]></category>
            <category><![CDATA[hacking]]></category>
            <dc:creator><![CDATA[Jonathon Cronin]]></dc:creator>
            <pubDate>Mon, 26 Feb 2024 00:33:36 GMT</pubDate>
            <atom:updated>2024-02-26T00:35:31.789Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eL-r_RyP_DysmA8Yl9nBGg.png" /></figure><h3>Overview</h3><p>This article covers a demonstration of using the Wi-Fi Pineapple to perform an attack on a wireless network. This uses a Kali Linux setup to perform the password cracking on the PCAP files created. However, this is not required for this process. The Wi-Fi Pineapple provides a base for this operation and this article is designed to demos its role it is also not required. I believe its a very useful platform for the management, professionalism and documentation of engagements. This article is meant as an introductory process using tools to create a lower ceiling of entry to this form of penetration testing.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3gWZkEjjAwEHx1pHw3F7oA.png" /></figure><h3>Assumptions</h3><ol><li>You have a working Wi-Fi Pineapple with access to its console</li><li>You have a wireless network using WPA2 or similar level protection</li><li>You have a device like a cell phone or laptop you can access the hacking targets network.</li></ol><p>This guide is not intended for use against networks that you do not own or do not have permissions to perform the penetration testing on.</p><p>This is an example of the network and what part of the process we are trying to interrupt to perform the attack.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1mx-atI5ibehX7_oeyYqUg.png" /></figure><h3>Process</h3><ol><li>Connect your test device to the wireless network you are planning to attack.</li><li>Connect to you Wi-Fi Pineapple and perform a recon scan.</li><li>Perform a recon scan. A note here in the settings of the Wi-Fi pineapple the default is to not enable the 5Ghz signal range. If you have the extension device to use this you will want to check on those settings first and then perform the scanning.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9wFohB_zh8Bt3-ar4nN0iQ.png" /></figure><ol><li>Select the wireless network and click on the Capture Handshakes (this will be red in the current version of the Wi-Fi Pineapple)</li><li>Now you need to capture the handshake. You can simply disconnect and reconnect to the network. However, I recommend using the Deauthenticate All button in the Wi-Fi Pineapple.</li><li>Once the device reconnect and if successful you will get a capture of the 4 way EAOPL handshake as a PCAP file. You are looking for a full PCAP capture.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zrMzWiSqEi17cdbcV-8K5A.png" /></figure><p>4. Now you can download this into you Kali Linux or anywhere you can run aircrack-ng from.</p><p>5. You can now perform password attacks against the capture file. This will be covered in the next section.</p><h3>Password Attack on PCAP File</h3><ol><li>This attack is easy to perform with a wordlist. Again in a demostration lab I would recommend using a PSK (pre-shared key) that is in a wordlist you are intending to use.</li><li>Use the following command on the PCAP file in order to use a word list file to crack the password.</li></ol><pre>aircrack-ng -w /use/share/wordlist/&lt;wordlist&gt; &lt;your PCAP capture&gt;.cap</pre><p>If this is successful you will see a screen showing the password recovered from the PCAP file.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/865/1*tdNheZSWqm5k7-BtHPtE3A.png" /></figure><p>If you would like to run this same process but using aircrack-ng without using the Wi-Fi pineapple see this article for a great breakdown.</p><p><a href="https://medium.com/students-in-infosec/how-to-find-the-password-of-a-wifi-network-aircrack-ng-tutorial-4a8a5dd15c5d">How to Find the Password of a WiFi Network | AirCrack-ng Tutorial</a></p><h3>Conclusion</h3><p>This article covers the utilization of the Wi-Fi Pineapple in the process of performing a password attack against a network. This is only a basic introduction to the most simple demonstration. If there is enough interest I will cover the more realistic attacks using Hashcat to work against more realistic passwords. The process for performing an Evil Twin attack and attacks using the MiTM methods to insert and reroute all traffic for users to your pineapple.</p><p>Overall the device provides a simplified base for running wireless network attacks and antenna and scanning all in a single package. There is also the C2 system which can allow you to setup and send a wifi pineapple to a customer and perform all operations while remote from the target for penetration testing ahead of any onsite work. We can cover these in more detail in later articles. Overall its a great product for the price point.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YQi1r9VNtQDuOEhYIegWig.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=647f0373b0df" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Historians as IT/OT Convergence]]></title>
            <link>https://medium.com/@jon_83595/historians-as-it-ot-convergence-3f48d16a1c46?source=rss-6b13a6faed29------2</link>
            <guid isPermaLink="false">https://medium.com/p/3f48d16a1c46</guid>
            <category><![CDATA[it]]></category>
            <category><![CDATA[it-ot-convergence]]></category>
            <category><![CDATA[sto]]></category>
            <category><![CDATA[industry-4-0]]></category>
            <dc:creator><![CDATA[Jonathon Cronin]]></dc:creator>
            <pubDate>Wed, 04 Oct 2023 13:26:29 GMT</pubDate>
            <atom:updated>2023-10-04T13:26:29.899Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-LibCh2Cux6rkKuuv3A_yg.jpeg" /></figure><h3>Summary</h3><p>When we are dealing with the issue of getting data out of the operational technology space into the corporate space of a industrial system we are faced with many issues. Security being paramount but also the issues of understanding the nature of the data and its format, the way OT works including scan times and speed of data flow. There are many solutions to this issue but for the IT department the task of code switching (changing the way we think) can cause serious pain in making this a reality. What I want to draw attention to is a pathway using a kind of OT system known as a historian. Using these it is possible to keep the IT staff and the OT staff happy and productive and providing the data for reporting, business and solution analysis without any compromise to security. I know, it sounds a little too good to be true but let me demonstrate a working example that I have implemented in multiple systems.</p><h3>Historian Replication Magic</h3><ol><li>Historian isn’t a database. The way historian work is using a type of flat file storage. This is for technical reasons including the use of the bufferwriter libraries but we will avoid the why here. What is important is that while there is SQL installed with many historians its not the storage. The files are.</li><li>OLE DB Connectors. These are what historians will install or have available. These Connection Servers (in SQL this isn’t a separate server) allow the use of OpenQuery or T-SQL to extract the flat files back into structured data. Once here we have a table of information. SQL being the standard language of data in computers we are now off. Once you have these queries understood you can extract data from the OT historian.</li><li>Tiered Historians. Many vendors allow the use of Tiered historians. Where data is replicated from one historian to a more central one. Now we can move the data we want but this is where the magic happens! We can setup a single connection from the OT network historian to replicate the data to one in the IT space. Now this is coming from the OT network into the IT network and is a replicated dataset. The OT data is preserved for OT operators and staff to use for trend data etc. Now there is a copy ready for the IT group and its on an IT server. This usually is a service with a single port needed. The direction is one way, the IT historian cannot write to the OT in this model and now you have data using the OLE DB as above you can query that data into your reporting system as SQL queries.</li></ol><h3>Story of a PLC that Gathered Data</h3><p>So now there is a pathway between the protected, OT network historian and the IT staff and accounting, c-suite or anyone needing to report, contextualize or handle any data from the OT systems. There is a catch, what data!</p><p>Well this is where the OT staff come in. In order to gather this data there needs to be data. Tags added to the PLC devices (or DeltaV etc.) that will gather the data. Usually this code will gather on set time, not on scan time on the PLC. However, this data once set in tags is all that is needed. Now the OT staff does need to write this code BUT, remember this is your wheelhouse. You don’t have to lift a finger outside of the control system. Once in place add these to the historian, then set them to replicate to the IT historian and you are set.</p><h3>IT Don’t Panic</h3><p>So IT you will need a historian of your own, also you will need to punch a hole in the firewall to allow the replication to be handled. Once these are in place and the OT PLC code changes are in place. That’s basically it. Now you have the data, use the SQL to use the OLE DB connector and you have structured data for reporting, for Azure DataLakes or whatever you want to do with the data. Now you are using a reflected copy of the data and if the device does go offline for any reason you will have gaps in your data but that is because there is a real gap in the real world. If you need the data held for longer, see your vendors retention settings. There are many ways to customize for your environment but the principals here will get you the data and empower your organization</p><h3>Conclusion</h3><p>I have not mentioned any specific vendors for Historians here and this is in purpose. This is a solution I have done with multiple vendors. The key with this model is that you never ask anyone to work outside their departments skill set and you empower your organization with data securely without needing to compromise or install new equipment. This is a cost saving, targeted approach to a complex problem but by using a historian with replication as the center of the wheel you create a model flexible enough to fit any organization.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3f48d16a1c46" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[LoRaWAN SenseCap K1100 into Home Assistant]]></title>
            <link>https://medium.com/@jon_83595/lorawan-sensecap-k1100-into-home-assistant-phase-1-6d3d7ad3368d?source=rss-6b13a6faed29------2</link>
            <guid isPermaLink="false">https://medium.com/p/6d3d7ad3368d</guid>
            <category><![CDATA[lorawan]]></category>
            <category><![CDATA[sensecap]]></category>
            <category><![CDATA[the-things-network]]></category>
            <category><![CDATA[home-assistant]]></category>
            <category><![CDATA[node-red]]></category>
            <dc:creator><![CDATA[Jonathon Cronin]]></dc:creator>
            <pubDate>Mon, 17 Jul 2023 04:26:38 GMT</pubDate>
            <atom:updated>2023-07-17T15:09:31.629Z</atom:updated>
            <content:encoded><![CDATA[<h3>Summary</h3><p>You want to start with some LoRaWAN systems or similar IoT, IIoT, XIoT devices. The easiest platform for integration of this style system is Home Assistant as it is free, reliable and has a wide base of users. So I set out to take the SenseCap K1100 test kit with Grove sensors using the LoRaWAN connection module to gather, send, parse and provide that information into a working Home Assistant page. This process covers how to accomplish this setup and similar setups for users new to most if not all of these systems. Additionally this utilizes NodeRED for the MQTT connection from The Things Network where we are sending the data to and the Home Assistant. All system setups will be covered in their own sections to help users of all technical levels. This is a working POC model in my own environment and is not theoretical but its also complex and your own implementation may not work perfectly out of the gate. I will do my best to cover ‘gotcha’ and other weird problems I ran into in putting this together.</p><h3>Topics Covered</h3><ol><li>LoRaWAN</li><li>Arduino</li><li>Raspberry Pi</li><li>SenseCap</li><li>Grove</li><li>The Things Network (TTN)</li><li>MQTT</li><li>JSON</li><li>NodeRED</li><li>Home Assistant</li></ol><h3>What I setup</h3><ol><li><a href="https://www.seeedstudio.com/Seeed-Studio-LoRaWAN-Dev-Kit-p-5370.html?queryID=1c3eb3e135593af75b4e845562d0e5f5&amp;objectID=5370&amp;indexName=bazaar_retailer_products">SenseCap K1100 Sensor Prototyping Kit</a></li><li><a href="https://wiki.seeedstudio.com/K1100-quickstart/">SenseCap K1100 Quick Start</a></li><li><a href="https://www.seeedstudio.com/Wio-WM1302-LoRaWAN-Gateway-Module-SPI-US-915-p-5454.html?queryID=56d6dbba5327fdebec9398099fe2beee&amp;objectID=5454&amp;indexName=bazaar_retailer_products">WM1302 SPI LoRaWAN card in a Raspberry PI 3 in a Pi Hat</a></li><li><a href="https://www.thethingsnetwork.org/">The Things Network community edition account</a></li><li><a href="https://www.home-assistant.io/">Home Assistant with NodeRED setup</a></li></ol><h3>Process Flow</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nTyMfwJFIQLFsv9EmGDZiA.png" /></figure><h3>Devices</h3><ol><li>SenseCap K1100 Prototyping Kit</li></ol><p>2. Wio-WM1320 LoRaWAN Gateway module SPI</p><p>3. Raspberry Pi 3B or better</p><p>4. Raspberry Pi Hat (for CPIO 30 Pin out)</p><h3>Part 1: Setting up the SenseCap K1100</h3><p>The SenseCap comes with a setup guide in the container as well as a great how to guide I will include. This also covers the SenseCap app method of connection using a SenseCap gateway as well as Ubidots and Azure IoT setups. I understand that this setup is more complex than those and stands as more of a POC or GTFO of what you can do with zero cost and exposing all of the inner workings of this process. No personal offense will be taken if you choose the other path.</p><ol><li>Connect the display to your computer using a data enabled USB C cable. If you don’t see the disk drive open on its own or available from the display then most likely it is the cable.</li><li>Enter the Bootloader mode on the device by clicking down (restart) on the power button 2 times quickly. See guide from Seeed Studio for more detail.</li><li>Load the latest <a href="https://files.seeedstudio.com/wiki/K1100-quick-start/SenseCraft-v0.2-20220907.uf2">UF2 file</a> and reboot the device</li></ol><p>That should be it for the display Arduino board for now. You can load custom programs to the display. There are sample code and an entire YouTube series provided by Seeed Studio on this that I highly recommend if you are interested in those elements. For our purposes we will push onward.</p><h3>Part 2: Setting up the WM1320 Card</h3><p>Now we will need a gateway setup for the LoRaWAN to connect to. I will skip over the setup of the Raspberry PI image as that is covered in many other places. What you need is a vanilla or Pi OS Lite on a Raspberry Pi 3B+ or better and a Raspberry pi hat and of course the SPI chip and antenna. You will want the device with the installed OS and card and USB connection for the supplementary power on the hat as show.</p><ol><li>Access your Raspberry Pi either through SSH or VNC or directly using a monitor etc.</li><li>Open bash and enter the command</li></ol><pre>sudo raspi-config</pre><p>3. Select Interface options</p><p>4. Enable SPI and I2C connectors in that menu</p><p>Once this phase is complete the card and its connections will be available to the Raspberry Pi and we can proceed in setting up the code and connections to TTN.</p><h3>Part 3: Setting up The Things Network Gateway and Connecting LoRa Gateway.</h3><p>Now the last prep step will be to setup an account with The Things Network for the gateway to connect to. You will need to go to The Things Network and setup an account under their Community Edition. This is a free account and will act as the hub for the data that we will be sending to it from our gateway.</p><p>You should follow the process in this Wiki provided by Seeed Studio for this setup. Its clear and concise and better laid out than anything I can provide.</p><p><a href="https://wiki.seeedstudio.com/WM1302_module/">Setup LoRaWAN Gateway Module and Connecting to TTN</a></p><p>What you will want to keep in mind if the Region you setup as this will inform the MQTT endpoint we will use later.</p><h3>Part 4: Connecting the Sensors to the Gateway locally</h3><p>Now back to the first device we worked on. The small display. You can go to the Uplink button and select LoRa, select the bands in use in your region. You will see a screen for registering this with SenseCap, just click the selection button a couple of time to bypass this. You will see the LoRa Init message come up. The device should register to the gateway you have setup. When it does send a message the LED will go blue. Once the packet has been successfully sent up to the TTN you will see a Send Success message on the device. In the TTN (which does not refresh automatically) you should see hexcode coming in from the device every 5 minutes. You can check this in the Live Message section of your application Live Data. You are looking for the HEX code data as shown here.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*w_QDZrWWHxxkuSaOsKBiyw.png" /></figure><h3>Part 5: Payload Decoder</h3><p>Now you have data coming from your K1100 device through your gateway and coming to the TTN application. This is the first half of the process for the setup. The data coming in is encoded data that needs to be decoded to be meaningful. Luckily this is just a matter of using the payload decoder in the application.</p><p>Raw Data Example</p><p>420927134C800080008000</p><p>You will need to setup the payload decoder. You will need to copy the javascript file. Now there is an older .js called K1100 which you would think is what we need. However, this is for an older standard of TTN. Ignore it! Instead use this JS to parse the data code.</p><pre>function decodeUplink (input, port) {<br>  // data split<br><br>  var bytes = input[&#39;bytes&#39;]<br>  // init<br>  bytes = bytes2HexString(bytes)<br>    .toLocaleUpperCase()<br>  let result = {<br>    &#39;err&#39;: 0, &#39;payload&#39;: bytes, &#39;valid&#39;: true, messages: []<br>  }<br>  let splitArray = dataSplit(bytes)<br>  // data decoder<br>  let decoderArray = []<br>  for (let i = 0; i &lt; splitArray.length; i++) {<br>    let item = splitArray[i]<br>    let dataId = item.dataId<br>    let dataValue = item.dataValue<br>    let messages = dataIdAndDataValueJudge(dataId, dataValue)<br>    decoderArray.push(messages)<br>  }<br>  result.messages = decoderArray<br>  return { data: result }<br>}<br><br>/**<br> * data splits<br> * @param bytes<br> * @returns {*[]}<br> */<br>function dataSplit (bytes) {<br>  let frameArray = []<br><br>  for (let i = 0; i &lt; bytes.length; i++) {<br>    let remainingValue = bytes<br>    let dataId = remainingValue.substring(0, 2)<br>    let dataValue<br>    let dataObj = {}<br>    switch (dataId) {<br>      case &#39;01&#39; :<br>      case &#39;20&#39; :<br>      case &#39;21&#39; :<br>      case &#39;30&#39; :<br>      case &#39;31&#39; :<br>      case &#39;33&#39; :<br>      case &#39;40&#39; :<br>      case &#39;41&#39; :<br>      case &#39;42&#39; :<br>      case &#39;43&#39; :<br>      case &#39;44&#39; :<br>      case &#39;45&#39; :<br>        dataValue = remainingValue.substring(2, 22)<br>        bytes = remainingValue.substring(22)<br>        dataObj = {<br>          &#39;dataId&#39;: dataId, &#39;dataValue&#39;: dataValue<br>        }<br>        break<br>      case &#39;02&#39;:<br>        dataValue = remainingValue.substring(2, 18)<br>        bytes = remainingValue.substring(18)<br>        dataObj = {<br>          &#39;dataId&#39;: &#39;02&#39;, &#39;dataValue&#39;: dataValue<br>        }<br>        break<br>      case &#39;03&#39; :<br>      case &#39;06&#39;:<br>        dataValue = remainingValue.substring(2, 4)<br>        bytes = remainingValue.substring(4)<br>        dataObj = {<br>          &#39;dataId&#39;: dataId, &#39;dataValue&#39;: dataValue<br>        }<br>        break<br>      case &#39;05&#39; :<br>      case &#39;34&#39;:<br>        dataValue = bytes.substring(2, 10)<br>        bytes = remainingValue.substring(10)<br>        dataObj = {<br>          &#39;dataId&#39;: dataId, &#39;dataValue&#39;: dataValue<br>        }<br>        break<br>      case &#39;04&#39;:<br>      case &#39;10&#39;:<br>      case &#39;32&#39;:<br>      case &#39;35&#39;:<br>      case &#39;36&#39;:<br>      case &#39;37&#39;:<br>      case &#39;38&#39;:<br>      case &#39;39&#39;:<br>        dataValue = bytes.substring(2, 20)<br>        bytes = remainingValue.substring(20)<br>        dataObj = {<br>          &#39;dataId&#39;: dataId, &#39;dataValue&#39;: dataValue<br>        }<br>        break<br>      default:<br>        dataValue = &#39;9&#39;<br>        break<br>    }<br>    if (dataValue.length &lt; 2) {<br>      break<br>    }<br>    frameArray.push(dataObj)<br>  }<br>  return frameArray<br>}<br><br>function dataIdAndDataValueJudge (dataId, dataValue) {<br>  let messages = []<br>  switch (dataId) {<br>    case &#39;01&#39;:<br>      let temperature = dataValue.substring(0, 4)<br>      let humidity = dataValue.substring(4, 6)<br>      let illumination = dataValue.substring(6, 14)<br>      let uv = dataValue.substring(14, 16)<br>      let windSpeed = dataValue.substring(16, 20)<br>      messages = [{<br>        measurementValue: loraWANV2DataFormat(temperature, 10), measurementId: &#39;4097&#39;, type: &#39;Air Temperature&#39;<br>      }, {<br>        measurementValue: loraWANV2DataFormat(humidity), measurementId: &#39;4098&#39;, type: &#39;Air Humidity&#39;<br>      }, {<br>        measurementValue: loraWANV2DataFormat(illumination), measurementId: &#39;4099&#39;, type: &#39;Light Intensity&#39;<br>      }, {<br>        measurementValue: loraWANV2DataFormat(uv, 10), measurementId: &#39;4190&#39;, type: &#39;UV Index&#39;<br>      }, {<br>        measurementValue: loraWANV2DataFormat(windSpeed, 10), measurementId: &#39;4105&#39;, type: &#39;Wind Speed&#39;<br>      }]<br>      break<br>    case &#39;02&#39;:<br>      let windDirection = dataValue.substring(0, 4)<br>      let rainfall = dataValue.substring(4, 12)<br>      let airPressure = dataValue.substring(12, 16)<br>      messages = [{<br>        measurementValue: loraWANV2DataFormat(windDirection), measurementId: &#39;4104&#39;, type: &#39;Wind Speed Sensor&#39;<br>      }, {<br>        measurementValue: loraWANV2DataFormat(rainfall, 1000), measurementId: &#39;4113&#39;, type: &#39;RainFall Hourly&#39;<br>      }, {<br><br>        measurementValue: loraWANV2DataFormat(airPressure, 0.1), measurementId: &#39;4101&#39;, type: &#39;Barometric Pressure&#39;<br>      }]<br>      break<br>    case &#39;03&#39;:<br>      let Electricity = dataValue<br>      messages = [{<br>        &#39;3000&#39;: loraWANV2DataFormat(Electricity)<br>      }]<br>      break<br>    case &#39;04&#39;:<br>      let electricityWhether = dataValue.substring(0, 2)<br>      let hwv = dataValue.substring(2, 6)<br>      let bdv = dataValue.substring(6, 10)<br>      let sensorAcquisitionInterval = dataValue.substring(10, 14)<br>      let gpsAcquisitionInterval = dataValue.substring(14, 18)<br>      messages = [{<br>        &#39;3000&#39;: loraWANV2DataFormat(electricityWhether),<br>        &#39;3001&#39;: `${loraWANV2DataFormat(hwv.substring(0, 2))}.${loraWANV2DataFormat(hwv.substring(2, 4))}`,<br>        &#39;3502&#39;: `${loraWANV2DataFormat(bdv.substring(0, 2))}.${loraWANV2DataFormat(bdv.substring(2, 4))}`,<br>        &#39;3900&#39;: parseInt(loraWANV2DataFormat(sensorAcquisitionInterval)) * 60,<br>        &#39;3911&#39;: parseInt(loraWANV2DataFormat(gpsAcquisitionInterval)) * 60<br>      }]<br>      break<br>    case &#39;05&#39;:<br>      let sensorAcquisitionIntervalFive = dataValue.substring(0, 4)<br>      let gpsAcquisitionIntervalFive = dataValue.substring(4, 8)<br>      messages = [{<br>        &#39;3900&#39;: parseInt(loraWANV2DataFormat(sensorAcquisitionIntervalFive)) * 60,<br>        &#39;3911&#39;: parseInt(loraWANV2DataFormat(gpsAcquisitionIntervalFive)) * 60<br>      }]<br>      break<br>    case &#39;06&#39;:<br>      let errorCode = dataValue<br>      let descZh<br>      switch (errorCode) {<br>        case &#39;00&#39;:<br>          descZh = &#39;CCL_SENSOR_ERROR_NONE&#39;<br>          break<br>        case &#39;01&#39;:<br>          descZh = &#39;CCL_SENSOR_NOT_FOUND&#39;<br>          break<br>        case &#39;02&#39;:<br>          descZh = &#39;CCL_SENSOR_WAKEUP_ERROR&#39;<br>          break<br>        case &#39;03&#39;:<br>          descZh = &#39;CCL_SENSOR_NOT_RESPONSE&#39;<br>          break<br>        case &#39;04&#39;:<br>          descZh = &#39;CCL_SENSOR_DATA_EMPTY&#39;<br>          break<br>        case &#39;05&#39;:<br>          descZh = &#39;CCL_SENSOR_DATA_HEAD_ERROR&#39;<br>          break<br>        case &#39;06&#39;:<br>          descZh = &#39;CCL_SENSOR_DATA_CRC_ERROR&#39;<br>          break<br>        case &#39;07&#39;:<br>          descZh = &#39;CCL_SENSOR_DATA_B1_NO_VALID&#39;<br>          break<br>        case &#39;08&#39;:<br>          descZh = &#39;CCL_SENSOR_DATA_B2_NO_VALID&#39;<br>          break<br>        case &#39;09&#39;:<br>          descZh = &#39;CCL_SENSOR_RANDOM_NOT_MATCH&#39;<br>          break<br>        case &#39;0A&#39;:<br>          descZh = &#39;CCL_SENSOR_PUBKEY_SIGN_VERIFY_FAILED&#39;<br>          break<br>        case &#39;0B&#39;:<br>          descZh = &#39;CCL_SENSOR_DATA_SIGN_VERIFY_FAILED&#39;<br>          break<br>        case &#39;0C&#39;:<br>          descZh = &#39;CCL_SENSOR_DATA_VALUE_HI&#39;<br>          break<br>        case &#39;0D&#39;:<br>          descZh = &#39;CCL_SENSOR_DATA_VALUE_LOW&#39;<br>          break<br>        case &#39;0E&#39;:<br>          descZh = &#39;CCL_SENSOR_DATA_VALUE_MISSED&#39;<br>          break<br>        case &#39;0F&#39;:<br>          descZh = &#39;CCL_SENSOR_ARG_INVAILD&#39;<br>          break<br>        case &#39;10&#39;:<br>          descZh = &#39;CCL_SENSOR_RS485_MASTER_BUSY&#39;<br>          break<br>        case &#39;11&#39;:<br>          descZh = &#39;CCL_SENSOR_RS485_REV_DATA_ERROR&#39;<br>          break<br>        case &#39;12&#39;:<br>          descZh = &#39;CCL_SENSOR_RS485_REG_MISSED&#39;<br>          break<br>        case &#39;13&#39;:<br>          descZh = &#39;CCL_SENSOR_RS485_FUN_EXE_ERROR&#39;<br>          break<br>        case &#39;14&#39;:<br>          descZh = &#39;CCL_SENSOR_RS485_WRITE_STRATEGY_ERROR&#39;<br>          break<br>        case &#39;15&#39;:<br>          descZh = &#39;CCL_SENSOR_CONFIG_ERROR&#39;<br>          break<br>        case &#39;FF&#39;:<br>          descZh = &#39;CCL_SENSOR_DATA_ERROR_UNKONW&#39;<br>          break<br>        default:<br>          descZh = &#39;CC_OTHER_FAILED&#39;<br>          break<br>      }<br>      messages = [{<br>        measurementId: &#39;4101&#39;, type: &#39;sensor_error_event&#39;, errCode: errorCode, descZh<br>      }]<br>      break<br>    case &#39;10&#39;:<br>      let statusValue = dataValue.substring(0, 2)<br>      let { status, type } = loraWANV2BitDataFormat(statusValue)<br>      let sensecapId = dataValue.substring(2)<br>      messages = [{<br>        status: status, channelType: type, sensorEui: sensecapId<br>      }]<br>      break<br>    case &#39;20&#39;:<br>      let initmeasurementId = 4175<br>      let sensor = []<br>      for (let i = 0; i &lt; dataValue.length; i += 4) {<br>        let modelId = loraWANV2DataFormat(dataValue.substring(i, i + 2))<br>        let detectionType = loraWANV2DataFormat(dataValue.substring(i + 2, i + 4))<br>        let aiHeadValues = `${modelId}.${detectionType}`<br>        sensor.push({<br>          measurementValue: aiHeadValues, measurementId: initmeasurementId<br>        })<br>        initmeasurementId++<br>      }<br>      messages = sensor<br>      break<br>    case &#39;21&#39;:<br>      // Vision AI:<br>      // AI 识别输出帧<br>      let tailValueArray = []<br>      let initTailmeasurementId = 4180<br>      for (let i = 0; i &lt; dataValue.length; i += 4) {<br>        let modelId = loraWANV2DataFormat(dataValue.substring(i, i + 2))<br>        let detectionType = loraWANV2DataFormat(dataValue.substring(i + 2, i + 4))<br>        let aiTailValues = `${modelId}.${detectionType}`<br>        tailValueArray.push({<br>          measurementValue: aiTailValues, measurementId: initTailmeasurementId, type: `AI Detection ${i}`<br>        })<br>        initTailmeasurementId++<br>      }<br>      messages = tailValueArray<br>      break<br>    case &#39;30&#39;:<br>    case &#39;31&#39;:<br>      // 首帧或者首帧输出帧<br>      let channelInfoOne = loraWANV2ChannelBitFormat(dataValue.substring(0, 2))<br>      let dataOne = {<br>        measurementValue: loraWANV2DataFormat(dataValue.substring(4, 12), 1000),<br>        measurementId: parseInt(channelInfoOne.one),<br>        type: &#39;Measurement&#39;<br>      }<br>      let dataTwo = {<br>        measurementValue: loraWANV2DataFormat(dataValue.substring(12, 20), 1000),<br>        measurementId: parseInt(channelInfoOne.two),<br>        type: &#39;Measurement&#39;<br>      }<br>      let cacheArrayInfo = []<br>      if (parseInt(channelInfoOne.one)) {<br>        cacheArrayInfo.push(dataOne)<br>      }<br>      if (parseInt(channelInfoOne.two)) {<br>        cacheArrayInfo.push(dataTwo)<br>      }<br>      cacheArrayInfo.forEach(item =&gt; {<br>        messages.push(item)<br>      })<br>      break<br>    case &#39;32&#39;:<br>      let channelInfoTwo = loraWANV2ChannelBitFormat(dataValue.substring(0, 2))<br>      let dataThree = {<br>        measurementValue: loraWANV2DataFormat(dataValue.substring(2, 10), 1000),<br>        measurementId: 4164 + parseInt(channelInfoTwo.one),<br>        type: &#39;Measurement&#39;<br>      }<br>      let dataFour = {<br>        measurementValue: loraWANV2DataFormat(dataValue.substring(10, 18), 1000),<br>        measurementId: 4164 + parseInt(channelInfoTwo.two),<br>        type: &#39;Measurement&#39;<br>      }<br>      if (parseInt(channelInfoTwo.one)) {<br>        messages.push(dataThree)<br>      }<br>      if (parseInt(channelInfoTwo.two)) {<br>        messages.push(dataFour)<br>      }<br>      break<br>    case &#39;33&#39;:<br>      let channelInfoThree = loraWANV2ChannelBitFormat(dataValue.substring(0, 2))<br>      let dataFive = {<br>        measurementValue: loraWANV2DataFormat(dataValue.substring(4, 12), 1000),<br>        measurementId: parseInt(channelInfoThree.one),<br>        type: &#39;Measurement&#39;<br>      }<br>      let dataSix = {<br>        measurementValue: loraWANV2DataFormat(dataValue.substring(12, 20), 1000),<br>        measurementId: parseInt(channelInfoThree.two),<br>        type: &#39;Measurement&#39;<br>      }<br>      if (parseInt(channelInfoThree.one)) {<br>        messages.push(dataFive)<br>      }<br>      if (parseInt(channelInfoThree.two)) {<br>        messages.push(dataSix)<br>      }<br><br>      break<br>    case &#39;34&#39;:<br>      let model = loraWANV2DataFormat(dataValue.substring(0, 2))<br>      let GPIOInput = loraWANV2DataFormat(dataValue.substring(2, 4))<br>      let simulationModel = loraWANV2DataFormat(dataValue.substring(4, 6))<br>      let simulationInterface = loraWANV2DataFormat(dataValue.substring(6, 8))<br>      messages = [{<br>        dataId: 34, &#39;3570&#39;: model, &#39;3571&#39;: GPIOInput, &#39;3572&#39;: simulationModel, &#39;3573&#39;: simulationInterface<br>      }]<br>      break<br>    case &#39;35&#39;:<br>    case &#39;36&#39;:<br>      let channelTDOne = loraWANV2ChannelBitFormat(dataValue.substring(0, 2))<br>      let channelSortTDOne = 3920 + (parseInt(channelTDOne.one) - 1) * 2<br>      let channelSortTDTWO = 3921 + (parseInt(channelTDOne.one) - 1) * 2<br>      messages = [{<br>        [channelSortTDOne]: loraWANV2DataFormat(dataValue.substring(2, 10), 1000),<br>        [channelSortTDTWO]: loraWANV2DataFormat(dataValue.substring(10, 18), 1000)<br>      }]<br>      break<br>    case &#39;37&#39;:<br>      let channelTDInfoTwo = loraWANV2ChannelBitFormat(dataValue.substring(0, 2))<br>      let channelSortOne = 3920 + (parseInt(channelTDInfoTwo.one) - 1) * 2<br>      let channelSortTWO = 3921 + (parseInt(channelTDInfoTwo.one) - 1) * 2<br>      messages = [{<br>        [channelSortOne]: loraWANV2DataFormat(dataValue.substring(2, 10), 1000),<br>        [channelSortTWO]: loraWANV2DataFormat(dataValue.substring(10, 18), 1000)<br>      }]<br>      break<br>    case &#39;38&#39;:<br>      let channelTDInfoThree = loraWANV2ChannelBitFormat(dataValue.substring(0, 2))<br>      let channelSortThreeOne = 3920 + (parseInt(channelTDInfoThree.one) - 1) * 2<br>      let channelSortThreeTWO = 3921 + (parseInt(channelTDInfoThree.one) - 1) * 2<br>      messages = [{<br>        [channelSortThreeOne]: loraWANV2DataFormat(dataValue.substring(2, 10), 1000),<br>        [channelSortThreeTWO]: loraWANV2DataFormat(dataValue.substring(10, 18), 1000)<br>      }]<br>      break<br>    case &#39;39&#39;:<br>      let electricityWhetherTD = dataValue.substring(0, 2)<br>      let hwvTD = dataValue.substring(2, 6)<br>      let bdvTD = dataValue.substring(6, 10)<br>      let sensorAcquisitionIntervalTD = dataValue.substring(10, 14)<br>      let gpsAcquisitionIntervalTD = dataValue.substring(14, 18)<br>      messages = [{<br>        dataId: parseInt(dataId),<br>        &#39;3000&#39;: loraWANV2DataFormat(electricityWhetherTD),<br>        &#39;3001&#39;: `${loraWANV2DataFormat(hwvTD.substring(0, 2))}.${loraWANV2DataFormat(hwvTD.substring(2, 4))}`,<br>        &#39;3502&#39;: `${loraWANV2DataFormat(bdvTD.substring(0, 2))}.${loraWANV2DataFormat(bdvTD.substring(2, 4))}`,<br>        &#39;3900&#39;: parseInt(loraWANV2DataFormat(sensorAcquisitionIntervalTD)) * 60,<br>        &#39;3912&#39;: parseInt(loraWANV2DataFormat(gpsAcquisitionIntervalTD))<br>      }]<br>      break<br>    case &#39;40&#39;:<br>    case &#39;41&#39;:<br>      let lightIntensity = dataValue.substring(0, 4)<br>      let loudness = dataValue.substring(4, 8)<br>      // X<br>      let accelerateX = dataValue.substring(8, 12)<br>      // Y<br>      let accelerateY = dataValue.substring(12, 16)<br>      // Z<br>      let accelerateZ = dataValue.substring(16, 20)<br>      messages = [{<br>        measurementValue: loraWANV2DataFormat(lightIntensity), measurementId: &#39;4193&#39;, type: &#39;Light Intensity&#39;<br>      }, {<br>        measurementValue: loraWANV2DataFormat(loudness), measurementId: &#39;4192&#39;, type: &#39;Sound Intensity&#39;<br>      }, {<br><br>        measurementValue: loraWANV2DataFormat(accelerateX, 100), measurementId: &#39;4150&#39;, type: &#39;AccelerometerX&#39;<br>      }, {<br><br>        measurementValue: loraWANV2DataFormat(accelerateY, 100), measurementId: &#39;4151&#39;, type: &#39;AccelerometerY&#39;<br>      }, {<br><br>        measurementValue: loraWANV2DataFormat(accelerateZ, 100), measurementId: &#39;4152&#39;, type: &#39;AccelerometerZ&#39;<br>      }]<br>      break<br>    case &#39;42&#39;:<br>      let airTemperature = dataValue.substring(0, 4)<br>      let AirHumidity = dataValue.substring(4, 8)<br>      let tVOC = dataValue.substring(8, 12)<br>      let CO2eq = dataValue.substring(12, 16)<br>      let soilMoisture = dataValue.substring(16, 20)<br>      messages = [{<br>        measurementValue: loraWANV2DataFormat(airTemperature, 100), measurementId: &#39;4097&#39;, type: &#39;Air Temperature&#39;<br>      }, {<br>        measurementValue: loraWANV2DataFormat(AirHumidity, 100), measurementId: &#39;4098&#39;, type: &#39;Air Humidity&#39;<br>      }, {<br>        measurementValue: loraWANV2DataFormat(tVOC), measurementId: &#39;4195&#39;, type: &#39;Total Volatile Organic Compounds&#39;<br>      }, {<br>        measurementValue: loraWANV2DataFormat(CO2eq), measurementId: &#39;4100&#39;, type: &#39;CO2&#39;<br>      }, {<br>        measurementValue: loraWANV2DataFormat(soilMoisture), measurementId: &#39;4196&#39;, type: &#39;Soil moisture intensity&#39;<br>      }]<br>      break<br>    case &#39;43&#39;:<br>    case &#39;44&#39;:<br>      let headerDevKitValueArray = []<br>      let initDevkitmeasurementId = 4175<br>      for (let i = 0; i &lt; dataValue.length; i += 4) {<br>        let modelId = loraWANV2DataFormat(dataValue.substring(i, i + 2))<br>        let detectionType = loraWANV2DataFormat(dataValue.substring(i + 2, i + 4))<br>        let aiHeadValues = `${modelId}.${detectionType}`<br>        headerDevKitValueArray.push({<br>          measurementValue: aiHeadValues, measurementId: initDevkitmeasurementId, type: `AI Detection ${i}`<br>        })<br>        initDevkitmeasurementId++<br>      }<br>      messages = headerDevKitValueArray<br>      break<br>    case &#39;45&#39;:<br>      let initTailDevKitmeasurementId = 4180<br>      for (let i = 0; i &lt; dataValue.length; i += 4) {<br>        let modelId = loraWANV2DataFormat(dataValue.substring(i, i + 2))<br>        let detectionType = loraWANV2DataFormat(dataValue.substring(i + 2, i + 4))<br>        let aiTailValues = `${modelId}.${detectionType}`<br>        messages.push({<br>          measurementValue: aiTailValues, measurementId: initTailDevKitmeasurementId, type: `AI Detection ${i}`<br>        })<br>        initTailDevKitmeasurementId++<br>      }<br>      break<br>    default:<br>      break<br>  }<br>  return messages<br>}<br><br>/**<br> *<br> * data formatting<br> * @param str<br> * @param divisor<br> * @returns {string|number}<br> */<br>function loraWANV2DataFormat (str, divisor = 1) {<br>  let strReverse = bigEndianTransform(str)<br>  let str2 = toBinary(strReverse)<br>  if (str2.substring(0, 1) === &#39;1&#39;) {<br>    let arr = str2.split(&#39;&#39;)<br>    let reverseArr = arr.map((item) =&gt; {<br>      if (parseInt(item) === 1) {<br>        return 0<br>      } else {<br>        return 1<br>      }<br>    })<br>    str2 = parseInt(reverseArr.join(&#39;&#39;), 2) + 1<br>    return &#39;-&#39; + str2 / divisor<br>  }<br>  return parseInt(str2, 2) / divisor<br>}<br><br>/**<br> * Handling big-endian data formats<br> * @param data<br> * @returns {*[]}<br> */<br>function bigEndianTransform (data) {<br>  let dataArray = []<br>  for (let i = 0; i &lt; data.length; i += 2) {<br>    dataArray.push(data.substring(i, i + 2))<br>  }<br>  // array of hex<br>  return dataArray<br>}<br><br>/**<br> * Convert to an 8-digit binary number with 0s in front of the number<br> * @param arr<br> * @returns {string}<br> */<br>function toBinary (arr) {<br>  let binaryData = arr.map((item) =&gt; {<br>    let data = parseInt(item, 16)<br>      .toString(2)<br>    let dataLength = data.length<br>    if (data.length !== 8) {<br>      for (let i = 0; i &lt; 8 - dataLength; i++) {<br>        data = `0` + data<br>      }<br>    }<br>    return data<br>  })<br>  let ret = binaryData.toString()<br>    .replace(/,/g, &#39;&#39;)<br>  return ret<br>}<br><br>/**<br> * sensor<br> * @param str<br> * @returns {{channel: number, type: number, status: number}}<br> */<br>function loraWANV2BitDataFormat (str) {<br>  let strReverse = bigEndianTransform(str)<br>  let str2 = toBinary(strReverse)<br>  let channel = parseInt(str2.substring(0, 4), 2)<br>  let status = parseInt(str2.substring(4, 5), 2)<br>  let type = parseInt(str2.substring(5), 2)<br>  return { channel, status, type }<br>}<br><br>/**<br> * channel info<br> * @param str<br> * @returns {{channelTwo: number, channelOne: number}}<br> */<br>function loraWANV2ChannelBitFormat (str) {<br>  let strReverse = bigEndianTransform(str)<br>  let str2 = toBinary(strReverse)<br>  let one = parseInt(str2.substring(0, 4), 2)<br>  let two = parseInt(str2.substring(4, 8), 2)<br>  let resultInfo = {<br>    one: one, two: two<br>  }<br>  return resultInfo<br>}<br><br>/**<br> * data log status bit<br> * @param str<br> * @returns {{total: number, level: number, isTH: number}}<br> */<br>function loraWANV2DataLogBitFormat (str) {<br>  let strReverse = bigEndianTransform(str)<br>  let str2 = toBinary(strReverse)<br>  let isTH = parseInt(str2.substring(0, 1), 2)<br>  let total = parseInt(str2.substring(1, 5), 2)<br>  let left = parseInt(str2.substring(5), 2)<br>  let resultInfo = {<br>    isTH: isTH, total: total, left: left<br>  }<br>  return resultInfo<br>}<br><br>function bytes2HexString (arrBytes) {<br>  var str = &#39;&#39;<br>  for (var i = 0; i &lt; arrBytes.length; i++) {<br>    var tmp<br>    var num = arrBytes[i]<br>    if (num &lt; 0) {<br>      tmp = (255 + num + 1).toString(16)<br>    } else {<br>      tmp = num.toString(16)<br>    }<br>    if (tmp.length === 1) {<br>      tmp = &#39;0&#39; + tmp<br>    }<br>    str += tmp<br>  }<br>  return str<br>}</pre><p>Copy this into the Uplink decoder. You can leave the down link formatted as none. You can copy the RAW example above and use the decoder to see the JSON that is output and represents what TTN will present to our next step in MQTT.</p><h3>Part 6 MQTT to Home Assistant</h3><p>Now you have formatted JSON data parsed from your payload available. How you access this is through MQTT. There are many ways to approach this connection but in this case I will be covering this through NodeRED.</p><ol><li>In your Home Assistant setup NodeRED and make sure its available.</li><li>Setup a Flow for MQTT Connection</li><li>Parse the JSON data</li><li>Create a Helper</li><li>Send the resulting data to the Helper for view in Home Assistant</li></ol><p>Once you have NodeRED Setup you will want to create a flow as shown here.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/851/1*Yw3Fk6Pqzy62hmzjTgXAgA.png" /></figure><ol><li>Setup the MQTT</li></ol><p>When you are in the TTN webpage click on Integrations and check MQTT. Every single application automatically exposes a MQTT for you to connect to. You will get the information you need to setup the MQTT in NodeRED.</p><p>2. NodeRED setup MQTT connection</p><p>Drag a MQTT into your flow and you will see there is a edit for the MQTT Server. You will use the information you gathered before to fill out this page.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/631/1*XWBw_kNwA0QqFQDwO8KmZg.png" /></figure><p>Then for the Security Tab you will use the information as shown here. Format is specified on the TTN page for community edition. This is different for the Enterprise Edition of the offering. The password is a API key you will want to create that before finalizing these settings.</p><p>Once this is complete you will see Connected as the status in NodeRED after deploying this setup.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/645/1*XJQy6chFUy6eqyD9vJfoKg.png" /></figure><p>Now you save this configuration for the MQTT server connection. Now we need to setup the Topic we are going to subscribe to.</p><p>3. Topic to subscribe to</p><p>You will need to in the first page of the MQTT configuration setup the Topic that you want to subscribe to. If you want to test you can subscribe to # and this will give you everything.</p><p>Also TTN only supports QOS = 0</p><p>Example Topic for data:</p><p>v3/&lt;APP NAME&gt;@ttn/devices/eui-&lt;DEVICE ID&gt;/up</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/505/1*AlwaDWl8zq19ZlSPE_keyA.png" /></figure><p>4. Parsing the JSON</p><p>Now you have the JSON from the Payload decoder available in Home Assistant. However, its just a bunch of JSON. I will provide the gross and incorrect but easiest to follow way to parse out some single information from the JSON. This is the Temp that I determined by analysis of the JSON coming in via MQTT. This again is just a brute force example of grabbing the data.</p><pre>if (msg.payload.uplink_message.decoded_payload.messages[0][0].measurementId == &quot;4097&quot;){<br>    var newMsg = { payload: msg.payload.uplink_message.decoded_payload.messages[0][0].measurementValue};<br>    return newMsg;<br>}</pre><p>5. Helper and Seeing the Data</p><p>Now the last step you will need a Helper. If you go to add integrations and click on the helper tab and setup a new number helper.</p><p>Now we will setup a Service Call in NodeRED and point the payload from the parsing JSON to the Helper.</p><p>Now that data will show up in that Helper and can be added to Home Assistant.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1013/1*NwcqSzbZN2scDBbqKfz2Pg.png" /></figure><h3>Conclusion</h3><p>So now you have a template to build a sensor network using LoRaWAN and your Home Assistant sitting as the collector or center of your IoT design. This again is all POC and the code, parsing and display should all be redesigned in your project but instead of starting out from scratch as I did at the beginning of this project. You should be starting with a working data model and with no cost outside of equipment cost.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6d3d7ad3368d" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>