CC BY
37
Научни трудове на Съюза на учените в България - Пловдив. Серия В. Техника и технологии. Том XVII, ISSN 1311 -9419 (Print); ISSN 2534-9384 (Online), 2019. Scientific Works of the Union of Scientists in Bulgaria - Plovdiv. Series C. Technics and Technologies. Vol. XVII., ISSN 1311 -9419 (Print); ISSN 2534-9384 (Online), 2019
LORAWAN BASED SYSTEM TOR MEASURMENT AND MONITORING OF TEMPERATUR AND HUMIDITY IN DATA CENTERS AND SERVER ROOMS
Dimitar Tokmakov, Sotir Sotirov , Slavi Gluhov University of Plovdiv "Paisii Hilendarski", Plovdiv, Bulgaria
Abstract: This paper describes the realization of a wireless temperature and relative humidity measurement node used in datacenters and server rooms that transmit data to a LoraWan gateway via a LoraWan communication link. Using a wireless sensor node gives a number of advantages, such as wireless installation, physical independence from the data center infrastructure, battery power supply with low power consumption. Data from the LoraWan gateway are transmitted into an Internet-based cloud application for post processing and visualization, which enables further optimization of air conditioning and ventilation systems.
Keywords: LoraWan node, measuring temperature and relative humidity in data centers and server rooms
Introduction: Modern server rooms and data centers have systems for monitoring, measuring and controlling environmental parameters such as temperature, humidity, leakage, smoke, and so on. These systems are built on 19 " chassis and require their mounting in modular cabinets and in 19" RAK systems. The sensors are connected to the measuring system via wires and the data being sent for processing and visualization to client applications using the data center's communication links. The power supply of these systems is provided by the data center infrastructure. The presence of connecting wires for the measurement sensors as well as the need to connect to an electrical supply in server rooms limits their installation at random locations in the server premises. The ability to mount wireless sensors that are independent of the room infrastructure is important when the data from these sensors is used to optimize the operation of climate and ventilation systems in order to ensure energy efficiency, in addition to critical alarm alerts.
The realization of a remote temperature and humidity measurement system in server rooms based on miniature battery-powered sensors and the transmission of data through the LorraWan Communication Network is an unresolved problem that we are trying to address with this development.
Materials and methods:
Figure 1 shows the architecture of the measurement system for temperature and relative humidity monitoring in server premises and data centers by using the LoRaWAN Media Access Control (MAC) protocol for WAN networks [3].
LoRa
MIT«
LoraWan Internet claud
sensor nodes LoraWan Gateway app|jcation
Remote Moflrtonng
Client applications
Fig. 1 Architecture of LoraWan measurement systems for temperature and relative
humidity
The measuring system consists of end devices - LoraWan sensor nodes built by a microcontroller, temperature and relative humidity sensor, and a LoraWan radio modem running at 868MHz and powered by a 3.7 V Li-Ion rechargeable battery type 18650.
The sensor node measures the temperature and relative humidity and sends the data through LoraWan communication channel to the LoraWan gateway located near the data center. From the LoraWan gateway, the data is transferred to an Internet-based cloud application via the IPsec protocol and from there through the HTTPS protocol can be used by various client applications. Figure 2 shows the block diagram of the wireless measuring node.
1-Wire
ÍF
3.7 V 18650 LDO 3.3V
Microcontroller ATMEGA328
868MHz
V
SPI
RFM95W Lora Radio Transceiver
DHT 22
Fig.2 Block diagram of the LoraWan measuring node
The temperature and relative humidity measuring node is built up by the microcontroller ATMEGA328, the Lora radio transceiver module RFM95W and the capacitive digital sensor DHT22 (AM2302). DHT22 measures a temperature of -40 ° to 80 ° C and relative air humidity in the range 0% - 99%. It has a pre-calibrated digital output and features great reliability and stability. The accuracy of humidity measurement is +/- 2% and at +/- 0.5%. The interface of the sensor is 1-Wire, the refresh rate is 1Hz. The power supply to the sensor node uses 3.7V Li-Ion battery type 18650 and linear low drop regulator for 3.3 V with low quiescent current.
The ATMEGA 328 microcontroller reads data from the DHT22 sensor, encodes them in the Cayenne LPP format, and sends them via the radio transmitter modem RFM95W to the LoraWan gateway using the Lora communication channel. The measuring node firmware is tuned to send the data every 5 minutes, the transmission time is about 5 seconds, and the remaining time node goes into low-power mode.
Figure 3 shows the block diagram of the LoraWan gateway. It is built on a single board computer with ARM-Cortex processor, USB, Ethernet and SPI interfaces, to which are connected 3G modem, GPS module and LoraWan concentrator RAK 831 based on Semtech SX1301. The LoraWan concentrator has an external dipole antenna with 5dbi amplification. The gateway uses a SX1301-based board that provides 8-channel for downlink and 1 uplink channel to the standard Europe-wide 868MHz plan. The Internet connection of the gateway is via an Ethernet network interface and a backup one from the USB 3G modem. The 5 V power supply for the single board computer is provided by a PoE converter from 48V to 5V. LoraWan gateway listens for all LoraWan protocol packets and transmit them to an Internet-based cloud service in which data is archived and translated to various client applications. By processing this data from the appropriate software, a room temperature profile can be constructed to effectively manage climate and ventilation systems for energy efficiency and to achieve normal server room working conditions.
Fig. Block diagram of 8-channel LoraWan gateway
Results and discussion: Fig. 3 shows the data measured by the system received by the LoraWan gateway and sent to an internet cloud server application. The figure shows the visualized data from the client application using the myDevices platform via the Cayenne LPP protocol.
Cayenne + [r„te nsw proiert
О Cayenne LPP g Cayenne LPP Humidity{2) Щ Location dill RSSI
Hill '.ИИ
| Temperature (1J
IM <1 IM О
Temperature {1} Humidity (2) 5NR RSSI
i 18,20 ♦'60.50 ■иНб.ОО ,1111-111.00
Celsius Percent (%) Decibels dBm
Карта Сателит
Ol
¿J / ,
нически верситет ' София, ллиал...
ПЛОВДИВСКИ УНИВЕРСИТЕТ j
пдисий ХИЛЕНДДРСКИ
§ +
Fig.3 Temperature and relative humidity data sent to the server every 5 min.
The measured temperature, relative humidity, signal-to-noise ratio, received signal strength indicator (RSSI) and the time of the last measurement are displayed. An interactive map shows the physical position of the LoraWan measuring node.
Conclusion: The measurement system presented in this paper measures real-time temperature and relative humidity in data centers and server rooms. Using LoraWan wireless communication link the data is forwarded by the LoraWan gateway and sent to the Internet cloud application for processing, archiving and visualization. The LoraWan gateway and sensor node described in the present work are installed in the University of Plovdiv Paisii Hilendarski, Bulgaria.
The implementation of LoraWan based measurement system has a number of advantages over traditional data center measurement and monitoring systems: possibility of mounting the measuring node at random location without connecting wires, battery power supply with low power consumption, collection and processing of measurement data for their use to optimize the operation of the air conditioning and ventilation system.
ACKNOLEDGMENT This work was funded by the University of Plovdiv "Paisii Hilendarski" science fund NPD reference No.MU17-FF-010
References:
[1]. LPWA Technology for IoT, Nable Communications, 2016
[2] LPWA Technologies(Unlock New IoT Market Potentiol), LoRaAlliance, 2015
[3] A technical overview of LoRa and LoRaWAN, LoRaAlliance, 2015
