A LoRaWAN network polling scheduling method and system

Abstract

The present invention discloses a LoRaWAN network polling scheduling method and system. The present invention divides the frequency band range of the LoRaWAN uplink channel into units of gateways, and binds the terminal address to the gateway in the network server; before the terminal is connected to the network, the network access multicast message is monitored at the agreed network access multicast frequency point, and the network access multicast message is responded to and listened to when the network access multicast message contains its own address; the network server sends a multicast message through the gateway, and before each round of sending the service polling multicast message, the terminal that did not reply in the last polling and the newly added terminal are screened out, and the network access multicast message is sent to these terminals; after receiving the service polling multicast message, the terminal reports data only when the message contains its own address, and when reporting, the working frequency in the frequency point group is determined according to the order of the terminal's own address in the address list. The present invention can solve the problems of packet loss, conflict, and collision in the LoRaWAN wireless network, and fully improve the utilization rate of the wireless channel.

Description

Technical Field

[0001] The present invention belongs to the field of LoRaWAN wireless access networking management, relates to a management method for improving wireless channel utilization during LoRaWAN network networking and transmission, and specifically relates to a LoRaWAN network polling scheduling method and system. Background Art

[0002] LoRa is a wireless technology in Low Power Wide Area Network (LPWAN), which has the characteristics of long distance, low power consumption, low speed and low cost, and is suitable for large-scale deployment. The coverage range can be a city or even a country, and the collected data can be integrated into the public data platform. Projects such as smart cities, smart transportation, and smart factories can also be established. These data can also be used for later big data mining, which is beneficial to projects such as urban management, social public management, industrial Internet, and development consulting.

[0003] The LoRaWAN network architecture consists of three parts, including terminals, gateways (also called base stations), and network servers (NS). When the LoRaWAN network terminal reports data, the terminal sends wireless data to the outside world; after the gateway receives the wireless signal, it reports the data to the network server through Ethernet or the wireless operator network; the network server saves the received data and identifies the type of data packet reported. If it is a data packet that needs to be replied, it pushes the reply message to the gateway, which is sent out by the gateway in a wireless form. The interaction protocol between the gateway and the network server is a complete TCP / IP protocol, and the terminal and the gateway interact wirelessly, using the LoRaWAN protocol belonging to the MAC layer.

[0004] In the LoRaWAN specification, two terminal access modes are defined, namely Over-The-Air Activation (OTAA) and Activation By Personalization (ABP). In OTAA mode, NwkSKey (Network Session Key) and AppSKey (Application Session Key) are automatically generated based on the DevNonce randomly generated by the terminal and the AppNonce randomly generated by the network server, and the network server assigns the terminal's unique address; in ABP mode, fixed NwkSKey and AppSKey are used, and the terminal's unique address is also burned into the chip at the factory and cannot be changed during online use. The terminal address is referred to as DevAddr. Compared with the two modes, OTAA has better security, while ABP is easier to access and use.

[0005] In the LoRaWAN specification, three working modes are defined for terminal nodes, Class A, Class B, and Class C. In Class A working mode, after sending data, the terminal will open a downlink data receiving window after 1 second, referred to as RX1. If no downlink data is received in RX1, then a RX2 receiving window will be opened after 1 second. The receiving frequencies of RX1 and RX2 are different. In Class B working mode, in addition to receiving data from the network server in RX1 and RX2, the terminal node can also periodically open a receiving window to receive data sent to it by the network server after synchronizing the timestamp with the network server. In Class C working mode, in addition to opening the RX1 and RX2 receiving windows after the terminal actively reports, the receiving window is always opened at the Class C frequency in other time periods, and it is in a constant receiving mode.

[0006] According to the LoRaWAN1.01 specification, the uplink frequency of the 470 frequency band in China ranges from 470.3MHz to 489.3MHz, with each 200KHz being a frequency, and one frequency corresponding to a working channel; for example, 470.3MHz is a channel, 470.5MHz is a channel, and so on, for a total of 96 channels. The SX1301 gateway module commonly used in LoRaWAN gateways is a half-duplex receiving module that can simultaneously receive 8 consecutive uplink frequencies when working. For example, a gateway can simultaneously receive data from terminal nodes with uplink frequencies of 470.3MHz, 470.5MHz, 470.7MHz, 470.9MHz, 471.1MHz, 471.3MHz, 471.5MHz, and 471.7MHz.

[0007] Regardless of the mode in which the terminal node works, there will be packet loss in the LoRaWAN network. The following analyzes the causes of packet loss from the gateway side and the terminal side respectively.

[0008] The half-duplex mode transceiver module used by the LoRaWAN gateway will close the reporting receiving channel when sending downlink messages, which will cause packet loss of data reported by some terminal sensors when sending downlink data. If the reporting data of two or more terminals is received at the same frequency, due to co-frequency interference during demodulation, it is better to be able to decode the reporting data of one of the terminals, but it is also possible that none of them can be decoded.

[0009] In the LoRaWAN specification, the reporting mode of the terminal node is an ALOHA method. Generally, terminals within the frequency range share these frequencies. They report data according to their own trigger mechanisms and have no constraints on each other. This will cause two or more terminal sensors to send data at the same time, resulting in conflicts and collisions, and finally neither of them can report successfully. There are some detection technologies that can pre-detect whether the frequency is occupied when reporting data. If it is occupied, change a frequency; if it is idle, report data directly. This detection will still cause conflicts and collisions when there are a large number of terminal nodes, resulting in packet loss and reporting failure.

[0010] The data reporting of the LoRaWAN network is determined by the automatic triggering conditions of the terminal sensor. In this case, there is an imbalance in channel utilization. In certain periods of time, the utilization is high or even overloaded, which will cause a large number of packet losses. If the sensor has a retransmission mechanism, it will cause a channel storm and paralyze the entire frequency band. In certain periods of time, the utilization may be very low. Such an imbalance in channel utilization is not conducive to the implementation of bearer services. Summary of the invention

[0011] Purpose of the invention: In view of the problems existing in the above-mentioned prior art, the purpose of the present invention is to provide a LoRaWAN network polling scheduling method and system to improve the utilization rate of frequency resources, solve the problems of packet loss, conflict and collision in the LoRaWAN wireless network, and fully improve the utilization rate of wireless channels.

[0012] Technical solution: In order to achieve the above-mentioned invention object, the present invention adopts the following technical solution:

[0013] A LoRaWAN network polling scheduling method comprises the following steps:

[0014] The frequency band range of the LoRaWAN uplink channel is divided into gateways. Each gateway works in one management domain, and different management domains have different working frequency bands.

[0015] Bind the terminal address to the gateway in the network server. The data of a specified terminal can only be received by a specified gateway. Different terminals bound to the same gateway use different working frequencies.

[0016] Before the terminal is connected to the network, it monitors the network multicast message at the agreed network multicast frequency. After receiving the network multicast message, it checks whether it has its own address. If not, it continues to monitor. If so, it responds and uses the multicast downlink frequency and spreading factor in the network multicast message to start listening to the service polling multicast message.

[0017] The network server sends service polling multicast messages and network access multicast messages through the gateway. Before sending service polling multicast messages in each round, the network server selects the terminals that did not respond in the last polling and the newly added terminals, and sends network access multicast messages to these terminals; the network access multicast message includes the multicast downlink frequency and spreading factor used by the gateway, the working frequency group of the gateway and the terminal address list. The terminal address in the service polling multicast message is an agreed fixed value, and the information payload content is the response terminal address table;

[0018] After receiving the service polling multicast message, the terminal will report data only when its own address is contained in the multicast message. When reporting, the working frequency in the frequency group is determined according to the order of the terminal's own address in the address list.

[0019] Preferably, the frequency band range of the LoRaWAN uplink channel is divided into 12 frequency bands, with 8 frequency points as a section. The network server divides 8 terminals into a group, and each gateway is bound to a maximum of 8 terminals.

[0020] Preferably, the working mode of the terminal is Class C mode.

[0021] Preferably, three network-entry multicast frequencies are agreed upon in the LoRaWAN downlink channel frequency band, and the terminal has an automatic frequency switching mechanism. If the network-entry multicast message is not monitored at one of the frequencies, the frequency is switched and the monitoring is performed again.

[0022] Preferably, the message header of the incoming multicast message consists of a version, a frequency group number, a multicast downlink frequency, a spreading factor and a reserved field.

[0023] Preferably, the response terminal address table in the service polling multicast message consists of a version, a frequency group number and a terminal address list bound to a gateway.

[0024] As a preferred embodiment, each gateway will leave a reception waiting time after sending down the service polling multicast message, and will not send a message to the terminal during the waiting time.

[0025] A LoRaWAN network polling scheduling system comprises a network server, a gateway and a terminal, characterized in that the frequency band range of the LoRaWAN uplink channel is divided by gateway, each gateway works in a management domain, and the working frequency bands of different management domains are different; the network server is used to bind the terminal address to the gateway, the data of the designated terminal can only be received by a designated gateway, and different terminals bound to the same gateway use different working frequencies; and the gateway sends a service polling multicast message and a network access multicast message, and before each round of sending the service polling multicast message, the terminals that did not reply in the last polling and the newly added terminals are screened out, and the network access multicast messages are sent to these terminals; the network access multicast message includes the multicast downlink frequency and spreading factor used by the gateway, the working frequency group of the gateway and the terminal address list, the terminal address in the service polling multicast message is an agreed fixed value, and the information load content is a response terminal address table;

[0026] Before accessing the network, the terminal monitors the network access multicast message at the agreed network access multicast frequency. After receiving the network access multicast message, it detects whether it has its own address. If not, it continues to monitor. If so, it responds and uses the multicast downlink frequency and spreading factor in the network access multicast message to start listening to the service polling multicast message. After receiving the service polling multicast message, the terminal will report data only if the multicast message contains its own address. When reporting, the working frequency in the frequency group is determined according to the order of the terminal's own address in the address list.

[0027] Beneficial effects: The present invention divides the entire reporting channel range into multiple frequency bands, specifies the working frequency band of the gateway, binds the terminal to the reporting gateway, and the network server groups the terminals, and sends network access multicast messages and service polling multicast messages to the terminals in the group in turn, so that each terminal has an exclusive reporting frequency idle time slot, avoiding the conflict and collision caused by the uncertainty of the sending time slot in the traditional LoRaWAN network, resulting in packet loss. The present invention is suitable for application scenarios with high acquisition rate, low packet loss rate and high reliability, and can reduce the packet loss of reported data while fully improving the channel utilization rate, and even achieve zero packet loss under normal operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a schematic diagram of the CN470-510 frequency band division.

[0029] Figure 2 It is a schematic diagram of a single service multicast polling process in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The present invention is further illustrated below in conjunction with the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention. After reading the present invention, modifications to various equivalent forms of the present invention by those skilled in the art all fall within the scope defined by the appended claims of the present application.

[0031] The embodiment of the present invention discloses a LoRaWAN network polling scheduling method, which first divides the frequency band range of the LoRaWAN uplink channel into units of gateways, and binds the terminal address to the gateway in the network server, so that the data of the designated terminal can only be received by a designated gateway, and different terminals bound to the same gateway use different working frequencies; the network server sends a service polling multicast message and a network access multicast message through the gateway, and before sending the service polling multicast message in each round, the network server screens out the terminals that have not responded in the last polling and the newly added terminals , send network access multicast messages to these terminals; before the terminal is connected to the network, it listens to the network access multicast message at the agreed network access multicast frequency. After receiving the network access multicast message, it detects whether it has its own address. If not, it continues to listen. If so, it responds and uses the multicast downlink frequency and spreading factor in the network access multicast message to start listening to the service polling multicast message; after receiving the service polling multicast message, the terminal will only report data if the multicast message contains its own address. When reporting, the working frequency in the frequency group is determined according to the order of the terminal's own address in the address list.

[0032] Specifically, during the implementation process, the LoRaWAN frequency band range is first divided. According to the definition of CN470-510 segment in LoRaWAN protocol specification 1.0.1, there are 96 uplink channels and 48 downlink channels. According to the characteristics of LoRaWAN gateway modules SX1301 and SX1255, the gateway can only receive 8 frequencies at the same time. Therefore, in the embodiment of the present invention, the uplink channel frequency band range is divided into 12 frequency bands with 8 frequencies as a segment. Figure 1 shown.

[0033] The embodiment of the present invention divides the frequency band based on the gateway. Each gateway works in a management domain. The receiving frequency band of each management domain is different. A management domain can only have one receiving frequency band. Gateways in different management domains need to be configured with different working frequency bands. That is, there are a maximum of 12 gateways within the wireless coverage influence range of the same deployment space area. Assuming the first gateway as an example, it works in the first frequency band, and the uplink frequencies it monitors are 470.3MHz, 470.5MHz, 470.7MHz, 470.9MHz, 471.1MHz, 471.3MHz, 471.5MHz, 471.7MHz, and so on. The gateway working in the second frequency band has a receiving frequency of 8 frequencies starting from 471.9.

[0034] Then when adding the terminal to the network server, bind the terminal address (DevAddr) to the gateway, and the data of the specified terminal must be received by a specified gateway.

[0035] In the embodiment of the present invention, multiple network-entry multicast frequency bands may be specified, such as specifying three network-entry multicast frequency points, 503.5 MHz, 505.5 MHz, and 507 MHz.

[0036] Before the terminal accesses the network, it randomly selects one of the three designated frequencies to monitor. After receiving the access packet on the monitored frequency, it detects whether there is its own address in the access data. If there is no address, the access packet is discarded; if there is, it needs to respond to the access packet. If the access packet is not received for a long time, the terminal needs to have a mechanism for automatic frequency switching and reselect the monitoring frequency. After sending the access response packet, the terminal uses the multicast downlink frequency and spreading factor (SF) in the access multicast message to start listening to the service polling multicast.

[0037] The network server groups the terminals bound to the gateway, and in this embodiment, each 8 terminals form a group. Multicast messages are sent to the terminals in these groups in turn. The network server sends service polling multicast messages and network access multicast messages through the gateway. Before sending service polling multicast messages in each round, the network server screens out the terminals that did not respond in the last polling and the newly added terminals, and sends network access multicast messages to these terminals.

[0038] The content of the incoming multicast message includes the multicast downlink frequency used by the gateway, the multicast downlink SF, the working frequency group of the gateway, and the address data of the 8 terminals. The incoming multicast message is sent using three agreed incoming multicast frequencies, and each frequency is sent once.

[0039] The format of the incoming multicast message is: Payload (35 Byte), message header (3 Byte), terminal address list (32 Byte). Payload is the information load, the message header is defined as shown in Table 1, and the terminal address list is defined as shown in Table 2.

[0040] Table 1 Message header definition

[0041]

[0042] Table 2 Terminal address list definition

[0043] DevAddr1 DevAddr2 …… DevAddr7 DevAddr8 4Byte 4Byte …… 4Byte 4Byte

[0044] When the terminal address list is less than 8, the latter part is not filled, and the list part must be an integer multiple of 4. The network access response message uses the FPort port number 251 as the identifier, the data can be empty, and the network server does not perform detection. For the three agreed frequency points, the Ch channel number can correspond to 0-2.

[0045] like Figure 2 As shown in the figure, after the terminal is normally connected to the network, the network server requires the terminal sensor data to be reported by the multicast polling mechanism. The network server sends a service polling multicast sending request to the gateway, and the gateway sends a service polling multicast message. All terminals bound to this gateway will receive the service polling multicast message information; after receiving the service polling multicast message, the terminal determines whether to discard or report the data according to the information in the message specification. The terminal sensor works in Class C mode, which is suitable for terminal sensors that can provide continuous power supply.

[0046] The service polling multicast message follows the LoRaWAN specification, the difference lies in the conventions of DevAddr and Payload. The DevAddr content in the multicast message is filled with FFFFFFFF.

[0047] The payload part of the service polling multicast message is the response terminal address table, which occupies a total of 33 bytes (the specific definition is shown in Table 3). Each terminal address indicates that the responding terminal matches the reporting data sending frequency to be used in the working frequency band according to the order in the address table.

[0048] Table 3 Definition of service polling multicast message

[0049] Version Ch DevAddr1 DevAddr2 …… DevAddr7 DevAddr8 4Bit 4Bit 4Byte 4Byte …… 4Byte 4Byte

[0050] Among them, in the version of the Version content, multicast message version 2 is represented by value 2, occupying the high 4 bits of the first byte

[0051] Ch channel number is from 0 to 15, occupying the lower 4 bits of the first byte, indicating the frequency group used by the terminal to report the frequency. In this example, there are 96 uplink channels, 8 in a group, up to 12, and the numbers can correspond to 0-12. The terminal determines its own reporting frequency in turn according to the frequency order of this frequency group. DevAddr1~8 are the addresses of the terminals, each occupying 4 bytes, represented in big endian.

[0052] Assuming that the gateway operates in the first frequency band, the frequency range is 470.3 MHz to 471.7 MHz, then the reporting corresponding frequencies used by the terminals receiving the service polling multicast message are shown in Table 4.

[0053] Table 4 Service polling multicast message terminal address corresponding frequency point

[0054] Terminal Address Uplink frequency DevAddr1 470.3 DevAddr2 470.5 DevAddr3 470.7 DevAddr4 470.9 DevAddr5 471.1 DevAddr6 471.3 DevAddr7 471.5 DevAddr8 471.7

[0055] If there are only three terminals when sending the service polling multicast message, then only three terminal addresses, DevAddr1, DevAddr2, and DevAddr3, are required in the address segment.

[0056] In this embodiment, the polling multicast message sent by the network server through the gateway each time contains 8 terminal addresses. After receiving the multicast message, the terminal will report data only when the multicast message contains its own address, and its own address is automatically matched to the working frequency point in the working frequency band according to the order in the address table. Under this scheme, 8 terminals use different frequency points in a frequency band to report data at the same time, ensuring that there will be no situation where multiple terminals report at the same frequency point at the same time; on the gateway, after sending the polling multicast message, a 5-second reception waiting time will be left. During this period of time, there will be no situation of sending messages, and the situation where the terminal is not received when reporting data can also be solved. The polling scheduling method of the present invention effectively solves the problem of packet loss caused by conflicts and collisions caused by asynchronous transmission of gateways and terminals.

[0057] In actual applications, for a gateway that can receive 8 frequencies at the same time, the network server sends a set of service polling messages every 5 seconds to collect data from 8 terminals. 96 terminal data can be collected every minute, and 5760 terminal data can be collected every hour (8×statistical duration / collection interval=8×3600 / 5=5760). In a traditional LoRaWAN network, no matter which working mode is used, if each terminal reports data, the packet loss rate will reach more than 50% to collect 5760 terminal data within an hour, which will be catastrophic.

[0058] Based on the same inventive concept, a LoRaWAN network polling scheduling system disclosed in an embodiment of the present invention includes a network server, a gateway and a terminal; wherein the network server is used to bind the terminal address to the gateway, the data of the designated terminal can only be received by a designated gateway, and different terminals bound to the same gateway use different working frequencies; and the gateway sends a service polling multicast message and a network access multicast message, and before each round of sending the service polling multicast message, the terminals that did not respond in the last polling and the newly added terminals are screened out, and the network access multicast messages are sent to these terminals; the network access multicast message includes the multicast downlink frequency and spreading factor used by the gateway, the working frequency of the gateway, and the working frequency of the gateway. The frequency group and terminal address list are made, the terminal address in the service polling multicast message is an agreed fixed value, and the information load content is the response terminal address table; before the terminal is connected to the network, it listens to the network multicast message at the agreed network multicast frequency point, and after receiving the network multicast message, it detects whether it has its own address, if not, it continues to listen, if so, it responds and uses the multicast downlink frequency point and spreading factor in the network multicast message to start listening to the service polling multicast message; after receiving the service polling multicast message, the terminal will only report data if the multicast message contains its own address, and when reporting, the working frequency point in the frequency group is determined according to the order of the terminal's own address in the address list. For specific implementation details, please refer to the above method embodiment, which will not be repeated here.

Claims

1. A LoRaWAN network polling scheduling method, characterized in that: The steps include: The frequency band range of the LoRaWAN uplink channel is divided into gateways. Each gateway works in one management domain, and different management domains have different working frequency bands. Bind the terminal address to the gateway in the network server. The data of a specified terminal can only be received by a specified gateway. Different terminals bound to the same gateway use different working frequencies. Before the terminal is connected to the network, it monitors the network multicast message at the agreed network multicast frequency. After receiving the network multicast message, it checks whether it has its own address. If not, it continues to monitor. If so, it responds and uses the multicast downlink frequency and spreading factor in the network multicast message to start listening to the service polling multicast message. The network server sends service polling multicast messages and network access multicast messages through the gateway. Before sending service polling multicast messages in each round, the network server selects the terminals that did not respond in the last polling and the newly added terminals, and sends network access multicast messages to these terminals; the network access multicast message includes the multicast downlink frequency and spreading factor used by the gateway, the working frequency group of the gateway and the terminal address list. The terminal address in the service polling multicast message is an agreed fixed value, and the information load content in the service polling multicast message is the response terminal address table; After receiving the service polling multicast message, the terminal will report data only when its own address is contained in the multicast message. When reporting, the working frequency in the frequency group is determined according to the order of the terminal's own address in the address list.

2. The LoRaWAN network polling scheduling method according to claim 1, characterized in that: The frequency band range of the LoRaWAN uplink channel is divided into 12 frequency bands, with 8 frequency points as a section. The network server divides 8 terminals into a group, and each gateway is bound to a maximum of 8 terminals.

3. The LoRaWAN network polling scheduling method according to claim 1, characterized in that: The working mode of the terminal is Class C mode.

4. The LoRaWAN network polling scheduling method according to claim 1, characterized in that: Three multicast frequencies for network access are agreed upon in the LoRaWAN downlink channel frequency band. The terminal has an automatic frequency switching mechanism. If the multicast message for network access is not heard at one of the frequencies, the frequency is switched and heard again.

5. The LoRaWAN network polling scheduling method according to claim 1, characterized in that: The message header of the incoming multicast message consists of a version, a frequency group number, a multicast downlink frequency, a spreading factor and a reserved field.

6. The LoRaWAN network polling scheduling method according to claim 1, characterized in that: The response terminal address table in the service polling multicast message is composed of a version, a frequency group number and a terminal address list bound to a gateway.

7. The LoRaWAN network polling scheduling method according to claim 1, characterized in that: After each gateway sends the service polling multicast message, it will leave a reception waiting time. During the waiting time, no message will be sent to the terminal.

8. A LoRaWAN network polling scheduling system, comprising a network server, a gateway and a terminal, characterized in that: The frequency band range of the LoRaWAN uplink channel is divided into units of gateways, each gateway works in a management domain, and the working frequency bands of different management domains are different; the network server is used to bind the terminal address to the gateway, the data of the designated terminal can only be received by a designated gateway, and different terminals bound to the same gateway use different working frequencies; and the gateway sends a service polling multicast message and a network access multicast message. Before each round of sending the service polling multicast message, the terminals that did not respond in the last polling and the newly added terminals are screened out, and the network access multicast message is sent to these terminals; the network access multicast message includes the multicast downlink frequency and spreading factor used by the gateway, the working frequency group of the gateway and the terminal address list, the terminal address in the service polling multicast message is an agreed fixed value, and the information load content in the service polling multicast message is the response terminal address table; Before accessing the network, the terminal monitors the network access multicast message at the agreed network access multicast frequency. After receiving the network access multicast message, it detects whether it has its own address. If not, it continues to monitor. If so, it responds and uses the multicast downlink frequency and spreading factor in the network access multicast message to start listening to the service polling multicast message. After receiving the service polling multicast message, the terminal will report data only if the multicast message contains its own address. When reporting, the working frequency in the frequency group is determined according to the order of the terminal's own address in the address list.

9. The LoRaWAN network polling scheduling system according to claim 8, characterized in that: The message header of the network-entry multicast message consists of version, frequency group number, multicast downlink frequency, spreading factor and reserved field; the response terminal address table in the service polling multicast message consists of version, frequency group number and a terminal address list bound to the gateway.

10. The LoRaWAN network polling scheduling system according to claim 8, characterized in that: After each gateway sends the service polling multicast message, it will leave a reception waiting time. During the waiting time, no message will be sent to the terminal.

Images