Source: RFID World
1. Introduction
The purpose of this article is to provide an introductory technical overview of LoRa and LoRaWAN. Low Power Wide Area Networks (LPWAN) are expected to support a significant portion of the billions of devices predicted for the Internet of Things (IoT). LoRaWAN is designed from the ground up, optimized for battery life, capacity, range, and cost. An overview of the LoRaWAN specifications for different regions is provided, along with a high-level comparison of various technologies competing in the LPWAN space. (Content has been edited and shortened compared to the original.)
2. What is LoRa?
LoRa is a physical layer or wireless modulation used to establish long-range communication links. Many traditional wireless systems use Frequency Shift Keying (FSK) modulation as the physical layer because it is a very effective modulation for achieving low power consumption. LoRa is based on linear frequency spread spectrum modulation, which retains the low power characteristics similar to FSK modulation but significantly increases communication range. Linear spread spectrum has been used in military and space communications for decades due to its ability to achieve long communication distances and robustness against interference; however, LoRa is the first low-cost implementation for commercial use.
Long Range (LoRa)
The advantage of LoRa lies in its long-range capabilities. A single gateway or base station can cover an entire city or hundreds of square kilometers. At a given location, the distance largely depends on the environment or obstacles, but LoRa and LoRaWAN have a link budget superior to any other standardized communication technology. The link budget, usually expressed in decibels (dB), is the primary factor determining distance in a given environment. Below is a coverage map of the Proximus network deployed in Belgium. With minimal infrastructure implementation, it can easily cover the entire country.
3. Where is LPWAN Suitable?
No single technology can meet all IoT project applications and volumes. WiFi and BTLE are widely adopted standards that work well for applications related to personal device communication. Cellular technology is well-suited for applications requiring higher data throughput and power supply. LPWAN offers years of battery life and is designed for sensors and applications that need to send small amounts of data a few times an hour over long distances from various environments.
Key Factors in LPWAN:
Network architecture
Communication distance
Battery life or low power consumption
Robustness against interference
Network capacity (maximum number of nodes in the network)
Network security
Unidirectional vs. bidirectional communication
Various service applications
4. What is LoRaWAN?
LoRaWAN defines the communication protocol and system architecture of the network, while the LoRa physical layer enables long-range communication links. The protocol and network architecture have the greatest impact on node battery life, network capacity, quality of service, security, and various application service qualities of the network.
Network Architecture
Many existing deployed networks use a mesh network architecture. In a mesh network, individual terminal nodes forward information from other nodes to increase the communication distance and size of the network area. While this increases range, it also adds complexity, reduces network capacity, and decreases battery life, as nodes may receive and forward information from other nodes that may not be relevant to them. When establishing long-distance connections, a long-distance star architecture is most meaningful as it protects battery life.
In a LoRaWAN network, nodes are not associated with dedicated gateways. Instead, the data transmitted by a node is typically received by multiple gateways. Each gateway forwards the data packets received from terminal nodes to a cloud-based network server via some backhaul (cellular, Ethernet, satellite, or Wi-Fi). Intelligence and complexity are placed on the server, which manages the network, filters redundant received data, performs security checks, schedules confirmations through the optimal gateway, and executes adaptive data rates. If a node is mobile or moving, there is no need to switch from gateway to gateway, which is an important feature applicable to asset tracking—a major target vertical application of IoT.
Battery Life
Nodes in a LoRaWAN network communicate asynchronously, sending data when it is ready, whether event-driven or time-scheduled. This type of protocol is often referred to as the Aloha method. In mesh or synchronous networks, such as cellular, nodes must frequently wake up to synchronize with the network and check for messages. This synchronization consumes energy significantly and is the primary driver of reduced battery life. In a recent study, GSMA compared different technologies in the LPWAN space, finding that LoRaWAN has a 3 to 5 times advantage over other technology choices.
Network Capacity
To enable long-distance star networks, gateways must have very high capacity or performance to receive messages from a large number of nodes. High network capacity is achieved through adaptive data rates and multi-channel multi-modulation transceivers in the gateway, allowing messages to be received simultaneously on multiple channels. Key factors affecting capacity include the number of concurrent channels, data rate (air time), payload length, and how often nodes send data. Because LoRa is based on spread spectrum modulation, when using different spreading factors, the signals are actually orthogonal to each other. When the spreading factor changes, the effective data rate also changes. Gateways take advantage of this feature, being able to receive multiple different data rates on the same channel at the same time. If a node has a good connection and is close to the gateway, there is no reason to always use the lowest data rate, as filling the available spectrum takes longer than it needs. The higher the data transmission rate, the shorter the time in the air, allowing more potential space for other nodes to transmit data. Adaptive data rates also optimize the battery life of nodes. For adaptive data rates to work, symmetric uplink and downlink require sufficient downlink capacity. These characteristics give LoRaWAN very high capacity, making the network more scalable. With minimal infrastructure, networks can be deployed, and when capacity is needed, more gateways can be added, data rates can be adjusted, and crosstalk can be reduced, scaling the network capacity by 6 to 8 times. Other LPWAN technologies do not have the scalability of LoRaWAN due to technical trade-offs that limit downlink capacity, making downlink distances asymmetrical to uplink distances.
Device Classes – Not All Nodes Are the Same
Terminal devices serve different applications and have different requirements. To optimize various terminal application specifications, LoRaWAN uses different device classes. Device classes balance the network’s downlink communication latency with battery life. In control or actuator-type applications, downlink communication latency is a critical factor.
Bidirectional terminal devices (Class A): Class A terminal devices allow bidirectional communication, so each terminal device’s uplink transmission is followed by two short downlink receive windows. The transmission time slots are scheduled by the terminal device based on its own communication needs with a slight variation based on a random time base (ALOHA-type protocol). For applications where the terminal device only needs to communicate briefly with the server after sending an uplink transmission, this Class A operation is the lowest power terminal system. Any other downlink communication from the server must wait for the next scheduled uplink.
Bidirectional terminal devices with scheduled receive time slots (Class B): In addition to the random receive windows of Class A, Class B devices open additional receive windows at scheduled times. To enable the terminal device to open its receive window at scheduled times, it synchronizes with a gateway beacon once. This allows the server to know when the terminal device is listening.
Bidirectional terminal devices with maximum receive time slots (Class C): Class C terminal devices have their receive windows almost continuously open, only closing when transmitting.
Security
Incorporating security is extremely important for any LPWAN. LoRaWAN uses two layers of security: one for network layer security and another for application layer security. Network security ensures the reliability of network nodes, while application layer security ensures that network operators cannot access end-user application data. Key exchange uses AES encryption with IEEE EUI64 identifiers. Each technology choice has trade-offs, but LoRaWAN’s characteristics in network architecture, device classes, security, capacity scalability, and mobile optimization meet a wide range of potential IoT applications.
Nanjing Renjue Intelligent is a company specializing in high-cost performance LoRa products, currently promoting lora1278, 1276, 1262, 1268 module products, 1301 base station serial module, and pure RF module.
5. Overview of LoRaWAN Regions
The LoRaWAN specifications vary slightly based on different regional spectrum allocations and regulatory requirements. Europe and North America have established LoRaWAN specifications, while other regions are still being developed by technical committees. Joining the LoRa Alliance as a contributor member and participating in technical committees offers significant advantages for companies targeting solutions in the Asian market.
Introduction to LoRa WAN Protocol
In addition to metropolitan area networks of 2G/3G/4G, wireless technologies in IoT applications include various communication technologies for local area networks and short distances, such as WiFi in the 2.4G frequency band, Bluetooth, Zigbee, and Sub-Ghz, etc. These short-range wireless technologies have very clear advantages and disadvantages. Moreover, from the perspective of wireless application development and engineering operations personnel, there has always been a dilemma: designers can only choose one between longer distances and lower power consumption. However, with the adoption of LoRa technology, designers can now achieve both, maximizing long-distance communication and low power consumption while also saving on additional repeater costs.
Generally speaking, transmission rate, operating frequency band, and network topology are the three main parameters affecting the characteristics of sensor networks. The choice of transmission rate will determine the transmission distance and battery life of the system. The choice of operating frequency band must consider trade-offs between the frequency band and the system’s design goals. In FSK systems, the choice of network topology is determined by the transmission distance requirements and the number of nodes needed by the system. Semtech’s new cost-effective transceiver solution using new spread spectrum technology will change the previous trade-off considerations, providing users with a simple system that can achieve long distances, long battery life, and increased system capacity, thus leading to the emergence of LoRa technology.
LoRa integrates digital spread spectrum, digital signal processing, and forward error correction coding technologies, achieving unprecedented performance. Previously, only high-grade industrial radio communications would integrate these technologies, but with the introduction of LoRa, the landscape of embedded wireless communications has changed dramatically.
Forward error correction coding technology adds some redundant information to the data sequence to be transmitted, so that error code elements injected during the data transmission process can be corrected in a timely manner at the receiving end. This technology reduces the need to create “self-repairing” data packets for retransmission and performs well in addressing burst errors caused by multipath fading.
Once the data packet is established and forward error correction coding is injected to ensure reliability, these packets are sent to the digital spread spectrum modulator. This modulator feeds each bit of the packet data into an “expander,” dividing each bit time into numerous chips. The LoRa modulator can be configured to range from 64 to 4096 chips/bit. AngelBlocks configured modulator can use the maximum spreading factor of 4096 chips/bit (12). In contrast, ZigBee can only divide in the range of 10-12 chips/bit.
By using a high spreading factor, LoRa technology can transmit small capacity data over a wide range of radio frequency spectrum. In fact, when measured with a spectrum analyzer, this data appears as noise, but the difference is that noise is uncorrelated, while data is correlated, based on which the data can actually be extracted from the noise. In fact, the higher the spreading factor, the more data can be extracted from the noise.
In a well-functioning GFSK receiver, an 8dB minimum signal-to-noise ratio (SNR) is required to reliably demodulate the signal. Using the AngelBlocks configuration, LoRa can demodulate a signal with an SNR of -20dB, while the GFSK method differs from this result by 28dB, which corresponds to a significant increase in range and distance. In outdoor environments, a 6dB difference can achieve twice the original transmission distance.
To effectively compare the performance of transmission ranges between different technologies, we use a quantitative metric called “link budget.” The link budget includes every variable affecting the received signal strength at the receiving end, in its simplified system including the transmission power plus the receiver sensitivity.
AngelBlocks’ transmission power is 100mW (20dBm), and the receiver sensitivity is -129dBm, resulting in a total link budget of 149dB. In comparison, GFSK wireless technology with a sensitivity of -110dBm (which is already excellent data) requires 5W of power (37dBm) to achieve the same link budget value. In practice, most GFSK wireless technology receivers can achieve a sensitivity of -103dBm, under which the transmitter must operate at a frequency of 46dBm or about 36W to achieve a similar link budget value as LoRa.
Therefore, using LoRa technology, we can achieve a wider transmission range and distance with low transmission power, which is precisely the low-power wide-area technology we need.
Transmission Rate and Distance
The transmission rate is a key variable in system design, determining many attributes of the overall system performance. The wireless transmission distance is determined by the receiver sensitivity and transmitter output power, and the difference between the two is referred to as the link budget. Output power is limited by standard specifications, so the only way to increase distance is to improve sensitivity, which is significantly affected by data rate. For all modulation methods, the lower the rate, the narrower the receiver bandwidth, and the higher the receiver sensitivity. Among the most widely used modulation methods in today’s cost-effective wireless transceivers are FSK or GFSK. To further reduce the receiver bandwidth of FSK systems, the only feasible method is to improve the accuracy of the reference crystal. Under equivalent data rate conditions, commercially available low-cost spread spectrum modulation methods can achieve 8-10dB higher sensitivity than traditional FSK modulation methods. Semtech will launch a new transceiver that integrates a spread spectrum modulation method called LoRa and traditional GFSK modulation. The graph shows the relationship curve of sensitivity relative to data rate in GFSK modulation and LoRa spread spectrum modulation systems.
Compared to FSK systems, this new spread spectrum method using the same low-cost crystal improves sensitivity by 30dB, theoretically equivalent to a 5-fold increase in transmission distance.
Network Architecture and Protocol
Choosing between star and mesh networks is a key factor affecting the performance of the entire wireless network system. Star networks are the simplest network structure with the lowest latency. Long-distance, co-channel synchronous transmission, improved co-channel suppression, and high selectivity are advantages of these spread spectrum methods, providing a high-performance system solution for sensor networks that traditional FSK modulation cannot achieve.
The advantages of spread spectrum modulation at the same rate can easily be used to improve the performance of existing mesh networks, while star networks will achieve optimal system performance. Utilizing star networks in suburban environments can achieve distances of 8-16km, eliminating the need for large mesh network structures to cover such wide areas.
A multi-channel, multi-modulation concentrator can adapt to different rates and powers of different nodes, thus maximizing network capacity and extending battery life. By using different spreading factors, the transmission rate of the spread spectrum system can be changed. The variable spreading factor increases the overall system capacity of the network, as signals using different spreading factors can coexist on a single channel. With network simulation tools, we can easily observe that, compared to traditional fixed-rate FSK systems, star networks using the above technologies can easily gain many advantages, such as 95% of nodes only occupying 10% of total energy consumption, while the overall system capacity will also increase by 5-6 times.
In summary, compared to other wireless systems, LoRa technology has several major advantages. It uses spread spectrum modulation technology, capable of demodulating signals with noise levels below 20 dB. This ensures high sensitivity, reliable network connections, while improving network efficiency and eliminating interference. Compared to mesh networks, the star topology of the LoRaWAN protocol eliminates synchronization overhead and hops, thus reducing power consumption and allowing multiple concurrent applications to run on the network. Additionally, the communication distance achieved by LoRa technology is much longer than that of other wireless protocols, allowing LoRa systems to operate without repeaters, thereby reducing overall ownership costs. Furthermore, compared to 3G and 4G cellular networks, LoRa technology offers greater scalability and cost-effectiveness for embedded applications.
LoRaWAN is a low-power wide-area network specification launched by the LoRa Alliance, which can provide regional, national, or global networks for battery-powered wireless devices. LoRaWAN targets some core needs in IoT, such as secure bidirectional communication, mobility, and local services. This technology allows smart devices to achieve seamless interoperability without complex local configurations, granting users, developers, and enterprises in the IoT field the freedom to operate.
The LoRaWAN network architecture is a typical star topology, where the LoRa gateway acts as a transparent relay connecting front-end terminal devices and back-end central servers. The gateway connects to the server via standard IP, while terminal devices communicate with one or more gateways in a single hop, and all nodes are bidirectional.
Communication between terminals and gateways is completed based on different frequencies and data transmission rates, with the choice of data rate requiring a trade-off between transmission distance and message latency. Due to the use of spread spectrum technology, communication at different data transmission rates does not interfere with each other and creates a set of “virtualized” frequency bands to increase gateway capacity. The data transmission rate range of LoRaWAN networks is from 0.3 kbps to 50 kbps. To maximize terminal device battery life and overall network capacity, the LoRaWAN network server controls the data transmission rate and RF output of each terminal device through an adaptive data rate (ADR) scheme. Networks covering the entire country suitable for IoT need to address security communication issues such as critical infrastructure, confidential personal data, or social public services, which are generally solved using multi-layer encryption:
Unique network key (EU164) ensuring network layer security;
Unique application key (EU164) ensuring end-to-end security at the application layer.
Nanjing Renjue Intelligent is a company specializing in high-cost performance LoRa products, currently promoting lora1278, 1276, 1262, 1268 module products, 1301 base station serial module, and pure RF module. For finished base station needs, please contact Engineer Yu: 15651028736.
Device Specific Keys (EUI128) LoRaWAN network nodes have multi-level security schemes to ensure different needs of various applications:
Class A
Bidirectional communication terminal devices (Class A): Class A terminal devices allow bidirectional communication, with each terminal device’s uplink transmission accompanied by two downlink receive windows. The transmission slots of terminal devices are based on their own communication needs, with slight adjustments based on a random time base (ALOHA protocol). In Class A terminal device applications, power consumption is the lowest, as downlink communication with the server can only occur after the terminal sends an uplink transmission, and any downlink communication from the server can only occur after the uplink communication.
Class B
Bidirectional communication terminal devices with preset receive slots (Class B): Class B terminal devices open additional receive windows at preset times. To achieve this, terminal devices synchronize with a gateway to receive a Beacon, allowing the server to know that the terminal device is “listening.”
Class C
Bidirectional communication terminal devices with maximum receive slots (Class C): Class C terminal devices have their receive windows almost continuously open, only closing when transmitting.
