How to Improve LoRaWAN® Gateway Performance

What are the best practices to improve the performance of LoRaWAN gateways? LoRa communication protocol depends on many Spreading Factors (SFs), data rates, coding rate, propagation conditions, and LoRaWAN specification. So, when considering the performance gateway can ensure, it is important to take into account transmission power, data transfer rate, spreading factor, duty cycle, channel occupancy rate, and distance between the gateways. Understanding these factors and optimizing them can help you get the best performance possible from your LoRa network. Listed below are some helpful tips to improve your LoRaWAN gateway performance.

Spreading Factor

One of the ways to improve LoRaWAN Gateway and network performance is to adjust the Spreading Factors (SFs) allocated to the Gateway. The spreading factors are critical to the capacity of the LoRa cell, as these are a factor of capture effects. The higher the spreading factors, the farther away the user is from the gateway, and the lower the spreading factors, the closer they are to the gateway.

In order to optimize the PRP (Packet Reception Probability) of LoRaWAN devices, the spreading factor is allocated to each of the EDs (End Devices) in a chirp spread spectrum. The ToA (Time on Air) of each end devices is different, which means that adjusting the Spreading Factors can enhance LoRaWAN network performance. Spreading factors allocation has a number of characteristics and can be optimized to decrease energy consumption. This communications technique is based on an evenly distributed genetic ADR algorithm.

The spreading factor is the ratio of chip and symbol rates. A high spreading factor improves the Signal to Noise Ratio (SNR) and increases the airtime of a packet. The spreading factor can be anywhere from 6 to 12 in different configurations. The lower the spreading factor, the higher the chirp rate. In addition to affecting SNR (Signal to Noise Ratio), the higher the spreading factor improves the sensitivity of the receiving antenna. If you want to improve the performance of your LoRaWAN Gateway, you should consider the spreading factor used in the system. The spreading factor determines how robust the signal is to interference and channel conditions. The higher the spreading factor, the lower the error rate will be during decoding. The spreading factor is a linear variation in frequency, which decreases as signal strength increases. Increasing the spreading factor allows the gateway to extend the range of transmission and LoRa communication.

Duty Cycle

The duty cycle is another factor to optimize LoRaWAN Gateway performance. By changing the duty cycle of each LoRa network node, the frequency and data transfer rate are optimized. Spreading factors also affect the number of confirmed frames sent by the gateway. The more frames a node can send per second, the better. The higher the duty cycle, the higher the amount of time it takes to complete the LoRa communication.

If you want to boost the performance of your LoRaWAN Gateway, you need to change the duty cycle parameter. This parameter limits the number of communications in transmitted packets that can occur in a single hour. In the EU 863-870 MHz band, a duty cycle of 1% per channel must be used. By modifying the duty cycle parameter, you can boost the performance of your LoRaWAN technology gateway by up to 50%.

The number of channels must be carefully designed to minimize the probability of packet collisions. Additionally, the number of channels must be tightly coupled with the number of end devices. The number of channels must be large enough to offer alternate communication channels for network nodes that collide with each other. This way, the duty cycle does not have a huge impact on the latency. The number of communication channels, in turn, is increasingly important for your LoRaWAN Gateway to work effectively.

A fixed duty cycle limits the number of transmissions in the downlink. The off-period immediately following transmissions also reduces the capacity of the LoRaWAN network. Using a fixed duty cycle will reduce the number of acknowledged frames, but may not be sufficient for ultra-reliable services. The duty-cycle regulation also limits the number of acknowledgeable frames, which can degrade performance.

To maximize the performance of your LoRaWAN gateway networks, you should definitely consider adjusting the duty cycle to accommodate the number of end devices and their data needs. A well-dimensioned LoRaWAN gateway can serve thousands of end devices, but it must still have the flexibility to meet the requirements of the different applications.

Channel Occupancy Rate

There are several ways to improve LoRaWAN Gateway performance, including reducing channel overlapping and optimizing channel occupancy rate. Increasing channel occupancy rate will ensure better performance in a dense deployment of Internet of Things (IoT) devices. The Computer-Aided Design (CAD) functionality provided by the LoRa transducer allows it to detect the LoRa preamble signal and match the configuration. A new method is defined in a recent paper by Kim et al., which involves selecting an adaptive spreading factor between two single-channel LoRa modems [2]. This method allocates pairs of transmitters to spreading factors in a manner that reduces collisions along with the multihop topology.

The authors propose to classify the environment before deployment. The current link quality estimation mechanism is not adaptive and does not consider the loss caused by collisions with other LoRa devices. It also fails to consider the effects of collisions with other LoRa networks. By increasing the channel occupancy rate, the number of end nodes that can communicate with each other will increase. This strategy can significantly increase LoRaWAN Gateway performance.

An increase in channel occupancy rate can increase network performance by up to 15%. In addition, it can improve the quality of service provided by a gateway. Increasing the channel occupancy rate on LoRaWAN gateways is a key feature that will increase user satisfaction and drive down operational costs.

Distance Between Gateways

A distance between two LoRaWAN gateways also significantly influences performance. While a high number of gateways can increase overall performance, a low number of gateways can reduce overall performance. As long as the distance between two gateways is within the network’s radius, the performance of the Internet of Things network will improve. The distance between gateways can be increased by placing the gateways at a closer distance, but this can only be beneficial for LoRaWAN networks that have a large number of LoRa nodes.

The longer the distance between two LoRaWAN gateways, the higher the sensitivity of the adaptive data rate packets. The farther the gateways are, the more collected data packets are subject to spreading factor values. This allows more LoRa nodes to transmit with a lower spreading factor, reducing interference and fast fading. But there is a limit on the distance, which radio signals can reach, and within which can transmit data.

The LoRaWAN protocol defines the maximum number of nodes and the maximum number of collisions that can occur before the LoRa network begins to experience problems with radio waves, chip rate, symbol rate, LoRa modulation, transmission, and communication range. This allows you to estimate the capacity of your communications channel while maintaining the high performance of LoRa technology. You should also consider the distance between LoRa technology gateways and their duty cycles, which will improve their overall performance. The distance between the LoRaWAN gateways should not be more than 200 m, as otherwise, it can make the network server transmission at large distances unstable.

A number of studies have analyzed the scalability of the LoRa network. The Georgiou et al. reference provides a mathematical model of a single gateway network, which includes the effects of demodulator channel saturation and a second gateway [1]. The resulting LoRa networks performance is higher than a single gateway network with only a few nodes. However, the distance between gateways can increase the number of nodes in a network.

Processing Gain

This increase in processing gain should be implemented in the gateway to increase its overall long-range throughput. The maximum number of LoRa technology nodes is close to 1000. The use of an in-band-full-duplex LoRa technology can minimize the control overhead of the distributed queueing (DQ). Depending on the configuration, the processing gain can be as much as three times the total throughput. The MAC (Medium Access Control) protocol is another factor to consider when improving LoRaWAN Gateway performance.

LoRaWAN networks can support millions of messages a day per each. The number of messages supported by a gateway depends on how many monitoring devices and Internet of Things (IoT) applications are on it. For example, a single eight-channel gateway may only support a few hundred thousand messages in a 24-hour period. A network with ten or more gateways can support one million messages in one day. This is not the limit for LoRaWAN networks, but rather the limit of what a single gateway can support.

As mentioned, the spreading factor affects the communication performance of LoRa networks. A large spreading factor results in more time on the air. Larger spreading factor improves link budget but reduces sensitivity. A smaller spreading factor improves the transmission range and increases the number of messages. However, the spreading factor should correspond to the modulation method used by the gateway. If not, the gateway will not be able to receive the data value at all.

Possible Issues Connected to LoRaWAN Networks

In-band and out-of-band RF (Radio Frequency) Interference is one of the most common LoRaWAN® issues that result in poor and unreliable LoRaWAN® coverage, reduced LoRa networks availability and dependability, and the need for re-transmission of packets (resulting in lower device battery life). These interfering RF signals are very common and are generated by nearby deployed 2G to 5G cellular base stations, FM (Frequency Modulation) transmitters, TV signals, GPS re-transmission stations, and fixed and mobile wireless transmitters. Poorly designed LoRa communications and LoRaWAN® gateways with insufficient or no rejection of these interference signals at their antenna ports are impacted, resulting in poor performance.

Increased network load and degraded goodput can degrade the performance of LoRaWAN networks. To overcome these issues, the LoRaWAN communication protocol will need to include new functionalities such as omnidirectional antennas, increased transmission data rate, antenna gain, low power consumption, and long-range radio waves. The Internet of Things application server is responsible for handling sensor application data and creating the application-layer downlink payloads for the connected IoT applications and devices. In addition, a joining server will manage the over-the-air activation process for LoRa technology devices.

There are many LoRaWAN® gateways deployed globally that provide no Rx selectivity and out-of-band rejection, resulting in degraded LoRaWAN® performance, reduced coverage (in many cases operators and enterprises have reported reduced coverage from 5-10 km to less than 1 km), increased Rx (Receive) noise level, reduced Rx SNR (Signal-to-Noise Ratio), high-level IM3 (Third-Order Intermodulation) and IM2 (Second-Order Intermodulation) generated within the LoRaWAN® Rx band, cross modulation, and damage (burn up) of the gateway LNAs (Low Noise Amplifier).

Combined, these RF impairments reduce the LoRaWAN® coverage, limit in-building penetration, impact LoRa networks’ reliability, and dependability, degrade the overall LoRa modulation performance to a level that is deemed unreliable for many practical applications (such as building smart cities, monitoring energy consumption, or randomly selected services), and cost the operator significantly more as they need to deploy more LoRaWAN® gateways to cover the same areas with more frequent maintenance and servicing of LoRa networks.

Many assume the poor RF performance is related to the LoRaWAN® standard itself, the real issue stems from poor LoRaWAN® gateway radio designs. The fix is rather simple and inexpensive – implement the same design considerations that the 1G – 4G base stations and most other wireless product developers have been implementing for the last 50 years.

What is Important to Consider Choosing IoT Gateway?

What are the most important considerations for selecting an Internet of Things (IoT) gateway? This section presents the importance of data storage, processing power consumption, and network security. The main purpose of the gateway is to ensure smooth long-range transmission of data through communication networks using a particular signal.

Data Storage

There are many factors to consider when selecting an Internet of Things gateway, including data storage. In some cases, a single gateway can be sufficient, while others may need more. The number of sensors required may vary as well, with some systems requiring as many as 10000 sensors, which take 30 readings per second. Before making a decision, it is important to evaluate the environment in which the IoT solution will be deployed. For example, a telecom company may need its sensors to be installed on top of a communication tower. The Internet of Things gateway needs to be installed at the highest point possible, so it should be able to handle high temperatures, while a carmaker may need one that is mounted in the room.

The data rate produced by IoT devices is in different formats, including discrete sensor readings, device health metadata, and large images or videos. These formats may be a challenge to store and transmit, and they can also be exposed to harsh environments. In such environments, data storage needs must be able to meet a wide range of requirements.

Another consideration is the amount of data stored by the gateway. A factory gateway, for example, is typically intended for use in an environment with reliable networks, but this doesn’t mean that the data rate will be stored forever. Oftentimes, factory gateways are not equipped with enough storage to store large amounts of data values. To be able to store this data, the gateway should have expandable memory capabilities and slots for additional micro-SD cards.

Processing Power

There are a few factors to consider when choosing gateways. Depending on the final objective and functionality needed, the ideal values will vary. For example, you may have a single sensor, while another might have ten thousand. You should consider the processing power of the gateway, as some may require up to thirty readings per second, which is not possible to perform with a low-power gateway. You may also need several gateways so that you can alternate transmission values (sending and receiving information).

Another factor to consider is the amount of legacy equipment your factory uses. Many factories still rely on older equipment, and obtained results from different studies show that overall upgrading it is not practical or economical. That is why it is important to make sure the gateway you choose supports legacy equipment. This will enable you to integrate all the data from multiple machines and monitor their performance. You may also want to consider other features such as energy-saving capabilities and LTE (Long-Term Evolution) support. A gateway with these features should be flexible and easy to use.

When choosing gateways, processing power is an important factor. The higher the processing power, the greater the number of applications it can run. However, more power means more vulnerabilities. Since IoT gateways are the system’s first line of sight, they must be secure, so it is important to choose those gateways which operate on the basis of reliable networks with the encrypted long-range transmission.

If your company relies on industrial data, you should choose gateways with sufficient processing power. The gateway must have enough processing power to process the data that streams in from monitoring devices. For example, you may want to automate irrigation on your farm. To do this, you would need to install multiple sensors to monitor soil moisture, crop yield, and general performance. These sensors, in turn, have to send obtained results using gateways with sufficient antenna gain, low-power, and long-range transmission. Besides, to provide the greatest value, your gateway should have built-in alerts that will notify you when it’s time to make a change.

Security

If you’re planning to integrate IIoT (Industrial Internet of Things) projects with your existing infrastructure, consider buying a gateway with real-time analytics capabilities. Real-time analytics help prevents defects in products. Choose a gateway that offers remote management capabilities for monitoring alerts and reviewing data. In addition, look for models that offer robust security and transmission encryption features. You’ll also want to choose a gateway that supports multiple communication protocols since they can ensure long-range transmission.

IoT gateways should meet security standards that meet your company’s needs. Look for DES (Data Encryption Standard), RSA (Rivest–Shamir–Adleman), AES (Advanced Encryption Standard), or Twofish Encryption Algorithm encryption. Your gateway must also have local storage capabilities. Cloud storage is not recommended. A local device is best. A security feature for IoT gateways includes encryption and authentication capabilities. While choosing an IoT gateway, consider a security rating from the IoT Foundation.

IoT gateways should have built-in encryption, a strong authentication protocol, and a secure remote software update platform. Edge antenna gain, processing, and analytics are also important for a secure data transfer route. Choose a gateway with these features to ensure your IoT data is secure from beginning to end. You should also look for a gateway that supports the latest connectivity protocols and supports mobile devices.

Edge Computing Applications

When choosing an IoT gateway, keep in mind that edge computing applications use processing power at the edge of a network to aggregate, pre-process, and score IoT data. It is imperative to secure these devices so that they can handle sensitive data and support decisions about the physical world. To help with security, edge computing gateways are available with encryption and access-control methods. They can also be equipped with failover management to ensure that data delivery continues even if one of the nodes fails in the line of sight.

Another consideration when choosing an IoT gateway is network throughput. Throughput is the amount of data transmitted per second. This metric is typically measured at the gigabit and megabit levels. Applications often require a certain minimum throughput in order to work properly, so low throughput can affect performance. In addition, high latency limits the functionality of an application. To address this issue, edge computing gateways also provide troubleshooting capabilities. These gateways are typically located outside of a secure data center.

Another consideration when choosing an IoT gateway is the ability to deploy the edge compute. The technology allows deployment in diverse physical environments, including outdoor environments and smelting factories. Edge computing is becoming more prevalent as the IoT becomes more widely adopted. Edge computing applications can reduce latency. The majority of the data coming from sensors is irrelevant to most IoT applications. If the temperature sensor is reporting a reading every second of twenty degrees Celsius, for example, edge computing will allow filtering the data at the sensor location before sending it to the cloud. This reduces network costs, as well as cloud storage and processing costs, increasing informational value.

Communication protocols

Choose an IoT gateway that supports the appropriate communication protocol. Unlike standard communication protocols, IoT protocols follow different architectures and have different capabilities. The choice of the protocol should be based on the type of application and the desired level of security and reliability. If your use case is resource-constrained, choose one with a low data rate and a high level of security. If you need a secure and reliable connection, choose one with the reliable long-range transmission.

Moreover, consider the range of the data that the IoT device can send. BLE (Bluetooth Low Energy) is not suitable for a WAN application. Besides, the range that the device can cover must be compatible with the protocol used for remote communications. In addition, choose a gateway with edge computing, which is an option for transmitting and analyzing data directly from the device. If you’re not sure, consider using a multi-hop gateway process.

What Does Make TEKTELIC Radio Waves Reliable?

The TEKTELIC team has been developing wireless radios and 1G-4G base stations for 25+ years (15 years at Nortel, ALU, and Ericsson amongst other companies, and 12 years at TEKTELIC) for global tier 1 integrators and operators, and we have never produced a single 1G to 4G radio, small cell, or base station without cavity RF filters or duplexers that provide sufficient Rx selectivity and out-of-band rejection.

Performing the radio RF impairments and system budgets derives the required level of Rx selectivity and out-of-band rejection and distribute them within the radio baseband, spreading factor section, and front end. The actual implementation is done with the help of digital filters, onboard ceramic and SAW (Surface Acoustic Wave)/BAW (Bulk Acoustic Wave) filters, and RF cavity filters and duplexers.

It is true that these implementations add cost, some complexity, incremental weight, and volume to LoRaWAN® Gateways. However, they ensure the best possible LoRaWAN® performance in any deployment condition close to a wireless transmitter. This results in significantly improved LoRaWAN® coverage, reliability, dependability, and transmission range in the presence of strong out-of-band interfering signals. In addition, the same LoRaWAN® operators can expect improved Rx linearity and lower Rx noise figure, ensuring more dependable and reliable LoRaWAN® networks performance, increased network capacity, and fewer device re-transmissions, and increased battery life due to low power consumption. There is no real performance drawback – only a small cost, weight and size increase at the gateway level, resulting in a substantially lower overall network cost (fewer gateways required to cover the same area) and much lower yearly operating costs (fewer issues and much lower overall maintenance).

Figure 1 is a picture of TEKTELIC Outdoor LoRaWAN® KONA Mega IoT Gateway with integrated custom-designed and built-in high-performance RF cavity duplexers. These duplexers combined with I/Q digital filters and on-board RF ceramic filters make the TEKTELIC KONA Mega Gateway perform exceptionally well in presence of strong interfering signals or any other real RF environment impacts. As one can see from the picture, these duplexers occupy ~60% of the gateway volume, almost half of the gateway weight, and are the most expensive items within the gateway, but they are worth every penny for customers that are deploying High Capacity, High Availability and Dependability Carrier Grade LoRaWAN® networks that are expected to last for 10 – 15 years without a single visit to a site.

To overcome similar issues with a number of customers that could not replace their original LoRaWAN® gateways for various reasons (time and cost being the most significant), TEKTELIC developed a proprietary IP67 ISM band RF front end filter that supports both FDD (Frequency Division Duplex) and TDD (Time Division Duplex) modes of operation with 3rd party vendor gateways. It does not offer the same performance as the TEKTELIC KONA Mega and Macro Gateways, but it does improve the overall LoRaWAN® performance significantly in the presence of strong interfering signals. Below is a picture of the ISM band RF front-end filter.

LoRaWAN® operators should be aware that if the LoRaWAN® network experiences Rx coverage, reliability, dependability, or other LoRaWAN® performance issues, they should scan for any strong signals present at the LoRaWAN antenna input, not just in the LoRaWAN band, but 200-300 MHz above and below the LoRaWAN ISM band as LoRaWAN gateway LNAs are very broadband devices. If such interfering signals are present, and in most urban and suburban deployments this is almost always the case if there are many other wireless systems deployed on the same tower, building, or pole (as well as rural LoRaWAN® deployments on the same tower), work with your System Integrator, LoRaWAN® gateway provider, another specialist, or contact the TEKTELIC team and we will be glad to assist you to improve your LoRaWAN® network coverage, reliability, dependability, and the overall LoRa® communication transmission performance.

As you see, in order to answer the question “how to improve LoRaWAN gateway performance?” you should consider a range of factors. It may be quite difficult to choose a gateway that will suit your needs, and you will definitely need a reliable partner for it. The TEKTELIC support team is ready to help you choose the best IoT gateway, so don’t hesitate and write to our email at info@tektelic.com.

1. Georgiou, A., Katkov, M., & Tsodyks, M. (2021). Retroactive interference model of forgetting. The Journal Of Mathematical Neuroscience, 11(1). doi: 10.1186/s13408-021-00102-6
2. Kim, S., Lee, H., & Jeon, S. (2020). An Adaptive Spreading Factor Selection Scheme for a Single Channel LoRa Modem. Sensors (Basel, Switzerland), 20(4), 1008. https://doi.org/10.3390/s20041008