4 Technical Approaches to Ensure Interference Resilience in LPWAN

For Low-Power Wide Area Networks (LPWAN) operating in the license-free spectrum, a major advantage is low network costs. Nevertheless, given the exponential increase in connected devices, the shared limited radio resources are becoming more and more congested. To enhance Quality-of-Service (QoS) and network scalability, ensuring interference resilience in LPWAN is a major undertaking.

Understanding Interference in License-Free Bands

Interference refers to the unwanted collision of two radio signals in the same frequency – causing data loss. Interference in license-free LPWAN, falls into two main categories:

1. Inter-system interference refers to disturbances caused by radio signals from other systems. As the license-free spectrum is available for everyone, multiple technologies co-exist and access the same frequency resources. For example, most LPWAN technologies including MIOTY, LoRa, and Sigfox commonly use the sub-gigahertz industrial, scientific and medical (ISM) radio bands. Similarly, Ingenu – another LPWAN player – shares the crowded 2.4 GHz band with Wi-Fi, Bluetooth, Zigbee, among others.

2. Intra-system interference, or self-interference, refers to disturbances caused by devices operating within the same network, such as within a MIOTY network or within a LoRa network. Self-interference is mainly attributable to asynchronous communication using ALOHA scheme in many LPWAN systems. Though greatly lowering power consumption, pure ALOHA-based networks generate significant self-interference due to uncoordinated, random data transmission among end devices.

Inter- and intra-system interference threaten to deteriorate network performance and hamper scalability.

Technical Approaches to Interference Resilience in LPWAN

Amid these challenges, a strong system design is key to ensuring high interference immunity in LPWAN. Below we explain four technical approaches to controlling and mitigating inter- and/or intra-system interference.

1.  Utilizing (ultra-) narrow bandwidths

Compared to wideband approaches based on spread spectrum, (ultra-) narrowband technology alleviates the problem of intra-system interference. Each narrowband message uses a very small bandwidth, allowing for high spectrum efficiency. More messages can hence fit into an assigned frequency band without overlapping with each other, enabling more devices to effectively operate at the same time without interfering with each other. This improves overall network capacity and system scalability. Minimal bandwidth usage additionally reduces noise level experienced by each signal.  

Think of narrow band messages as motorbikes and wideband messages as trucks. On a highway, we can afford a much larger number of motorbikes than trucks without incurring traffic accidents.

2.  Reducing on-air time

In many LPWAN systems, the transmission time or on-air time of a signal can last up to 2 seconds. This is problematic since messages with long on-air time are much more prone to collisions. Longer transmission times also increases opportunities for malicious and sophisticated attacks like selective jamming.

3.  Frequency hopping

By rapidly switching a message among different channels during transmission, frequency hopping improves resistance against inter-system interference. Constant frequency change helps avoid congested channels and makes signals difficult to intercept. On the downside, frequency hopping is very spectral inefficient as larger bandwidth usage is required. Wideband signals transmitted at low rates can easily overlap with each other, causing self-interference and data loss.

4.  Forward Error Correction (FEC)

Applying channel coding or forward error correction allows for detection and correction of transmission errors due to noise, interference, and fading. In unreliable or noisy channels, FEC helps reduce packet error rate and avoid costly data re-transmissions.

So far, no traditional LPWAN systems have succeeded in leveraging all of these approaches in their system design. LPWAN using an (ultra-) narrowband approach offers high spectrum efficiency, but extends on-air time due to very slow data rates. Spread spectrum systems capitalize on the benefits of frequency hopping, but suffer from self-interference and scalability issues due to wide bandwidth usage.

By splitting an ultra-narrowband message into multiple smaller sub-packets and distributing them at pseudo-random time and frequency patterns, Telegram Splitting brings the benefits of all four mentioned approaches to one system. Thanks to its much smaller size, each sub-packet has an extremely short on-air time of only 15 milliseconds. The chance of colliding with other inter- and intra-system signals is hence drastically minimized. Additionally, built-in FEC enables successful message retrieval even if up to 50% of sub-packets are lost along the way.

With the ever-growing device density and communication traffic in the IoT era, interference resilience in LPWAN will continue to be a top priority; as will selecting a robust technology without compromising cost and power efficiency.