Balancing the Energy Consumption and Latency of Over-the-Air Firmware Updates in LoRaWAN
Siddhartha S. Borkotoky · 2025
Read Balancing the Energy Consumption and Latency of Over-the-Air Firmware Updates in LoRaWAN by Siddhartha S. Borkotoky (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-10588.txt.
Introduction
Siddhartha Borkotoky tunes the delivery of firmware updates to battery-sipping radio networks so operators can choose their price in time or energy. LoRaWAN devices are energy-starved and legally duty-cycled, yet over-the-air updates are essential for security. Energy spent and delivery delay fight each other directly. The proposed scheme walks the spreading factors in sequence.
A fixed count of frames goes out at one spreading factor before the scheme steps to the next. Two knobs, the smallest factor used and the frames per factor, set the trade-off. Urgent security patches can ride the fast, thirstier setting. Routine updates can crawl cheaply. Flexibility is the contribution: one mechanism spans the frontier instead of a single compromise point. The analysis quantifies how the knobs move energy and latency.
Catalog record
| CREATOR | Siddhartha S. Borkotoky |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 47 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-10588
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Balancing the Energy Consumption and Latency of Over-the-Air Firmware Updates in LoRaWAN Siddhartha S. Borkotoky Abstract Over-the-air firmware updates are crucial for mitigating security threats and maintaining up-to-date device functionality in Long Range Wide Area Networks ( LoRaWANs ). LoRaWAN end devices are usually energy-constrained, and LoRaWAN transmissions are subject to duty-cycle restrictions. Consequently, controlling the energy expenditure and update-delivery latency of FUOTA are key challenges. We propose a flexible scheme that achieves a tunable trade-off between the energy consumption and delivery delay. The scheme employs the LoRa spreading factors sequentially to transmit update-carrying frames, sending a fixed number of frames with a given spreading factor before moving to the next. By adjusting the smallest spreading factor to be used and the number of transmissions per spreading factor, a suitable energy-delay trade-off can be achieved. Thus, time-sensitive updates, such as security patches, may be sent with a low-delay-high-energy setting, whereas a more energy-efficient but higher-delay setting may be used for non-critical updates. I Introduction LoRaWANs support low-rate, long-distance, and low-power transmissions of small amounts of data by end devices (EDs), and exhibit lifetimes of up to several years. In a long-lived network, it is necessary to occasionally update the firmware at the EDs to enhance security features or implement new functionalities. Since a LoRaWAN can feature many EDs, potentially in hard-to-reach locations, it is beneficial to update them wirelessly without manual intervention. This paradigm, referred to as Firmware Update Over The Air (FUOTA), has been standardized by the LoRa Alliance’s FUOTA working group [ 1 ] . In FUOTA, a firmware update server instructs the LoRaWAN gateway to transmit the firmware image to all eligible EDs. However, this process involves multiple challenges. First, LoRaWAN transmitters must comply with strict duty-cycle regulations. For example, in the European Union, the maximum duty cycle must not exceed 1%. Given the substantial size of a typical firmware image, duty-cycling has the effect that the gateway needs a long time to deliver the complete update to an ED. If the EDs are updated sequentially, it would take an unreasonably long time to update all devices. Therefore, FUOTA specifications suggest simultaneous delivery to all recipients by exploiting the broadcast nature of the wireless medium. Secondly, owing to its size, the update must be split into many fragments, such that each fragment fits into a single link-layer frame. Keeping track of the fragment delivery status at each recipient and retransmitting undelivered fragments via an acknowledgment-based mechanism can be extremely complex in a massive network. Therefore, FUOTA specifications advocate the use of forward erasure correction (FEC). In FEC, redundant fragments are transmitted to enable an ED to compensate for lost fragments without explicitly requesting retransmissions [ 2 ] . Yet another challenge, which is the focus of this work, involves the selection of physical-layer parameters for multicast transmissions. The noise tolerance of LoRaWAN transmissions depends on a parameter called the spreading factor (SF), which takes integer values between 7 and 12 [ 3 ] . Higher SFs provide better receiver sensitivity, meaning that frames can be received over longer distances. However, higher SFs increase the frame duration, which in turn increases the time needed to deliver the update and the energy expenditure at the transmitter and receiver. With the EDs distributed over a large region, different SFs may be appropriate for different devices. Most existing works on FUOTA assume that the gateway uses an SF that is adequate even for the farthest device [ 4 , 5 , 6 , 7 , 8 ] . The authors of [ 9 ] show that this approach is not energy efficient for nearby devices. They propose an energy-efficient mechanism in which the EDs are grouped according to their optimal SFs. The update is delivered separately to each group. However, this approach incurs large delivery delays, since a group begins receiving frames only after all previous groups have been updated. To address the limitations of the fixed-SF and group-based approaches, we propose a transmission scheme that provides fine-grain control over the energy-latency trade-off. The scheme employs multiple SFs, with a fixed number of transmissions assigned to each SF. This number, along with the set of SFs, can be adjusted to control the energy-latency trade-off according to the needs of the update application. The scheme provides large reductions in the update time compared to the group-based scheme. Its energy consumption is only slightly higher than that of the group-based scheme and significantly lower than that of the fixed-SF scheme. I-A Related Works In [ 4 ] , the authors analyzed the impact of LoRaWAN transmission settings on FUOTA performance. It is shown that the higher data-rates (i.e., lower SFs) facilitate faster update and lower energy consumption at the cost of reduced update efficiency (i.e., the fraction of devices that receive the complete update). This observation underscores the need for careful SF selection. The trade-off between data rate and update efficiency is also highlighted in [ 5 ] . A practical evaluation of FUOTA is reported in [ 6 ] . Efficient information delivery in wireless networks requires a mechanism to compensate for inevitable frame losses. For this, the FUOTA working group proposed an FEC scheme [ 2 ] that transmits redundant coded fragments (linear combinations of the original fragments) using a low-density parity-check (LDPC) code. The redundant fragments may enable an ED to correct for some lost data. A related FEC technique called rateless coding is used in [ 7 ] . A salient feature of a rateless code is that it allows the source to continue sending redundancy until all recipients recover all lost data. In contrast, the LDPC mechanism of [ 2 ] transmits only a predefined amount of redundancy, hence some recipients may fail to recover all lost data. In [ 7 ] , the gateway multicasts rateless coded fragments up to a predefined time. The nodes that have yet to receive the complete update are then individually delivered the remainder of the update using a beamforming strategy. The transmission scheme of [ 7 ] is improved in [ 8 ] by incorporating a security mechanism. The aforementioned techniques employ a fixed SF for all transmissions. However, fixed-SF assignments can be inefficient. If the SF is large, nodes near the gateway spend more energy by receiving long frames although they could successfully receive shorter frames. With a small SF, nearby nodes spend less energy, but distant nodes experience excessive frame losses and require very large update times. As a remedy, [ 9 ] suggests grouping the nodes according to their appropriate SFs. The update is delivered group-wise. Once all nodes using a certain SF receive the update, transmissions to the next group begin. However, as mentioned earlier, the sequential handling of groups increases the overall delivery time. I-B Contributions of this Paper In contrast to fixed-SF and group-based schemes, which aim for either low energy consumption or low delay, we present a flexible transmission scheme that can tune the trade-off between the two metrics. By adjusting a pair of design parameters, the operating point can be set as per the need of the update application (e.g., a high-priority update can employ less energy-efficient operation for faster completion, and vice