PaperlayerView exact revision on arXiv

Source-linked AI summary

A Survey on LoRa Networking: Research Problems, Current Solutions and Open Issues

Jothi Prasanna Shanmuga Sundaram, Wan Du, Zhiwei Zhao

arXiv:1908.10195v1eess.SPeess.SY

TL;DR

LoRa networks must support low-cost, large-scale, energy-efficient IoT deployments while addressing coordination, allocation, reliability, and security challenges. This paper surveys LoRa deployments and technical research, organizing existing measurements and solutions through a taxonomy. It identifies open issues for practical large-scale deployment and reports LoRa node lifetimes of 2-5 years under stated operating conditions.

  • Problem

    Practical LoRa networks still face research challenges in link coordination, resource allocation, reliable transmissions, and security.

  • Method

    The paper surveys LoRa deployments, analyzes performance measurements, develops a taxonomy of research problems, reviews recent solutions, and discusses open issues.

  • Results

    LoRa lifetime is estimated at 2-5 years when transmission consumes around 120-150 mW, MCU operations consume 10-15 mW, and duty cycles vary from 0.1% to 10%.

  • Takeaways & Limitations

    The survey organizes LoRa networking challenges and solutions to clarify open issues relevant to practical large-scale deployments.

Abstract

from arXiv · show

Wireless networks have been widely deployed for many Internet-of-Things (IoT) applications, like smart cities and precision agriculture. Low Power Wide Area Networking (LPWAN) is an emerging IoT networking paradigm to meet three key requirements of IoT applications, i.e., low cost, large scale deployment and high energy efficiency. Among all available LPWAN technologies, LoRa networking has attracted much attention from both academia and industry, since it specifies an open standard and allows us to build autonomous LPWAN networks without any third-party infrastructure. Many LoRa networks have been developed recently, e.g., managing solar plants in Carson City, Nevada, USA and power monitoring in Lyon and Grenoble, France. However, there are still many research challenges to develop practical LoRa networks, e.g., link coordination, resource allocation, reliable transmissions and security. This article provides a comprehensive survey on LoRa networks, including the technical challenges of deploying LoRa networks and recent solutions. Based on our detailed analysis of current solutions, some open issues of LoRa networking are discussed. The goal of this survey paper is to inspire more works on improving the performance of LoRa networks and enabling more practical deployments.

I. INTRODUCTION

LPWAN addresses IoT demands for low cost, large-scale deployment, and high energy efficiency, while LoRa offers open, autonomous networking. This survey classifies LoRa research challenges, reviews recent solutions, and identifies open issues.

  • Motivation: IoT deployments are increasing in density and scale, motivating wireless networks for connected sensing applications.Connected IoT devices were projected to grow from seven billion in 2017 to 20 billion in 2023.
  • LPWAN Motivation: LPWAN uses simple topologies, long-distance communication, and low data rates to achieve high energy efficiency beyond shorter-range technologies such as Zigbee and Bluetooth.
  • LoRa Positioning: LoRa is an open-source technology that enables low-cost, autonomous LPWAN networks without third-party infrastructure for diverse IoT deployments.Applications include smart buildings, smart cities, smart agriculture, smart meters, and water-quality measurement.
  • Survey Scope: The survey focuses on LoRa performance measurements, develops a taxonomy of research problems, reviews recent solutions, and discusses open issues for practical deployments.Its analysis covers challenges including link coordination, resource allocation, reliable transmissions, and security.

II. A BRIEF TECHNICAL BACKGROUND OF LORA

LoRa operates in unlicensed sub-GHz bands with CSS modulation, supporting long-distance, low-power communication through configurable operating modes and spreading factors. Its practical properties include multi-channel gateway reception, long node lifetime, and robustness to Doppler effects.

  • Technical Architecture: LoRa uses unlicensed sub-GHz ISM bands, configurable bandwidths of 125KHz, 250KHz, and 500KHz, and payloads up to 250 bytes over 5-15 Km.A LoRa system contains end-devices, gateways, network servers, and application servers.
  • Operating Modes: Class A devices use receive windows after confirmed uplinks, while Class B adds gateway-scheduled windows and Class C permits reception whenever the device is not transmitting.
  • Physical Layer: LoRa’s CSS physical layer varies chirp frequency linearly over bandwidth, making signals resilient to noise, fading, and interference.Spreading Factor determines how many data bits each chirp can encode; for example, SF9 encodes nine bits per chirp.
  • LoRaWAN: LoRaWAN centralizes MAC decisions at servers and uses either ALOHA with exponential back-off or TDMA scheduling to divide airtime.
  • Performance Properties: LoRa achieves distances up to 9Km in LoS at SF12 with PRR > 70%, while NLoS deployments reach around 2Km.Communication distance depends on bandwidth, spreading factor, transmission power, and coding rate.
  • Performance Properties: LoRa transmission consumes around 120-150 mW, while MCU operations consume 10-15 mW, supporting an extrapolated 2-5-year lifetime at 0.1%-10% duty cycles.
  • Performance Properties: Gateways can receive concurrently on eight channels, and transmissions using different orthogonal spreading factors from SF7-SF12 can share a channel.
  • Performance Properties: CSS modulation makes LoRa robust to Doppler effects, with mobile LoRa devices in constant-speed LoS conditions achieving PRR > 85%.

III. EXISTING DEPLOYMENTS OF LORA NETWORKS

LoRa networks have been deployed across urban management, household monitoring, smart metering, golf-course operations, and water management. These deployments demonstrate practical applications and reported reductions in resource use, operating costs, or energy consumption.

  • Smart cities and Urban Deployments: LoRa deployments support smart parking, lighting, waste management, solar-power monitoring, and household power monitoring in urban settings.The surveyed deployments include Seoul, Carson City, Lyon, and Grenoble.
  • Smart cities and Urban Deployments: 66% lower waste-collection frequency, 83% lower costs, and 46% higher recycling were reported for LoRa-enabled smart bins in Seoul.
  • Smart cities and Urban Deployments: 16% lower household power consumption was reported after power-consumption monitors were deployed in Lyon and Grenoble.
  • Smart meters: Around 7000 households in Gehrden, Germany, used LoRa-enabled smart meters to transmit periodic meter readings and reduce monitoring labor.The deployment served a population of 15,000.
  • Smart golf course: LoRaWAN sensors on Calgary golf carts provided real-time movement and location data to detect play delays and provide assistance.The application targeted a large course requiring periodic updates with low power consumption.
  • Smart Islands: 25% water savings were reported after LoRaWAN sensors were installed for water-quality and water-level management in Mallorca.

IV. TAXONOMY OF RESEARCH PROBLEMS

The survey organizes LoRa networking challenges into energy consumption, coverage, multiple access, and reliable transmission problems. It links these challenges to device longevity, diverse deployment conditions, scalability, and collision-induced data loss.

  • Energy Consumption: Power consumption is a major challenge because LoRa end-devices are expected to operate for 5–10 years with minimal maintenance.The paper distinguishes micro-controller operations from wireless transmissions as sources of consumption.
  • Communication Range: Coverage must address attenuation, propagation loss, and fading across environments ranging from open air to concrete- or steel-enclosed locations.The goal is improved signal penetration and network coverage.
  • Multiple Access: Multiple access comprises link coordination and resource allocation for allowing many end-devices to share limited spectrum.The relevant transmission parameters include Spreading Factor, transmission power, bandwidth, and channel.
  • Link Coordination: ALOHA and TDMA scheduling cannot handle collisions when thousands of devices connect, motivating new coordination techniques for denser deployments.
  • Error Correction: Long-distance transmissions can be corrupted or lost through channel effects, environmental conditions, or collisions, while existing Hamming-code correction is inefficient.The paper identifies channel coding and interference cancellation as solution categories.
  • Interference Cancellation: Netscatter cancels collisions from 256 concurrent transmissions, but this remains inadequate for thousands of end-devices served by one gateway.The survey therefore identifies interference cancellation as an open problem.

E. Security

The supplied passages describe security concerns alongside surveyed LoRa performance and energy-saving solutions. They identify key management, authorization, and practical constraints affecting LoRa deployments.

  • Security: LoRa security uses symmetric AES-128 encryption, but its keys are generated once and never updated.The paper identifies key generation and key-update mechanisms as major concerns.
  • Security: Eavesdropping, selective forwarding, and node impersonation attacks can target encryption keys, while differing application and network providers require third-party authorization.
  • Energy Consumption: LoRa Backscatter reports 2.8 Km transmission at 9.25 µwatts and 37.5 Kbps, with nearly 1000× lower power consumption than standard LoRa.
  • Energy Consumption: Ambient energy harvesting and backscatter transmission are proposed to reduce or eliminate battery dependence in LoRa devices.PLoRa uses harvested solar energy and backscatters ambient LoRa transmissions without external excitation signals.
  • Energy Consumption: Charm extends battery lifetime up to 4× standard LoRa by avoiding retransmissions, while FADR reduces standard LoRaWAN power consumption by 22%.
  • Energy Consumption: Wireless power transfer is described as infeasible for low-power, low-cost LoRa modules because of complex and costly hardware extensions.

B. Communication Range

LoRa communication-range studies examine deployments across dense urban, industrial, and outdoor settings, while proposed solutions extend coverage through constructive interference and cooperative gateway decoding.

  • Measurements: More than 95% packet delivery was achieved across three dense-urban trial phases, including 1,000,000 packets at 417 packets per hour.Ten gateways operated on eight channels with 100 sensors emulating 10,000 end-devices.
  • Measurements: 15 Km ground range and 30 Km water range were measured, with packet-loss ratios of 34% and 32%, respectively.The measurements were collected around Oulu, Finland, and used to calculate a channel attenuation model.
  • Research solutions: Choir exploits small carrier-frequency deviations to disentangle collided signals and extend range to 2.64× standard LoRa.Coarsely synchronized, geographically co-located nodes create constructive interference that raises the signal above the noise floor.
  • Research solutions: Charm combines weak signals from multiple gateways at the cloud to improve range 3× and battery life 4× over standard LoRa.Auxiliary gateway hardware detects very weak signals, and a heuristic jointly selects signals for cloud decoding.
  • Key insights: LoRa measurements identify gateway density, radio imperfections, weak-signal decoding, and commercial-device implementation as continuing range considerations.Commercial LoRa operation of GNU-radio coverage extensions has not yet been studied.

C. Error Correction

Error-correction solutions use channel coding and interference cancellation to recover lost or collided LoRa data, but redundancy-based recovery can become inefficient under heavier packet loss.

  • Channel coding: DaRe adds redundant information using Convolutional and Fountain codes so lost frames can be calculated from received frames.Its application-layer design requires buffering previous data units to compute the redundancy.
  • Channel coding: Higher packet-loss probabilities make lower code rates increase interference and reduce data-recovery efficiency.The analysis also motivates less complex decoding to reduce latency.
  • Key insight: Redundancy-based data recovery cannot be used alone to address packet loss because additional transmissions increase collisions.This limitation is especially relevant when packet losses are large or bursty.

2) Interference Cancellation:

LoRa interference-cancellation research decodes concurrent transmissions through constructive interference and distributed coding, addressing both collision recovery and scalability.

  • Interference cancellation: Choir and FTrack use constructive interference to disentangle collided signals and improve throughput 6.84× over standard LoRa.Test-bed evaluation also reports better results for Choir than multiple-antenna deployments.
  • Interference cancellation: NetScatter decodes 256 concurrent transmissions and improves throughput 14–62× and latency 15–67× over existing interference-cancellation techniques.Its distributed CSS coding decodes concurrent transmissions below the noise floor in a single FFT operation.
  • Key insights: Choir’s scalability limit of 5–10 concurrent devices motivates NetScatter’s distributed error-correction approach.The survey identifies NetScatter as a scalable successor to Choir’s collision-decoding strategy.
  • Key insights: Channel-coding research remains relevant because efficient coding could improve reception rate and network lifetime.DaRe and its analysis are cited as evidence of this continuing research opportunity.

2) Resource Allocation:

Resource-allocation studies tune spreading factors, bandwidth, transmission power, and channel assignments to balance packet reception, energy use, fairness, and scalability across deployment conditions.

  • Measurements: 96.7% static and 95% mobile packet delivery was measured in Oulu experiments without acknowledgements, retransmissions, or adaptive data rate.End-devices transmitted every five seconds across six channels.
  • Research solutions: The probing algorithm finds a near-optimal energy-efficient setting using only 44% more energy than ideal within 285 probes.The algorithm probes alternatives while limiting transmission-power exploration and iteration count.
  • Research solutions: Adaptive parameter allocation can scale networks through more gateways, personalized data rates, and channel assignments, while EXPLoRa methods outperform basic ADR in simulation.EXPLoRa-AT additionally equalizes time-on-air through ordered waterfilling.
  • Measurements: Energy consumption varied by up to 50% between lowest and highest transmission power for the same packet.The reported implication is to use the lowest possible transmission power for maximum energy saving.
  • Measurements: Increasing spreading factor does not always improve communication performance; bandwidth increases can improve packet reception in some settings but not universally.Performance depends on spreading factor, bandwidth, environment, and deployment location.
  • Measurements: Only a 10% packet-reception difference separated fastest and slowest PHY settings for farthest devices, supporting lower resource consumption.The survey links this result to using lower spreading factors and transmission power where feasible.

E. LoRa Security

LoRa security research identifies multiple attacks and examines key-management, authentication, and trust mechanisms. The survey highlights both proposed protections and unresolved application-specific security needs.

  • Key Generation: Initial and Non-Initial Join requests use tokens and reboot-triggered recovery to address replay risks and token loss, with security improvements supported by theory and experiments.
  • Threats: Physical compromise of an end-device can expose security keys and enable message decryption across the network.
  • Threats: LoRa devices remain susceptible to jamming, replay, beacon synchronization, traffic-analysis, and man-in-the-middle attacks.
  • Key Generation: Increasing DevNonce from 16 to 24 bits reduces join-attack risks, although the change is incompatible with the current LoRaWAN specification.
  • Research Solutions: Proposed security approaches include separated application and network key management, proxy-based exchange, third-party authorization, and blockchain-based trust mechanisms.

VI. OPEN ISSUES

Some LoRa solutions still leave performance improvements unresolved because dynamic communication-parameter selection does not account for ambient temperature.

  • VI. OPEN ISSUES: Dynamic communication-parameter solutions may omit ambient temperature, despite its major role in reducing signal strength.

A. Optimal placement of multiple gateways:

Multiple gateways can improve scalability and interference performance, but their deployment requires application- and hardware-specific placement decisions.

  • A. Optimal placement of multiple gateways:: Multiple gateways improve scalability and interference performance, yet optimal placement depends on the application and hardware constraints.
  • A. Optimal placement of multiple gateways:: A generic gateway-placement solution organized by application category could further improve LoRa performance.

B. Link Co-ordination:

Reliable long-range LoRa communication remains constrained by downlink acknowledgments, retransmission timing, communication-range limits, and unresolved security requirements. The survey identifies dynamic coordination and deployment approaches as continuing needs for practical large-scale networks.

  • B. Link Co-ordination:: Delayed or corrupted downlink ACKs can prevent end-devices from transmitting subsequent packets, motivating dynamic ACK mechanisms.
  • B. Link Co-ordination:: Static retransmission policies degrade performance when ACK arrival exceeds the retransmission timer; dynamic, modular policies remain unresolved.
  • C. Communication Range: Reliable transmissions are needed for long-range LoRa links, while future chip functionality may further enlarge communication range.
  • C. Communication Range: Choir requires potentially modified commercial radios, and NetScatter still needs hardware modifications despite supporting concurrent transmission by 256 nodes.
  • D. Security: Security requirements remain application-dependent, with several attacks still needing attention in LoRa networks.
  • VII. CONCLUSION: The survey analyzes deployment challenges and recent solutions to identify open issues for practical large-scale LoRa networks.
Loading 1908.10195v1…