[Report] When the Spectrum Becomes a Battlefield: Defending Urban IoT Under Radio Jamming - CODE BLUE 2026 #codeblue_jp #codeblue2026

[Report] When the Spectrum Becomes a Battlefield: Defending Urban IoT Under Radio Jamming - CODE BLUE 2026 #codeblue_jp #codeblue2026

A report on the CODE BLUE 2026 session "When the Spectrum Becomes a Battlefield: Defending Urban IoT Under Radio Jamming."
2026.11.18

This page has been translated by machine translation. View original

Hello, this is Usuda.

Everyone, are you taking security measures? (Greeting)

This is a report on the following session held at CODE BLUE 2026.

Whether you are handling radio waves in urban infrastructure or in any other context, this session offered lessons on what to do in an emergency.

When the Spectrum Becomes a Battlefield: Defending Urban IoT Under Radio Jamming

Smart cities rely on telemetry over LoRaWAN and LPWAN to maintain the observability (visibility) of critical infrastructure, especially during failures and disruptions. However, in harsh environments such as conflict zones, the radio layer faces jamming (radio interference) from electronic warfare (EW), causing packet loss, while a naive countermeasure of "simply retransmitting repeatedly" ends up exhausting batteries and communication bandwidth (airtime).

Based on operational experience on the front lines of a municipal LoRaWAN network that faced electronic warfare interference during wartime, this talk presents a lightweight operational model that does not require measurement equipment such as spectrum analyzers. Focusing on the SINR (signal-to-interference-plus-noise ratio) at reception, it explains how to detect and classify the likelihood of jamming using only the telemetry collected by ordinary gateways and network servers (RSSI/SNR, bursts of CRC errors, per-channel loss, gateway diversity). It further shows how to identify the "fingerprints" (characteristics) of broadband (wideband), narrowband, and sweep-type jamming, and to apply a mitigation package suited to each.

Finally, it organizes the trade-off between robustness and communication capacity, and explains why the combination of "partial fortification" and a "degraded mode" at the service layer is superior to "aiming for maximum robustness everywhere." Note that this talk is structured to be vendor-neutral (independent of any specific vendor), and sensitive parameters have been generalized.

Speakers: Pavlo Chernikov

Report

  • What happens when the radio spectrum itself becomes a battlefield
  • Urban infrastructure has been affected
    • Data centers, optical fiber lines, power, mobile networks
  • Kyiv Digital (city telecom service)
    • 80 craftsmen/engineers
  • When everything stopped
    • The invasion began in 2022
    • Everyone tried to leave at once by car
    • The transportation network was at a standstill
    • Members came to work
    • It worked precisely because they came to work
    • The city's data center
      • Being destroyed by a missile was not written in the disaster drills
      • It was brought back to an operational state in 5 days
      • Even if protected from hackers, if not protected from missiles, you are only handling one threat
      • Staff stayed overnight in cars and similar places
      • When building a backup site, you must imagine that people will live there
    • Buried optical fiber cables were cut
      • Emergency response personnel at the site
      • They worked for about 2 hours
      • There was fighting right nearby, which sped up the work
    • Following a request from the government, the direction of cameras was changed
      • Working with flashlights in complete darkness
      • Does the sniper know about it?
      • But they don't know
      • Communication methods between civilians and the military must be decided in advance
      • Both light and radio waves can be noticed
    • Power outage
      • Drone attacks begin
      • Batteries last only a few hours
      • Without power, communication is cut off
      • Satellite terminals were a great help
      • Started building out a TETRA radio network
      • Power and dependencies need to be sorted out
    • Mobile carriers
      • Full recovery took more than 7 days
      • Things were running on SIM cards from a single carrier
      • Switching control panels to non-proprietary LoRaWAN
    • Common challenges
      • When critical foundations were affected, they distributed across multiple options or migrated to their own systems
  • What happens if the radio spectrum is affected
    • Why do cities use LoRaWAN
    • Low cost and wide coverage
    • But what happens during a crisis
    • What matters is that power outages are not a factor
    • As long as it can receive, communication continues
    • It is needed precisely when the network is needed most
    • LoRa uses SF7-12
    • Raising the SF extends the range
    • Can you receive under jamming
      • Strengthen the received signal
      • Reduce interference
      • Lower the threshold
    • These are the three mitigation measures
    • Lessons from electronic warfare
      • Someone intercepts and then jams
      • Countermeasures quickly become obsolete
      • Anything can be jammed
      • LoRa is not a technology exclusive to cities
      • It is also used by the military
      • They share the same spectrum
      • Urban spectrum cannot be swapped out quickly
      • Battery capacity is also a severe constraint
  • Operational model
    • Let's think from the perspective of a network administrator
    • The lab has a spectrum analyzer
    • The operations side does not
    • But there are logs
      • RSSI, SNR, etc.
    • They act as a spectrum sensor
    • We read through these
    • First, detection
      • Confirm that radio performance is declining
    • Notice and respond
    • Finally, review
    • Fixed settings lose
    • Detection starts from a small number of values
      • Look at the distribution of RSSI and SNR
      • Percentage of packets that fail the CRC check
      • Number of receptions per channel
      • Gateway diversity
      • Delivery rate per class
    • The first question during an incident is whether it is a radio problem
      • There are various other factors
      • The symptoms differ
      • Network failures, interference, etc.
      • Check by things like whether the gateways are running
    • Classification
      • Classify the causes
        • If it is across the board, broadband jamming is possible
        • If it is specific channels, an attack on only those channels
      • We created three classes based on the symptoms
        • A is across the board
        • B is partial channels
        • C is localized
          • Affects a specific area
        • These alone cannot explain everything, but they are enough to decide what to do
      • For A, look at spatial diversity
      • For B, frequency diversity
        • Spread across multiple channels
      • For C, respond by focusing on the target area
        • Change placement and use directional antennas
    • When packets are lost, you want to retransmit
      • Under interference, that becomes an aggravating factor
      • The more traffic there is, the more interference increases
      • Airtime available for downlink is short
      • Retransmission leads to problems with the radio space and batteries
      • It has a negative impact on the whole network, so it is not enough for only the devices with countermeasures
      • Set a time limit on hardening
      • Even maximum interference resistance slows down the network
      • Devices automatically become low speed, so manage the changes
    • Radio is half of the countermeasures
    • The other half is done at the service level
      • Low-priority telemetry can be delayed
      • Buffer it and send later
        • High-priority data also uses alternative routes
      • Narrow down alerts
        • Because they keep going off
    • Retrospective
      • Record four things for each incident
        • Time, channel, how it was recovered
        • Record the normal state
        • If you do retrospectives, you can notice signs of obsolescence
  • Is this only a wartime matter?
    • The same applies to other disasters and the like
  • What can be done now
    • Record the normal state from gateway logs
    • Countermeasures differ for each case
    • Do not rely on retransmission
    • Build degraded operation not only into the radio but also into the service
    • Hold a retrospective for each incident
  • Any communication can be jammed

Impressions

We need to think about infrastructure and radio operations with emergencies in mind.

Retransmission design is the same in Wi-Fi design and the like, so I could understand it well.

Share this article