Secure Ranging: Delay, Distance, and Policy

Ada converts the chapter’s ranging times into distances and checks each access-policy case against the unlock zone

foundations
math-foundations
uwb
security
intermediate
Ada ADA · CALCULATION AUDIT

Secure Ranging: Delay, Distance, and Policy

Ada: Secure ranging is only useful when the timing number and the policy threshold agree. The chapter already gives the numbers needed to audit the physics and the access decision without inventing a new measurement.

A secure cabinet sets a policy threshold of 1.2 m: a phone presents a valid credential and the cabinet receives a UWB result of 0.74 m with a fresh timestamp, yet that same reading must be denied if it comes from the wrong side of the cabinet or if the ranging record is 2.5 seconds old when policy allows only 500 ms. IEEE 802.15.4z’s Scrambled Timestamp Sequence is meant to stop an attacker from forging an early pulse, and the chapter shows a 20 ns extra round-trip delay reading as about 3 m of extra apparent path. This audit asks the question those policy cases invite: does the chapter’s own timing-to-distance arithmetic actually support each accept, deny, and fallback verdict it claims?

Companion to the chapter UWB Applications and Security — every number here comes from that chapter.

Convert the timing evidence and check each policy case

  • Convert timing to distance: 1 ns x 0.30 m/ns = 0.30 m of one-way travel.
  • Check the relay-delay example: a 20 ns extra round-trip delay becomes 0.30 m/ns x 20 ns / 2 = 3.0 m of extra apparent one-way path.
  • Compare the unlock record to the zone: 1.2 m - 0.83 m = 0.37 m of distance margin before freshness, side-of-door, and confidence checks are applied.
  • Check the stale-record case: 1.6 s = 1600 ms, which is 1600 ms - 180 ms = 1420 ms older than the fresh example record, so the policy should not treat the old distance as current evidence.
  • Check the delayed-path denial: 3.8 m - 1.2 m = 2.6 m beyond the unlock zone, so the distance evidence should deny or fall back even if the credential itself is authorized.

The audit conclusion is bounded: the arithmetic supports the chapter’s policy examples, but the action still depends on credential authorization, confidence, side-of-door evidence, timestamp freshness, and fallback behavior.

Every number above is taken from the chapter’s own ranging and policy examples and re-derived step by step.