When a monitoring system loses its data, the only safe output is an explicit "cannot determine". A system that reports "no anomaly found" because it could not measure has not failed quietly — it has actively misled the operator, and it has done so in the direction of reassurance.
Need product specifications, use cases or references?
Call +86 130 4890 6506 for selection guidance and comparable project types Email usConsider two states of the world:
| What is true | What the screen says (in many systems) |
|---|---|
| The cable is fine; every measurement point was read | Nothing wrong |
| The fibre was cut three days ago; nothing has been measured since | Nothing wrong |
These two conditions are operationally opposite and visually identical.
The second is worse than having no monitoring at all. With no system, the operator knows they are relying on patrols. With a system that silently stopped measuring, the operator believes they have continuous coverage — and will deprioritise the patrol that would have caught the problem.
This failure is silent, permanent until noticed, and it fails toward false comfort. It deserves to be treated as the most serious defect a monitoring system can have.
Missing data in a DTS installation is not exotic. It arrives through routine events:
| Source | What the system sees |
|---|---|
| Damaged or disconnected fibre | No profile, or a truncated profile |
| Device power loss or reboot | Gap in the scan sequence |
| Paged-read timeout | Partially read profile |
| Connector disturbance | Degraded signal over part of the fibre |
| Sensor deployed but never mapped | Data arriving that belongs to no known asset |
| Coverage gap | A section of cable with no fibre at all |
| Policy not released | Data present, but no rule exists to judge it |
Each of these is a different condition with a different remedy. Collapsing them all into "no alarm" destroys the information the operator needs to fix them.
A monitoring system needs an explicit state meaning "I cannot judge", separate from "I judged, and it is fine" and separate from "I judged, and it is not fine".
In our five-level scheme that state is "cannot determine", and it sits alongside no anomaly found, attention advised, anomaly present and serious anomaly.
The design rule is blunt: when data, coverage, mapping or policy is insufficient, the system must not display "no anomaly found".
Quality must be carried as part of the measurement, not applied as a cosmetic filter afterwards. Our pipeline keeps four layers distinct:
The important consequence: a conclusion can never be detached from the quality of the data underneath it.
If the answer is "cannot determine", the operator's next question is "why not, and what do I do about it?"
So the verdict must carry its own explanation: which object was assessed, at what time, over what coverage, with what data quality, under which policy, with which limitations — and what action would restore the ability to judge.
"Unable to determine" without a reason is barely better than silence.
Capability tiers are legitimate: more data buys better judgement.
But the boundary must hold in both directions: when current data is absent, the system must not assert a definitive cause such as overload or connection fault. It may report what it observed; it may not claim to know why.
A system that guesses a cause will eventually have that guess acted upon. The cost of a wrong confident answer in industrial operations is not a bad customer experience — it is equipment damage or an outage.
A related temptation: if spatial resolution is coarse, present it as fine; if sampling is sparse, present it as dense. Our rule is plain: capability is not inflated according to sampling density.
If a section is covered by fewer measurement points than the operator assumes, that is a coverage fact that must be surfaced — not smoothed over.
We can provide: selection guidance · integration and network design guidance · product documentation on request
Call +86 130 4890 6506 for selection guidance and comparable project types Email usTell us your site conditions and what the data needs to connect to — we reply within one business day.
It might seem that a system which admits uncertainty is a weaker product than one that always gives an answer. The opposite is true in industrial operations, for three reasons:
This is also why our systems do not trip breakers, interrupt supply or shed load. Detecting, judging and acting are separated deliberately, and the operator authorises the action.
An explicit data-quality or coverage condition — never "no anomaly found". Loss of communication is identified within the design target of 15 seconds and surfaced as a distinct state that requires attention.
A sensor-fault alarm tells you something is wrong. A data-quality state tells you what you can and cannot conclude from the data you still have — which is what determines whether the cable is currently unmonitored. The second is the operationally important question.
Yes. Basic temperature monitoring runs in full without current data. When current data is available, load-correlated enhancement is enabled. What the system will not do without current data is assert overload or connection fault as a confirmed cause.
Yes — coverage is tracked and reported. If a section of cable has no fibre, or has fewer measurement points than the surrounding sections, that is surfaced rather than averaged away.
A partially read profile is treated as a data-quality condition, not silently completed. The reasoning is validated on both sides of a paged read to prevent frame mixing across scans.
Yes. A verdict carries object, time, location, value, trend, data quality, coverage, mapping, policy, limitations and a recommended action. "Cannot determine" without a reason would be nearly as unhelpful as silence.
Yes, and it stays labelled as such. Data is retained in tiers — per-frame profiles for 7 days, minute-level profiles for 1 year, assessment and alarm evidence for 3 years — with evidence bundles immutable and backed up off the machine. A later review can see both the reading and the caveat that applied when it was taken.
If you need to know whether your cable sections are genuinely covered — not just whether the system is quiet — tell us your layout and the DTS device you are reading, and we will tell you what the coverage and quality picture would look like.
Call +86 130 4890 6506 · export@rebe-tec.cn