Cambridge A-Level ICT: Monitoring and Control — Revision Notes

Condensed for the final weeks. For the full explanation, use the Monitoring and Control study guide.

Monitoring vs control — keep them separate

	Monitoring	Control
What it does	Reads/records a value for review	Triggers an automatic response
Example	A weather station logging humidity	A greenhouse vent opening at a temperature threshold
Human involvement	Human reviews the data	No human decision needed

The most common way marks are lost: describing a monitoring scenario when the question asks about control, or vice versa.

3.1 Sensors and calibration

Named sensors: light/UV, temperature, pressure, humidity, pH, gas (oxygen, CO₂, CO, nitrogen oxides), sound, infrared, touch, (electro)magnetic field, proximity.

Monitoring uses: environmental (water pollution, weather stations, ambient temperature, atmospheric pressure, humidity, sunlight) and patient monitoring.

Calibration type	What it does
One-point	Checks against a single known reference value
Two-point	Checks at two points (low/high range), adjusts offset and scale error
Multipoint	Checks several points across the range for highest accuracy where response isn’t perfectly linear

3.2 Control: matching sensor to real-world use

Sensor	Named application	Why it fits
Touch	Fluid level (nuclear plant cooling water)	Physical contact detection
Light	Car park barrier	Detects presence/level of light
Moisture	Soil water content	Measures moisture directly
Infrared	Burglar alarm	Detects body heat, no contact needed
Proximity	Smartphone screen near ear	Detects closeness without contact
Sound	Burglar alarm	Detects noise

Always justify the fit — naming a sensor without saying why it suits the scenario loses marks. A vague “temperature sensor for conditions” answer is weaker than “temperature sensor because it specifically measures temperature, triggering the vent at a defined threshold.”

Actuators carry out: linear, rotary, soft, hydraulic, pneumatic, electric, thermal, magnetic, mechanical actions.

Microprocessor-controlled systems (named): greenhouses, central heating, air conditioning, burglar alarms, traffic/pedestrian flow (smart motorways), car park barriers, traffic lights, wireless sensor/actuator networks, smart homes.

Worked example: sensor → microprocessor → actuator chain

A greenhouse vent system:

1. SENSOR:        Temperature sensor takes a reading
2. MICROPROCESSOR: Compares reading against a stored threshold
3. DECISION:       Reading > threshold?
4. ACTUATOR:       If yes, motorised/pneumatic actuator opens vent
                    (rotary or linear movement)
5. LOOP:            System re-checks the sensor continuously --
                    reverse signal closes vent once temperature
                    drops back below threshold

A flowchart answer must show the sensor reading, a decision box comparing it to the threshold, and the actuator action as the decision’s outcome — this is exactly the Section 4 (Algorithms and Flowcharts) link the syllabus expects.

Second worked example: a car park barrier system

A car park uses a light sensor and an induction loop to control entry.

1. SENSOR:        Light sensor detects a vehicle blocking a light
                   beam at the barrier; induction loop detects the
                   change in magnetic field caused by a vehicle's
                   metal body passing over it.
2. MICROPROCESSOR: Confirms a vehicle is present using both signals
                   together (reducing false triggers from, e.g., a
                   pedestrian or small object).
3. DECISION:       Is a valid ticket/payment also registered?
4. ACTUATOR:       If yes, a motorised (rotary) actuator raises the
                   barrier arm; a timer-based process lowers it again
                   once the vehicle has fully passed the loop.

Notice this example uses two sensor types together rather than one – a useful reminder that named real-world systems in this syllabus often combine several sensor types to increase reliability, not just a single sensor in isolation.

Why calibration matters even for a “correctly working” sensor

A sensor can be electronically functional but still give inaccurate readings if it has never been calibrated, or has drifted out of calibration over time. Calibration compares the sensor’s output against a known reference value and adjusts for any systematic offset — this is conceptually distinct from the sensor simply “working” in the sense of producing a signal at all. A pH sensor left uncalibrated for months, for example, may still respond correctly to changes in acidity, but every reading it gives could be shifted from the true value by a consistent offset — exactly the kind of systematic error that only calibration, not repeated readings, can correct.

Worked example: choosing a calibration type

A pressure sensor is known to respond in a genuinely linear way across its whole range, while a humidity sensor is known to respond non-linearly near the extremes of its range.

Pressure sensor:  Two-point calibration is sufficient -- since the
                   response is linear, checking at a low and a high
                   reference point is enough to correct any offset
                   and scale error across the whole range.
Humidity sensor:   Multipoint calibration is more appropriate --
                   because the response is non-linear near the
                   extremes, checking only two points would leave
                   inaccuracies uncorrected in the parts of the range
                   where the response curves away from a straight
                   line.

Matching calibration type to the sensor’s actual response shape, rather than defaulting to the same calibration method for every sensor, is the specific judgement this sub-topic tests.

Exam traps

• Confusing a sensor (takes a reading) with an actuator (produces physical movement/action).
• Naming a sensor without stating the specific property it measures.
• Forgetting control systems are a closed loop — the system must keep re-checking the sensor after acting, not act once and stop.
• Describing an application (smart home, burglar alarm) without breaking it into the sensor → processing → actuator chain.
• Mixing up which calibration type suits a linear vs non-linear sensor response.

Self-test

1. What is the key difference between monitoring and control?
2. Name the three calibration types and what distinguishes them.
3. Why does an infrared sensor suit a burglar alarm specifically?
4. What is the “closed loop” property of a control system, and why does forgetting it lose marks?
5. What three elements must a control-system flowchart show?

Answers: 1. Monitoring reads and records a value for a human or system to review; control uses a sensor reading to trigger an automatic response without waiting for a human decision. 2. One-point (checks against a single reference value), two-point (checks low and high range, adjusts offset/scale error), multipoint (checks several points for highest accuracy on a non-linear response). 3. Because it detects the heat given off by a human body without requiring contact, unlike sensors that need direct touch. 4. The system must continuously re-check the sensor and respond to changes, not simply act once and stop; forgetting this means describing a one-off action rather than an ongoing automatic response. 5. The sensor reading, a decision box comparing it to a threshold, and the actuator action as the outcome.