CCTV Systems and Video Surveillance Fundamentals
This domain tests whether you understand how a camera, lens, encoder, recorder and storage chain together produce a picture - and which settings limit what that picture can prove.
The signal chain
Camera → transmission (coaxial, UTP, fibre, or PoE over IP) → recorder (a DVR for analogue/HD-over-coax, an NVR for IP) → storage → client workstation. Every link can degrade evidence: a poor crimp, an inadequate PoE budget, a failing HDD, or a downscaled substream.
Pixel density, not megapixels
Usefulness is set by pixels-per-metre (px/m) on the target, per IEC 62676-4 guidance: Detect ≈25 px/m, Observe ≈62.5, Recognise ≈125, Identify ≈250. Worked example: a 1080p camera (1920 px wide) covering a 20 m wide field of view gives 1920 ÷ 20 = 96 px/m - enough to recognise a known person's clothing, not enough to identify a stranger.
Settings an auditor must record
- Resolution and lens (2.8 mm wide vs 12 mm tele), field of view.
- Frame rate (commonly 12–25 fps), bitrate, codec (H.264/H.265), keyframe interval.
- Recording mode: continuous vs motion/schedule - gaps matter.
- Night performance, IR, WDR, shutter, privacy masks.
- Time source: NTP server, time zone, DST, measured drift.
- Retention and overwrite policy.
Storage arithmetic
4 Mbps × 86,400 s ÷ 8 = 43.2 GB per camera per day. Sixteen cameras ≈ 691 GB/day; 30 days ≈ 20.7 TB. That tells you whether the promised retention is physically possible.
Traps
Confusing DVR and NVR; quoting megapixels as proof of identification quality; assuming audio is recorded; ignoring clock drift; assuming the file you were shown is the original.