CCTV System Design for Buildings — A Step-by-Step Guide with Calculations

Designing a CCTV system is both an engineering discipline and a practical art. A well-designed system delivers precisely the right coverage, image quality, and retention — without over-engineering the network, over-specifying storage, or overspending on unnecessary equipment. This guide walks through the complete design process with worked calculations, so that a facility manager, consultant, or procurement officer can independently verify vendor proposals or design a system from scratch.

1. The CCTV System Design Process

Professional CCTV system design follows a structured, sequential process. Skipping any stage — particularly the requirement gathering and site survey — invariably leads to a system that fails to meet expectations.

2. Understanding Requirements

Before selecting any equipment, the designer must answer these critical questions — ideally in consultation with the building owner, facility manager, and security team:

Purpose of the CCTV System

• Detection: Detecting that someone or something is present in the scene. Requires lower resolution; wider fields of view are acceptable.
• Recognition: Determining that the person seen is the same person seen previously (but not necessarily identifying who they are). Requires moderate resolution.
• Identification: Positively identifying a specific individual — typically requiring a face capture that fills at least 120 pixels per metre of height in the image (as per the international standard EN 62676-4). Requires higher resolution, appropriate lens selection, and correct camera positioning.

Different locations within the same building may have different objectives. An entrance face-capture camera needs identification-level quality. A corridor camera may need only recognition. A perimeter camera may need only detection.

Key Questions to Answer

• What is the primary purpose — deterrence, evidence, real-time monitoring, or all three?
• Who will operate the system — dedicated 24-hour security, part-time operators, or remote monitoring?
• What is the retention period — 30 days (commercial), 60 days (PSU), 90+ days (banks)?
• Is integration with other systems required — access control, fire alarm, BMS, visitor management?
• What are the regulatory requirements — RBI guidelines (banks), BIS/STQC compliance, CVC procurement rules?
• What is the realistic budget — and does it include 5-year AMC and storage expansion costs?

3. Site Survey

The site survey is a physical inspection that converts the abstract requirements into concrete design inputs. It cannot be done remotely from floor plans alone — the designer must physically walk every area of the building, at different times of day if possible, to understand lighting conditions and activity patterns.

Site Survey Checklist

4. Camera Placement Design

Camera placement is the most critical design decision. Every camera must have a clearly defined purpose, a calculated field of view, and an achievable cable route back to the nearest network switch.

Pixel Density — The Key Metric

The international standard EN 62676-4 defines surveillance quality in terms of pixels per metre (PPM) — the number of image pixels that represent one metre of real-world height at the target distance. This is the most objective way to determine whether a camera will deliver the required level of detail:

How to Calculate Pixel Density

Camera Mounting Heights

• Face capture at entrances: 2.2–2.5 metres height, tilted slightly downward, with a narrow field of view focused on the doorway. The goal is to capture every face at roughly the same angle.
• Indoor general surveillance: 2.8–3.5 metres (standard ceiling height). Higher mounting gives a wider overview but reduces facial detail.
• Corridors: 2.5–3.0 metres. Consider corridor mode (rotating the camera 90° for a 9:16 aspect ratio) to maximise coverage of long, narrow spaces.
• Outdoor perimeter: 3.5–5.0 metres to prevent tampering and provide a commanding view.
• PTZ cameras: 5–8 metres for maximum pan range and stability.

5. Bandwidth Calculation

Every IP camera continuously transmits video data over the network. Accurate bandwidth calculation is essential for sizing switches, uplinks, and recording server network interfaces.

Factors That Determine Bandwidth

• Resolution: Higher resolution = more pixels = more data per frame
• Frame Rate (FPS): More frames per second = more data. 15 fps is standard for surveillance; 25–30 fps for high-security areas
• Compression Codec: H.265 reduces bandwidth by ~30–50% vs H.264; Smart Codec variants reduce further
• Scene Complexity: A busy parking lot generates more data than a static corridor because the compression algorithm must encode more changes between frames

Typical Bitrate Reference Table

6. Network Switch Sizing

IP CCTV systems require two tiers of switches: edge switches (PoE switches on each floor that connect cameras) and core switches (that aggregate all edge switch uplinks and connect to servers).

Edge Switch (PoE) Sizing

For each floor or zone, calculate:

1. Port count: Number of cameras on that floor + 1–2 spare ports for future expansion
2. PoE budget: Sum of power requirements of all connected cameras (typically 12–15W per fixed camera, 30–60W per PTZ)
3. Uplink bandwidth: Sum of all camera bitrates on that switch must not exceed the uplink speed

Edge Switch Selection Guide

Core Switch Sizing

The core switch aggregates all edge switch uplinks and connects to recording servers, VMS servers, and operator workstations. Key requirements:

• Port count: Number of edge switch uplinks + server ports + management ports + spare capacity
• Switching capacity: Must be non-blocking — the total switching fabric must handle all ports at full line rate simultaneously
• Redundancy: For mission-critical deployments (banks, PSU HQ), use redundant core switches in a stacked or failover configuration
• 10GbE uplinks: Connections from core switch to recording servers should be 10GbE (SFP+) to handle the aggregate bandwidth of all cameras being written to storage simultaneously

7. Data Flow in an IP CCTV Network

The following diagram illustrates how video data flows from cameras through the network infrastructure to the recording servers and operator workstations. The animated data packets show the continuous flow of video streams through each network layer.

8. Storage Calculation

Storage is the most expensive recurring cost in a CCTV system and the component most frequently underestimated. A rigorous calculation prevents both under-provisioning (losing footage before the retention period expires) and over-provisioning (wasting budget on unnecessary hard drives).

Storage Calculation Formula

Worked Example

64 cameras, 4MP resolution, H.265 at 15 fps (average 4 Mbps per camera), 60-day retention, RAID 6:

• Daily per camera: 4 Mbps × 10.8 = 43.2 GB/day
• Total daily: 43.2 × 64 = 2,765 GB/day (2.77 TB/day)
• 60-day raw: 2.77 × 60 = 166.1 TB raw storage
• RAID 6 usable: 166.1 ÷ 0.67 = 247.9 TB physical disks required
• Add 10% file system overhead: 247.9 × 1.1 = ~273 TB total disk capacity

9. Recording Server Sizing

The recording server receives video streams from cameras (via the network), processes them, and writes them to storage. Correct server sizing ensures smooth recording without dropped frames or performance degradation.

Server Sizing Guidelines

Failover Server

For mission-critical installations (banks, PSU headquarters, data centres), a failover recording server provides automatic takeover when the primary server fails. The failover server monitors the primary server's heartbeat and, upon detecting a failure, immediately begins recording from the same cameras. When the primary is restored, recordings are automatically merged. Major VMS platforms (Milestone, Genetec, NUUO) support this natively.

10. Complete Design Example — 8-Storey Office Building

To bring all the calculations together, here is a complete worked example for a typical 8-storey commercial office building with a basement car park, ground floor reception, and perimeter security.

Requirement Summary

Bandwidth Calculation

• 80 × 4MP H.265 @ 4 Mbps = 320 Mbps
• 8 × 4K H.265 @ 10 Mbps = 80 Mbps
• 4 × PTZ 4MP H.265 @ 6 Mbps = 24 Mbps
• 4 × ANPR 2MP H.265 @ 3 Mbps = 12 Mbps
• Total: 436 Mbps + 25% margin = 545 Mbps design bandwidth

Switch Sizing

• Edge switches: 6 × 24-port PoE+ managed switches (one per two floors + one for basement + one for perimeter) — each with 2 × 10GbE SFP+ uplinks
• Core switch: 1 × 24-port Layer 3 managed switch with 10GbE SFP+ ports — provides VLAN segmentation, QoS, and connects to all edge switches and servers
• Total PoE budget: (88 fixed × 15W) + (4 PTZ × 60W) + (4 ANPR × 15W) = 1,320 + 240 + 60 = 1,620W across all edge switches

Storage Calculation

• Total daily: (80 × 4 + 8 × 10 + 4 × 6 + 4 × 3) × 10.8 = (320 + 80 + 24 + 12) × 10.8 = 436 × 10.8 = 4,709 GB/day (4.71 TB/day)
• 60-day raw: 4.71 × 60 = 282.5 TB
• RAID 6 + 10% overhead: 282.5 ÷ 0.67 × 1.1 = ~464 TB physical capacity
• Equivalent: approximately 29 × 16TB surveillance-grade HDDs (in RAID 6 across 2 storage shelves) or 24 × 20TB HDDs

Server Specification

• 2 × Recording Servers (48 cameras each): Intel Xeon Silver 4310 (12-core), 32GB DDR4, 2 × 10GbE SFP+ NIC, 2U rackmount
• 1 × Failover Server: Same specification as recording servers — activates automatically if either primary fails
• 1 × VMS Management Server: Intel Xeon E-2388G, 32GB DDR4, 1 × 10GbE — handles VMS licensing, system health, user management
• SAN Storage: 2 × 24-bay storage shelves with 16TB surveillance HDDs, RAID 6 configuration, iSCSI or Fibre Channel connectivity to recording servers
• UPS: 6 kVA online UPS with 15-minute battery runtime for all servers and core switch

Estimated Bill of Quantities (BOQ) Summary

11. Documentation, Installation & Commissioning

Design Documentation Package

A complete CCTV design should include the following documents before procurement:

• Camera Schedule: Tabulated list of every camera — location name, camera type/model, resolution, lens, mounting type, PoE requirement, and zone/purpose.
• Floor Layout Drawings: Architectural floor plans with camera positions marked, showing fields of view (coverage cones), cable routes, and switch/cabinet locations.
• Network Single-Line Diagram: Complete network topology showing cameras, edge switches, core switch, servers, storage, and workstations with port assignments, VLAN configuration, and IP address allocation.
• Storage Calculation Sheet: Detailed calculation showing bitrate assumptions, daily storage, retention period, RAID configuration, and total disk capacity required.
• Bill of Quantities (BOQ): Complete list of all equipment, cabling, accessories, and installation materials with quantities.
• Technical Specifications: Detailed specifications for each equipment category for inclusion in the tender document.

Commissioning Checklist

After installation, systematic commissioning ensures every component functions correctly:

• Every camera image verified: correct coverage, focus, day and night quality
• Every recording channel confirmed: continuous recording, correct retention
• PoE delivery verified on every switch port
• Network performance tested: no packet loss, acceptable latency
• Analytics (if configured) validated under real conditions
• Remote access tested from outside the building network
• UPS tested: simulate power failure, verify graceful shutdown/switchover
• Failover server tested: simulate primary failure, verify automatic takeover
• Operator training delivered and documented
• As-built drawings and documentation handed over