Enterprise Video Surveillance System Design Guide: From Site Survey to Go-Live

Executive Summary

A poorly designed video surveillance system costs far more than a well-designed one — not in hardware, but in operational gaps, storage failures, and failed investigations when you need footage most. This guide provides a structured, engineering-led methodology for designing an enterprise IP video surveillance system from initial site survey through to operational handover.

We cover every design decision: camera placement, resolution selection, storage calculation, network architecture, VMS selection, and commissioning protocol. Bangladesh-specific guidance includes the Bangladesh Bank 90-day recording retention requirement, local cabling standards, and climate considerations for outdoor camera deployment in Dhaka and Chittagong.

Whether you are designing a 20-camera branch network or a 500-camera campus system, this guide provides the framework your team needs to deliver a system that performs reliably for years. SSAL has designed and deployed video surveillance systems across banking, healthcare, manufacturing, and government sectors throughout Bangladesh, and this guide reflects that accumulated expertise.

Chapter 1: Site Survey — The Foundation of Every Successful System

1.1 Threat Assessment

Every camera placement decision must begin with a threat assessment, not an equipment catalogue. A threat assessment asks: what are you protecting, against whom, and what does a successful security outcome look like?

Start by identifying your protected assets: cash vaults, server rooms, executive offices, stockrooms, loading docks, and parking areas all represent different threat profiles. Differentiate between insider threats — employees with authorised access misusing that access — and external threats such as unauthorised entry, theft, or vandalism. For manufacturing facilities in EPZ zones or RMG factories, cargo theft during loading and unloading is consistently the highest-risk event.

Document each threat scenario with a probability rating (low/medium/high) and an impact rating. This produces a risk matrix that drives camera priority and specification. High-probability, high-impact zones — such as a bank's cash counting room — justify higher-resolution cameras, redundant recording, and more frequent footage review. Lower-risk areas like car parks may accept lower resolution with extended retention.

1.2 Camera Placement Methodology

Camera placement follows a principle called overlapping coverage. Every critical area should be covered by at least 2 cameras from different angles, ensuring that if one camera is obstructed or fails, the second continues to capture the scene. For high-security areas such as vault doors and ATM machines, 3-camera coverage is best practice.

Define each camera's primary purpose before selecting its type. Entry/exit cameras capture faces of people entering or leaving — they require a narrow field of view aimed at face height (1.5–1.8 m). Overview cameras capture the entire floor plan of an open area — they require wide-angle lenses and higher mounting positions (3–5 m). Detail cameras capture specific objects such as cash counters or server racks — they require close placement and high resolution.

Use a scale floor plan to mark camera positions. Mark the camera's field of view cone on the plan using the manufacturer's lens angle data. Identify blind spots — areas where a camera's view is blocked by structural columns, shelving, or machinery — and add supplemental cameras to eliminate them. The final placement plan must demonstrate complete coverage of all high-risk zones with no uncovered entry or exit points.

1.3 Coverage Calculations — Pixels Per Metre (PPM)

The Pixels Per Metre (PPM) metric defines how many pixels in the camera's image represent one metre of width at the target distance. PPM determines whether a camera can fulfil its intended purpose.

The British Standard BS EN 62676-4 defines four performance levels:

• Detection (≥ 6 PPM): Detect that a human presence exists in the scene.
• Observation (≥ 25 PPM): Observe general body features and clothing.
• Recognition (≥ 62 PPM): Recognise a person previously known to you.
• Identification (≥ 250 PPM): Identify an unknown person from the image alone.

Calculate PPM using this formula: PPM = (Horizontal Pixel Count × Object Width Fraction) ÷ Object Width in Metres. For a 1920-pixel wide camera covering a 10-metre wide scene, the horizontal PPM is 192 — sufficient for observation and recognition. For ATM face capture at 1 metre with a 1920-pixel camera filling 80% of frame width, PPM exceeds 1,500 — well above identification threshold.

Always calculate PPM for each camera before finalising the specification. Cameras that fail to meet the required PPM for their stated purpose will not deliver usable investigation footage.

1.4 Lighting Analysis

Cameras need light. This is the single most-overlooked factor in CCTV design, and it is the root cause of the majority of unusable nighttime footage complaints. Conduct your site survey during both day and night to understand the actual lighting conditions your cameras will face.

Measure ambient light levels in Lux at each camera position. Most standard IP cameras perform well above 50 Lux. Below 10 Lux, standard cameras struggle; below 1 Lux, you need either IR illumination, starlight sensors, or thermal imaging. Bangladesh's outdoor environment presents specific challenges: monsoon overcast reduces daytime light levels, while interior areas in older buildings are routinely under-lit.

Identify strong backlight sources — windows, fluorescent shop fronts, vehicle headlights — that can silhouette subjects and make them unidentifiable. Consider cameras with Wide Dynamic Range (WDR) rated above 120 dB for backlit scenes. For car park entrance cameras, WDR is essential when vehicles enter from bright daylight into shaded interiors.

Chapter 2: System Architecture

2.1 IP Camera Types

Fixed cameras are the workhorse of any surveillance system. They are aimed at a specific area and do not move. Fixed cameras are available in dome, bullet, turret, and box form factors. Dome cameras blend into ceiling environments and resist vandalism; bullet cameras are suited for outdoor long-distance coverage; turret cameras offer flexible indoor mounting. For the majority of enterprise coverage requirements, fixed cameras deliver the best value per dollar.

PTZ (Pan-Tilt-Zoom) cameras can rotate horizontally, tilt vertically, and optically zoom, giving a single camera the ability to cover large open areas. PTZ cameras are best deployed to complement a fixed camera network rather than replace it. A fixed camera provides continuous evidence-grade recording; the PTZ allows operators to zoom into incidents in real time. PTZ cameras are cost-effective for large open spaces — parking lots, factory floors, outdoor perimeters — where a single PTZ replaces 10–15 fixed wide-angle cameras.

Fisheye cameras use an ultra-wide 180° or 360° lens to cover an entire room from a single ceiling-mounted camera. De-warping software within the VMS converts the distorted circular image into usable rectilinear views. Fisheye cameras are ideal for open-plan offices, retail floors, and foyers where full-room coverage from a single device minimises installation cost and ceiling penetrations.

Thermal cameras detect heat rather than light, making them effective in complete darkness, smoke, or fog. Thermal cameras are used for perimeter intrusion detection at fence lines and for detecting fires at early stages. They cannot read number plates or identify faces, so thermal cameras are always paired with standard optical cameras for evidence capture.

2.2 NVR vs. Server-Based Recording

A Network Video Recorder (NVR) is a dedicated appliance that receives video streams from IP cameras and stores footage on internal or attached hard drives. NVRs are cost-effective for small to medium systems — up to 64 channels — and require minimal configuration. Most NVR vendors (Hikvision, Dahua) offer ecosystem-locked NVRs optimised for their own cameras.

Server-based recording uses standard IT server hardware running Video Management Software. Server-based systems scale to thousands of cameras, support cameras from multiple vendors, and integrate more deeply with enterprise IT infrastructure. For systems above 64 cameras, for multi-site operations, or where advanced analytics and integration are required, a server-based VMS is the appropriate architecture.

Hybrid architectures combine NVRs at remote branch locations (where local recording provides resilience if WAN connectivity fails) with a central VMS server that provides unified monitoring, search, and reporting across all sites. This hybrid model is the standard architecture for Bangladesh's banking sector, where dozens of branches each require local recording with central oversight.

2.3 Bandwidth and Storage Calculations

Bandwidth per camera depends on resolution, frame rate, and compression codec. Use this formula: Bandwidth (Mbps) = Resolution (pixels) × Frame Rate × Bit Depth × Compression Factor. For a practical calculation, Hikvision's H.265+ codec at 4 MP resolution and 15 fps typically consumes 1.5–2.5 Mbps per camera under normal scene conditions.

Total system bandwidth = Sum of all camera bitrates. Add 20% overhead for network management traffic, VMS recording control, and peak-scene bitrate spikes. Ensure your network backbone and WAN uplinks are sized to accommodate this total.

Storage = Bitrate × Cameras × Seconds per day × Retention days ÷ 8 (bits to bytes) ÷ 1,073,741,824 (bytes to TB)

Example: 100 cameras at 2 Mbps each, 24 hours per day, 90-day retention:

• 100 × 2 × 86,400 × 90 = 1,555,200,000,000 bits
• ÷ 8 = 194,400,000,000 bytes = 194.4 TB raw storage
• Add 20% RAID overhead = approximately 233 TB usable storage required

Chapter 3: Camera Specification

3.1 Resolution Selection

Resolution selection must be driven by the PPM calculation from your site survey, not by a blanket policy of "install the highest available." Over-specifying resolution unnecessarily increases bandwidth consumption, storage requirements, and system cost. Under-specifying delivers footage that fails investigation requirements.

2 Megapixel (1920×1080, Full HD) cameras are appropriate for general interior coverage, corridor monitoring, stairwell observation, and low-risk perimeter areas. At 10 metres distance, a standard 2.8 mm lens 2 MP camera delivers approximately 80–100 PPM — comfortably above the recognition threshold.

4 Megapixel (2560×1440) cameras are the current enterprise standard, offering a 30% wider field of view than 2 MP at the same PPM. Use 4 MP cameras for entrance/exit doors, open-plan office areas, and anywhere you need both a wide view and enough detail for recognition. The bitrate increase over 2 MP is modest with H.265+ — approximately 20–30% higher.

8 Megapixel (4K, 3840×2160) cameras deliver identification-grade resolution across wide scenes. They are appropriate for cash counter areas, ATM vestibules, high-value stockrooms, and outdoor perimeters where you need to identify individuals across a 30–50 metre distance. 8 MP cameras require more bandwidth and storage; 4K-capable NVRs or server-grade GPUs for real-time decoding.

12 Megapixel and above cameras are specialist devices for large open areas — stadium concourses, warehouse floors, airport terminals — where a single camera must cover what would otherwise require 4–6 lower-resolution cameras. The cost per channel is high, but total system cost may be lower when including cabling and installation for the cameras they replace.

3.2 Compression Standards — H.265+

H.265 (HEVC) is a video compression standard that delivers the same image quality as H.264 at approximately 50% of the file size. H.265+ is Hikvision's proprietary enhancement that adds scene-adaptive bitrate reduction, achieving a further 50–70% storage reduction compared to H.264 by analysing static portions of the frame and compressing them more aggressively.

In practical terms, a 90-day retention system that requires 200 TB of storage on H.264 may require only 60–80 TB on H.265+. This translates directly to hardware cost savings — fewer or smaller hard drives — and ongoing electricity savings for drive operation. All enterprise systems designed by SSAL use H.265+ as the default recording codec.

The trade-off is that H.265+ decoding requires more processing power than H.264. Ensure that workstations used for live monitoring and playback have adequate GPU capabilities. Modern i5/i7 workstations with dedicated graphics cards handle 16–32 simultaneous H.265+ streams without difficulty.

3.3 Night Vision Technologies

IR (Infrared) illumination is the most common night vision technology. IR LEDs mounted around the camera lens emit infrared light invisible to the human eye but detectable by the camera sensor. IR cameras produce clear black-and-white images in complete darkness up to the range of their IR LEDs. IR range varies from 10 m (small indoor cameras) to 100+ m (long-range outdoor cameras). IR illuminators mounted alongside cameras can extend range further. The limitation of IR is the "headlight effect" — objects close to the camera are overexposed while the background is dark.

Starlight cameras use large-format sensors with high light sensitivity to produce colour images in very low light — as low as 0.001 Lux. Starlight cameras are preferred over IR for environments where colour information aids identification (clothing colour, vehicle colour) and where IR's black-and-white limitation is unacceptable. They perform well in street-lit outdoor environments and parking structures with some ambient light.

Thermal cameras detect temperature differences rather than reflected light, making them completely immune to darkness, glare, and camouflage. Thermal cameras are essential for fence-line perimeter detection across large outdoor areas. SSAL recommends pairing thermal cameras with starlight or IR cameras: the thermal camera detects intrusion and triggers the optical camera's PTZ to follow and record the intruder.

Chapter 4: Network Design

4.1 PoE Switch Sizing

Power over Ethernet (PoE) delivers both data and electrical power to IP cameras over a single Cat6 cable, eliminating the need for separate power supplies at each camera location. Correctly sizing your PoE switches is critical to system reliability.

Each camera draws a defined wattage. Standard PoE (802.3af) provides up to 15.4 W per port. PoE+ (802.3at) provides up to 30 W, required for PTZ cameras with heaters, multi-imager cameras, and cameras with built-in IR illuminators. PoE++ (802.3bt) provides up to 60 W or 90 W, required for high-wattage PTZ cameras.

Calculate the total PoE budget required: sum the wattage of every device connected to each switch. Do not exceed 80% of the switch's total PoE budget to allow headroom for startup surges. A 24-port switch with a 370 W PoE budget can support 24 cameras if average camera draw is below 12.3 W each — but a mix of PTZ cameras drawing 25 W will reduce actual capacity to 14 devices on the same switch.

4.2 VLAN Segmentation for Cameras

IP cameras are network-connected devices and must be treated as network security endpoints. Camera firmware vulnerabilities have been used in multiple high-profile cyberattacks globally. Isolating cameras on a dedicated VLAN prevents a compromised camera from accessing other network resources.

Configure a dedicated Camera VLAN (for example, VLAN 50) on all switches that carry camera traffic. This VLAN should have no routing access to your corporate data network. Camera streams route only to the NVR or VMS server. Management access to cameras for configuration and firmware updates should be restricted to a separate Management VLAN accessible only from the security operations workstation.

Implement 802.1X port authentication on camera switch ports where supported, ensuring that only authorised devices can connect to the camera network. Disable unused switch ports. Use DHCP with reservations or static IP assignment for all cameras so that IP changes alert you to unauthorised device swaps.

4.3 Bandwidth Planning

Design your network to carry 150% of your calculated camera bandwidth — the additional 50% provides headroom for analytics traffic, VMS management overhead, remote access, and peak-scene bitrate increases when complex motion is detected.

For multi-site deployments, calculate the WAN bandwidth requirement for each branch. If a branch has 20 cameras at 2 Mbps each, live monitoring from the central site requires 40 Mbps of WAN uplink from that branch. If live monitoring is limited to triggered cameras only, effective WAN bandwidth may be far lower. Configure Quality of Service (QoS) to prioritise camera traffic over general internet use on shared WAN links.

For very large campus systems or bandwidth-constrained WAN links, consider edge recording: cameras record locally to onboard SD card or local NVR, and the central VMS retrieves only flagged footage or provides remote live view on demand, reducing WAN consumption dramatically.

4.4 Cybersecurity for IP Cameras

Change all camera passwords from factory defaults immediately upon installation. Use strong, unique passwords for each camera and record them in a secure credential vault. Factory-default passwords are publicly documented; cameras left at default are routinely compromised within hours of connecting to any internet-accessible network.

Enable HTTPS for all camera web interfaces and disable HTTP. Disable unused camera services such as FTP, Telnet, and UPnP. Place cameras behind a firewall with outbound traffic restricted to the VMS server only — cameras should have no direct internet access.

Implement a firmware update policy: review manufacturer security bulletins monthly and apply firmware updates to all cameras within 30 days of release. Hikvision, Dahua, and Axis all provide automated firmware notification services. SSAL's Annual Maintenance Contracts include quarterly firmware reviews as a standard deliverable.

Chapter 5: Storage Design

5.1 Retention Period and Bangladesh Bank Requirements

Bangladesh Bank's ICT Security Guideline v4.0 mandates that all CCTV footage in bank premises be retained for a minimum of 90 days. This is among the most demanding retention requirements in the region and is the primary driver of storage sizing for the banking sector.

For non-banking enterprises, SSAL recommends a minimum of 30 days retention for general areas and 90 days for high-security areas such as cash handling, server rooms, and executive access areas. Retail environments benefit from 90-day retention to support investigation of inventory discrepancy claims that often surface weeks after the event.

Calculate retention storage requirements using the formula from Chapter 2. Always add 20% overhead to your calculated raw storage figure before specifying hard drive capacity, to account for file system overhead, pre-allocated spare space, and drive replacement time during which the system operates at reduced capacity.

5.2 RAID Configuration

RAID (Redundant Array of Independent Disks) protects footage against hard drive failure — the most common cause of data loss in NVR systems. Hard drives are mechanical devices with a finite lifespan; in a 24/7 recording environment, drives typically last 3–5 years.

RAID 1 (mirroring) duplicates data across 2 drives. If one fails, the other contains a complete copy. RAID 1 is simple and effective for small NVRs with 2–4 drives but has 50% storage overhead.

RAID 5 distributes data and parity information across 3 or more drives. Any single drive can fail and the system continues to operate; data is rebuilt when the failed drive is replaced. RAID 5 has approximately 33% storage overhead for a 3-drive array and scales efficiently with larger arrays. RAID 5 is the standard configuration for enterprise NVRs and recording servers.

RAID 6 adds a second parity stripe, allowing 2 simultaneous drive failures without data loss. RAID 6 is recommended for large storage arrays where the rebuild time after a single drive failure (which could exceed 24 hours for large drives) creates a window of vulnerability to a second failure.

Implement drive health monitoring (S.M.A.R.T. monitoring) and configure email alerts when a drive begins showing failure indicators. Schedule proactive drive replacement after 3 years of continuous 24/7 operation, even if the drive has not yet failed.

5.3 Cloud Backup

Cloud backup provides a secondary copy of footage at a geographically separate location, protecting against site-level events such as fire, flood, or physical destruction of the recording infrastructure. For Bangladesh banks, cloud backup aligns with Bangladesh Bank's requirements for offsite data protection.

Implement selective cloud backup: upload footage from high-priority cameras — vault areas, ATMs, entrance/exit cameras — rather than all cameras, to manage egress bandwidth costs. Configure retention policies on the cloud storage tier to automatically delete footage beyond the retention policy, managing ongoing storage costs.

For organisations without cloud backup agreements, a physical offsite backup using removable hard drives transported to a secure location weekly provides a minimum level of offsite protection. SSAL recommends encrypted drives for all portable backup media.

Chapter 6: Video Management Software

6.1 VMS Selection Criteria

Video Management Software (VMS) is the software platform that manages camera connections, records footage, enables live monitoring, and provides investigation and playback tools. The VMS is the operational centre of your surveillance system, and a poorly chosen VMS limits your ability to use the system effectively regardless of camera quality.

Evaluate VMS platforms on four dimensions: scalability (how many cameras and sites can it manage), integration (which access control, alarm, and analytics platforms does it support natively), usability (can your operators navigate it efficiently under pressure), and vendor stability (will the vendor still support this product in 10 years).

6.2 Milestone XProtect

Milestone XProtect is an open-platform VMS supporting cameras from over 13,000 device models. Its open architecture makes it the preferred platform for multi-vendor environments and complex integrations. XProtect offers tiered licensing — Essential (free), Express, Professional+, Expert, and Corporate — making it accessible for systems of all sizes.

XProtect's integration ecosystem includes access control platforms (Gallagher, Lenel, CCURE), alarm systems (Bosch, Honeywell), and analytics engines (BriefCam, Avigilon Appearance Search). For enterprise deployments with complex integration requirements, XProtect Corporate is SSAL's recommended platform.

6.3 Genetec Security Center

Genetec Security Center is a unified platform that combines VMS, access control, and ANPR (Automatic Number Plate Recognition) in a single software environment. Security Center's unified data model means that a door forced-open event in the access control system automatically retrieves and displays the associated camera footage on the operator screen without requiring manual lookup.

Security Center is particularly strong in multi-site enterprise deployments and is widely used by banking and government sectors globally. In Bangladesh, SSAL has deployed Genetec for clients requiring unified video and access control management across multiple branch locations.

6.4 Hikvision iVMS-4200 / HikCentral

For Hikvision-centric deployments, HikCentral Professional provides a comprehensive VMS with native deep learning analytics, access control management, and alarm management within a single platform. HikCentral is cost-effective for organisations deploying exclusively Hikvision hardware, as it eliminates third-party VMS licensing costs.

HikCentral supports up to 100,000 channels in a federated multi-server deployment, making it enterprise-scalable. Its web-based client allows access from any browser without software installation. The limitation compared to Milestone or Genetec is reduced third-party device support — non-Hikvision cameras require ONVIF connection and lose advanced analytics features.

Chapter 7: Installation Standards

7.1 Cable Management

All camera cabling must be installed in conduit or cable tray from the equipment room to the camera location. Exposed cable runs are vulnerable to physical damage, rodent attack (a significant concern in industrial and food processing environments in Bangladesh), and weather ingress for outdoor runs.

Use Cat6 UTP cable for indoor PoE camera runs up to 90 metres. For outdoor runs, use Cat6 outdoor-rated cable with UV-resistant jacket. Where cable must cross areas subject to electromagnetic interference — near motors, generators, or high-voltage switchgear — use shielded Cat6 (STP) with proper grounding of the shielding at one end only to prevent ground loops.

Label every cable at both ends immediately upon installation using heat-shrink labels or permanent markers. Label format should match the camera numbering scheme used in the design documentation. Maintain a cable schedule spreadsheet tracking cable number, origin, destination, length, and test result. This documentation is essential for fault-finding and future system modifications.

7.2 Conduit and Pathway

Rigid PVC conduit (20 mm minimum, 25 mm preferred) provides adequate mechanical protection for indoor camera cabling in commercial environments. For industrial environments with risk of impact damage, use steel conduit or rigid galvanised steel cable tray. In Bangladesh's high-humidity climate, use grey (exterior-grade) PVC conduit for all outdoor and rooftop cable runs, sealed at all penetrations with weatherproof compound.

Maximum cable pull distance without a junction box is 30 metres in straight runs; reduce by 30% for each 90-degree bend. Do not exceed 3 bends in a single run without an intermediate junction or pull box. Junction boxes must be IP65-rated for outdoor locations and must be accessible for maintenance.

All conduit penetrations through fire-rated walls or floors must be sealed with approved fire-stopping material immediately after cable installation. Failure to fire-stop cable penetrations can invalidate your FSCD NOC and create genuine fire risk.

7.3 Weatherproofing and Grounding

For outdoor cameras, ensure the IP rating matches the environment. IP66 (dustproof, high-pressure water resistant) is the minimum for exposed outdoor locations in Bangladesh's monsoon climate. IP67 (temporary immersion) is appropriate for ground-level cameras in flood-prone areas. IP68 (continuous submersion) is required for underwater or underground installations.

Camera housings should face away from prevailing weather where possible. Install drip shields above outdoor camera housings if they are mounted on walls subject to water runoff. Clean camera domes quarterly; Dhaka's dust and pollution accumulates on dome surfaces within weeks, visibly degrading image quality.

Ground all outdoor camera housings to the building's earthing system using 4 mm² green-yellow cable. Install surge protection devices (SPDs) at each outdoor camera's PoE connection to protect against lightning-induced transients. Bangladesh's lightning exposure is substantial, and lightning damage to unprotected outdoor cameras is a common service call.

Chapter 8: Commissioning and Handover

8.1 Testing Protocol

Commissioning begins only after 100% of physical installation is complete — cameras mounted, cables terminated, NVR/server installed, and network configured. A partial commissioning of a partially-installed system creates confusion and delays.

Conduct the following checks for every camera:

1. Power verification: Confirm camera receives power (PoE link LED active)
2. Network connectivity: Ping camera IP from VMS server
3. Image quality check: Review live image for focus, field of view compliance, exposure, and night vision function
4. PPM verification: Compare live image against design specification using a measuring reference placed in the scene
5. Recording verification: Confirm footage is recording to the correct storage location at the specified bitrate and frame rate
6. Playback test: Retrieve 5 minutes of recorded footage and confirm playback without corruption
7. Motion detection test: Trigger motion in the camera's detection zone and confirm event is logged

Document the result of every test for every camera in a commissioning checklist. Non-compliant cameras must be corrected before the commissioning record is signed.

8.2 System-Level Tests

Beyond individual camera testing, verify system-level functions:

• Failover test: Simulate a server failure (if redundant servers are specified) and confirm automatic failover within the specified time.
• Storage capacity test: Calculate current storage fill rate and project days of retention. Confirm this meets the specification.
• Remote access test: Connect from a remote workstation or mobile device and confirm live view and playback access.
• Analytics test: For systems with video analytics configured, trigger each analytic rule and confirm correct alert generation.
• Integration test: If integrated with access control or alarm, trigger each integration event and confirm the VMS responds correctly.
• UPS test: Disconnect mains power and confirm that all network equipment, NVRs, and servers remain operational on UPS power for the specified duration.

8.3 Operator Training

Even a perfectly designed and installed system fails if operators cannot use it effectively. Schedule a minimum of 4 hours of operator training for every system handover, covering: live monitoring, PTZ camera control, alert response procedures, footage export for investigations, and report generation.

Provide a laminated quick-reference card at each monitoring workstation with key procedures: how to search footage, how to export a clip, how to contact the system supplier in an emergency, and key system configuration information.

For complex multi-site VMS deployments, SSAL recommends role-based training: security manager training (full system administration), operator training (monitoring and response), and IT team training (system administration, network monitoring, firmware updates). Training records must be signed and retained as part of the system documentation package.

8.4 Documentation Handover

The documentation package delivered at handover must include:

• As-built camera layout drawings (PDF and DWG format)
• Cable schedule with test results
• Equipment register (serial numbers, firmware versions, IP addresses, passwords stored in sealed envelope)
• NVR/server configuration backup
• VMS licence certificates and activation keys
• Commissioning checklist signed by both the SSAL engineer and the client representative
• Operator training manual
• AMC contact details and escalation procedure

Store the documentation in both physical (hard copy) and digital (USB drive or client portal) formats. The system documentation is a critical asset — it is the reference used by future maintenance personnel and the starting point for any future system expansion.

Conclusion

A high-performing enterprise video surveillance system is the product of systematic design, not equipment selection. The methodology in this guide — from threat-based camera placement through PPM-verified specification, correctly sized storage, segmented networking, and structured commissioning — ensures your system delivers investigation-grade footage when you need it, stores it for the required retention period, and remains secure against cyber threats.

Bangladesh's regulatory environment, particularly Bangladesh Bank's 90-day retention mandate and the PDPO 2025's data governance requirements, makes professional system design not just a best practice but a compliance obligation. A properly documented, professionally commissioned system also provides demonstrable due diligence in the event of a security incident or regulatory inspection.

SSAL's engineering team brings over 16 years of group experience designing and deploying video surveillance systems across Bangladesh's banking, industrial, healthcare, and government sectors. We are authorised partners for Hikvision, Dahua, Axis, Avigilon, Milestone, and Genetec — and we recommend the right technology for your specific requirements, not the largest margin product.

About SSAL

Smart Safety & Automations Ltd. (SSAL) is Bangladesh's leading enterprise security, safety, and automation solutions provider. As a sister concern of Smart Technologies (BD) Ltd., we bring over 16 years of technology integration experience to security projects of every scale — from single-site SME installations to multi-city enterprise deployments.

Our security engineering team covers the full project lifecycle: security risk assessment, system design, supply, installation, commissioning, operator training, and Annual Maintenance Contract support. We hold authorised integrator status for the world's leading security brands and have successfully delivered projects for Bangladesh's major banks, garment manufacturers, healthcare institutions, and government bodies.

Ready to design your enterprise surveillance system?

Contact SSAL today for a free site survey and system design consultation.

• Website: www.smartsafety.com.bd
• CTA: [Get Free Consultation] [Talk to Our Expert] [Request a Quote]