CEO of AxxonSoft assesses why and how distributed video surveillance systems are unified

A distributed video surveillance system uses multiple cameras and other sensors to monitor different areas or facilities, writes Alan Ataev, CEO, AxxonSoft.

As such, a system can be made up of thousands of cameras and sensors spread across hundreds of sites. Unifying them to transmit video and other data to a central location for processing and analysis can be a complex task…

The reasons for unification

A reasonable question is: Why unify? Why not let the systems at different sites work independently and use local operators to monitor and manage each one? There are various reasons for unification. Here are some of the most common:

• Certain businesses, such as some retail chains, have multiple small to mid-size facilities with a few cameras and having an operator at each location is economically impractical. Moreover, some facilities, such as cell towers, cannot allocate space for the operator

• Many tasks, such as incident investigations or business-process monitoring, are more appropriate to handle at the organisational level, not at a single facility. Here, remote access to video footage and other collected data becomes imperative

• Critical facilities, such as power plants, may require centralised command and control combined with local monitoring rooms for maximum security.

• Large facilities with numerous areas and buildings may need to aggregate data to see what is happening across an entire territory

Unification tasks

What capabilities should a unified video surveillance system deliver to ensure efficient monitoring and swift response to alarms? Remote viewing of live and recorded videos (from one site at a time) is only a basic function. When it comes to setting up monitoring centres, different cameras in various locations must be viewed simultaneously.

The same applies to alarms: An operator should be able to receive alarm events and related videos from multiple sites in a synchronised manner without the need for switching. An advanced feature of unified systems is multilevel alarm management. This is used to establish a hierarchical structure of monitoring centres, for instance, for a state-wide gas station or retail chain. This functionality provides capabilities such as escalating critical alarms and reporting on alarm processing to monitor the operators’ work.

Real time monitoring that alerts an operator to any hardware, software or network issues is crucial for large installations, especially when staff are not constantly present at every facility (like a cell tower). Centralised reporting based on video analytics (alarms, visitor counts, queue lengths etc.) and system health parameters are essential for optimising business processes and maintaining the system. Finally, when multiple users work with different cameras, reports and other functionalities, it is imperative to have flexible user rights management as well as storage of individual user settings.

Technical implementation: Cloud

In general, the approaches to deploying video surveillance solutions fall into two major categories: Cloud-based systems or video surveillance as a service (VSaaS) and on-premises systems.

VSaaS systems are distributed and unified by nature. Video and metadata are streamed directly from cameras or via bridge appliances to the cloud, where all the data are processed and stored. No matter where cameras are located, the only requirement is a stable internet connection with sufficient bandwidth.

To view videos, users connect to the cloud via the internet, mostly using web browsers and mobile apps. Most cloud services provide alarm notifications. Some offer health monitoring, which is typically limited to alerting an operator to the loss of a video feed. Depending on the provider, VSaaS may also deliver user rights management, video analytics and reporting.

While cloud-based video surveillance is quite convenient for most SMB and home users and is the easiest way to monitor multiple distributed sites, it has significant constraints for enterprise applications. Camera numbers at the sites are capped by upstream internet bandwidth. Health monitoring is usually basic and cannot identify the cause of a failure; it also cannot monitor the full set of parameters and resource utilisation in real time and it does not provide corresponding reports. User rights management is not designed to handle hundreds of users with different access levels.

Finally, video analytics, even when using advanced technologies like AI, are very limited in terms of customisation, as are the reports. It’s also worth noting that many large enterprises impose strict security requirements, requiring that video surveillance work over closed-circuit networks, which makes VSaaS not applicable.

Technical implementation: On-premises

On-premises video surveillance systems use servers with installed VMS or NVRs that receive, process and store video and metadata from cameras and other sensors.

Several VMS servers can be joined into a single system with a common database for events, users, settings etc. The system can be designed in various ways, depending on the VMS developer. For example, each server may have its own copy of the database, synchronising it with the other servers peer-to-peer or via a sync server. Also, a dedicated server (or group of servers) may be allocated for storing the database. In any case, a stable network connection between servers is necessary to transmit events and sync databases.

When the VMS servers are in different remote facilities with no dedicated communication lines between them, it is not possible to make all the servers operate as a whole. Any network issues may cause system malfunctioning, and in addition, issues may occur when increasing the number of servers.

A combined approach

Is there a solution for unifying large, distributed on-premises video surveillance systems? Yes, and this is the combined approach. This approach assumes that every system at each location is independent and self-contained, with its own database, local monitoring etc.

A dedicated software suite, usually hosted in the monitoring hub, handles all the tasks of unification listed above. It stores user settings and implements user rights management, provides user connections to remote systems and collects data for health/alarm monitoring and reporting. These functionalities may also be orchestrated via a hybrid cloud that combines on-premises servers with a private cloud–based monitoring service.

This approach offers multiple benefits. It offers ease of access to any site combined with the security, control and resource customisation of an on-premises infrastructure. A variety of client applications are available to users, including desktop clients with all their advanced features. Any number of users can work with the system using individual GUI settings and in accordance with their access levels.

The solution can work over private networks (virtual or physical), while video footage is stored on-premises. This is crucial for organisations with strict security standards. Network issues or bandwidth limitations between the sites and the monitoring center do not affect the operation of local systems. Finally, such solutions are primarily developed for enterprise-scale applications. Therefore, they provide a flexible combination of local and central monitoring capabilities, customisable reporting and advanced alarm handling.

One of the main challenges of unifying a distributed video surveillance system is ensuring that the data from the cameras and sensors are transmitted effectively and efficiently to the central location. Therefore, network infrastructure is an instrumental factor in implementation. Networks need to be fast and reliable enough to handle data transmission and minimise latency. If not all connections are of equal quality, one can consider specific solutions that support low-bandwidth links.

The user interface is another significant consideration in the implementation of a distributed system. It should be convenient for working with multiple cameras and sensors while providing real time access to data and videos when necessary. It is worth looking at solutions with specially optimised interfaces for large installations.

The conclusions

In today’s world, more and more businesses and public institutions are realising the benefits of unifying distributed video surveillance systems. A unified system can improve situational awareness and enhance security while being easy to use, scalable and cost-effective.

Increasingly popular cloud-based solutions (VSaaS) offer a convenient way to get many of the benefits of unification. However, many large organisations still prefer on-premises systems that better comply with their security policies and customisation requirements. Purpose-built software suites provided by enterprise level VMS developers and hosted on-premises or in the private cloud help such organisations obtain high scale video surveillance solutions that meet their needs.

This article was originally published in the February 2023 edition of International Security Journal. To read your FREE digital edition, click here.