One of my goals of the New Year is to include more access control coverage in this column. In the past, I have focused on general business issues and where a specific product line has been singled out as an example, it’s pretty much been CCTV. This parallels my own experience, naturally. You tend to talk (or write) about subjects that are the most familiar to you and I’ve been involved with video far longer than access control.
However, at the enterprise level there is an equal reemphasis on controlling entry and egress at facilities, and we want to be sure we are addressing all areas of concern for our readers. Because I often use events that are happening in my consulting practice as topics for this column, the timing couldn’t be better. We are currently involved in several major access control projects and no doubt facing the same sort of issues that commonly affect you. In fact, an issue that we’ve seen with three different clients in as many weeks brings us to this month’s column.
Feed Alarms Into the CCTV System
The three projects I speak of are all integrated systems. They include CCTV cameras, recording and control, access control and alarm point monitoring. As is customary with this type of system, two of the three projects have some level of redundancy in that both the CCTV and access control head-ends are capable of controlling and responding to a great number of alarm points. The third project has a separate alarm processing system, since wireless alarm switches are used. There are two types of alarms: duress (panic buttons) and door switch.
One of the first things we do is determine the best way of handling each type of alarm and ensure that all alarms are fed back to the CCTV system in some way. This allows the operator to visually identify the cause of the alarm (where possible) and respond, accordingly. There are many choices, including alarm interfaces on the CCTV head-end, alarm contacts on the pan/tilt/zoom (p/t/z) dome cameras and contact on the access control panels.
In the past, such connections were simple. Any door switches would have two sets of contacts and would be wired to both the access control system and the CCTV system alarm interface.
On rare occasions, dome alarm contacts were used, as it requires bidirectional communication from the dome, which isn’t as reliable as alarm point monitoring requires. Any duress alarms would also be run to the CCTV alarm interface or, if needed, a separate alarm annunciator panel.
This method is still common and has the benefit of being simple to implement, extremely reliable and easy to troubleshoot. The drawbacks include expense (it’s a lot of extra wire to run in a large facility) and, unless carefully programmed, duplicate alarms as both systems respond to the same event.
High-level Interfaces Save Money
An intermediate step taken by some integrators has been to connect the CCTV system alarm interface to dry contacts, provided for that purpose on the access control system panels. This can be effective as well, since it eliminates the need to run a duplicate set of wires to each door.
Most access panels can be programmed to respond when there is an actual alarm event (door is propped open, bad card read) rather than every time the door is opened. Of course, doors that are not controlled through the access control system need to be accommodated some how, as do duress alarms. Drawbacks still include expense; you are still wiring alarm points in two directions.
Over time, many integrators have moved to a high-level interface in which the access control system “talks” directly to the CCTV system and sends alarm information directly from one brain to another.
This is a very cost-effective alternative: alarm points only need to be connected to one system and a single communications cable sends all of the alarm information from one system to another. Most CCTV manufacturers have fairly robust RS-232 ASCII communications protocols, and access control manufacturers have developed drivers for the various systems, allowing for great interoperability. The only work involved is in programming one system to send the right commands and ensuring the other system is listening and responding.
In many cases, end users and integrators have taken this to the next level, relying on the access control system to handle alarms and auxiliary doors through alarm panels, while also serving as an alarm monitoring system. In fact, as more solutions eliminate the need for a CCTV matrix switching system in favor of clusters of DVRs or Internet protocol (IP)-based video systems, the access control system is often the only centralized system on the premises to handle alarms. While this may be cost-effective and expedient, it creates a system vulnerability that is often missed until it is too late.
Be Wary of Distributed Intelligence
Access control systems rely heavily on distributed intelligence. Panels that are distributed around a building are empowered to make entry decisions, even (or especially) when they are offline and not communicating with the server. This minimizes network traffic and allows a server to control multiple buildings around a campus or around the world without continuous connectivity.
In fact, if the server fails, the access control system is designed to keep working. Contrast this to a CCTV matrix switching system: if the brains of the system go down, the entire system is generally useless.
This works very well for the primary function of an access control system — providing secured entry to a building or room — but can be a disaster when the system is relied upon for alarm monitoring. If the server fails for any reason, the panels will not have anywhere to send the alarms and will usually store them until they are again able to communicate with the server. Since the system remains functional when the server is down, there is no way to notify people that their panic buttons have stopped working and alert them to pick up the phone.
While server failures may not be common, there are still enough reasons for the server to be down — software upgrades, maintenance or other problems — that this should be a consideration. It is for this reason we urge our clients to wear both a belt and suspenders.
Critical alarms should be wired to both systems, or a separate alarm-monitoring panel should be installed in cases where no centralized CCTV system is present. Certain alarms, such as duress alarms, must absolutely, positively go through in real-time and do so reliably. While an access control system is an excellent alarm monitoring point, it is no substitute for a dedicated system in cases where lives or personal safety are on the line.
User Knowledge Is Key
Education is also critical. If there is a scheduled upgrade, or the server will be down for any length of time, let people know. The server should be in a place where people will see it and the monitor should never be turned off. A dark screen should mean immediate attention and not that the monitor has gone into “sleep” mode. By understanding when the system is vulnerable, areas that may be impacted can be put on alert and you can otherwise compensate until full functionality has been restored.
Integrated systems with high-level methods of sharing information are clearly a benefit to our industry and will continue to gain in popularity. But understanding system limitations and designing safety nets around them is a critical role that the consultant, integrator, security director or other end user must play.
After all, nobody wants to get caught with their pants down!