The steps of designing for digital
Aug 1, 2000 12:00 PM
Charlie R. Pierce
The average security director or CCTV design person will face many challenges when making the digital plunge. Changes within our industry suggest a need to change your strategy relating to CCTV.
Ten years ago, we thought a CCTV system with 25 to 50 cameras was quite large. Today, we are dealing with systems that contain hundreds or thousands of cameras within the reach and control of a single operation system. Consequently, the interactive design of such systems becomes especially important. Not even thinking about budget, let's just take a look at the logistics of dealing with the incredible onslaught of visual information being obtained.
We'll address some simple questions and then go from there.
Question 1: How many sections can I divide my system into? It's important to understand that there is no way to design these new mega-systems without first breaking them down into smaller, individual systems. We can then tie the individual systems together into groups and tie the groups into the whole. Such an approach divides the overall project into manageable smaller pieces.
Question 2: How many locations do I plan to incorporate into my overall system? Five years ago, the idea of handling a multitude of CCTV systems from a single point several miles - or several hundreds of miles - apart was a nice dream. Today, however, with the introduction of digital transmission and management systems, it is a reality. When there are several locations, we must design each system individually, according to the needs of each location, and then design the controlling system, or central station, in an efficient, automated manner.
Question 3: How many people do you plan to employ to monitor, maintain, and/or operate your new system? Consider the following points:
n The average person can only monitor up to four images simultaneously.
n The average person can only watch an average of 30 to 45 minutes of continuous black and white imagery before he loses complete comprehension of what he is looking at.
Apply these points to the logic of your new super-CCTV system and you'll be hit between the eyes. Right from the start, we must design, plan and expect our new digital super-system to be almost completely self-managed. It must be able to determine what to record, when to record, what to show on a screen and when to show it. Our system must be able to sound an alarm on its own, and to assist the operator in deciding what is real and what is a false alarm. We must design response criteria in ways that were never imagined just 10 years ago. If we do not perform this planning in the beginning - prior to a single equipment choice - our net result won't be a comprehensive security system but a confusing maze. We do it but we can't do it alone. Here, we must sit down and literally design the entire aspect of each piece of equipment within the system:
n Camera 1 Location - central lobby Purpose: to obtain an image of people entering and leaving between the hours of 7 a.m. and 5p.m., Monday through Friday. To obtain general images - for identification - of all persons who enter through the front lobby area at all times.
Now, the key is to incorporate camera 1 into the whole security process, automate it, and make it useful before we add a second camera. The first lunge is to place and aim the camera according to the need of the application. Since our purpose is to obtain general activity during a first shift and specific identification at other times, we know we must aim the camera either toward the main entrance or into the lobby from the main entrance.
Since our lobby area is large, it is impossible to have both general activity and identification-level activity (at the door) handled by a single camera position. Therefore, in this case my advice is to divide the responsibility between two cameras.
Camera 1 is located above the door, looking into the lobby area (general activity observation) while camera 2 is aimed at the door (specific identification information about those entering the lobby).
With the objectives and position of each camera in mind, camera 1 should not require much in the line of special features, while camera 2 will require some form of backlighting or electronic iris so that it can look at the bright lobby entrance and still see the frontal detail of those individuals entering or standing in the lobby area in front of the door.
And so the process is started. Now we start to create our various shopping and organizational lists. These will include a camera list, action list, interaction needs list and others. In such a way, we keep track of various features, actions, and interactions when it comes time to choose our equipment and to test our system on paper. Note that in the camera list (at left), camera 1 has no special features while camera 2 must have back-light compensation features.
The next step is to determine the interactive responses that will be needed to make these cameras as automated as possible. Sure, if they were the only two cameras in my system and I had a security guard on duty 24 hours a day, I could put up a monitor, multiplexer, and a recorder of some sort and call it a day. However, these are the first two cameras of a hundred or a thousand, so I must automate them as much as possible. So let's look. Monday through Friday from 7 a.m. to 5 p.m., we will record at a rate of 21 pictures per second. At all other times, we will assume the lobby will be empty and record information only when there is activity. We will track these operations on our activity sheets (see above right).
Having begun our activity listing, we can now start our controlling system requirements list (at right). It is not difficult, and it would be redundant, if we added something from each camera in the system. However, if we build a shopping list as we design based on each new system, then we end up with a solid basis to compare specification sheets when it comes time to pick equipment.
It is really simple, isn't it? Of course it is. That's because we are organized. Now, however, we start to enter into the tricky aspects of the system. We know the purpose and the trigger of the camera. But, we don't know the proposed result other than recording images of activity. Are there other times that camera 1 and/or 2 may come into play? I think there may be. However in order to see them, we must proceed to Camera 13, which is located in a very sensitive area in the center of our main building. The purpose of the camera is to monitor specific activities at the filing system against the south wall. We have it triggered by digital video motion detection (built into the camera, since it is in fixed lighting). The hours of active monitoring for this camera are from 4 p.m. to 5 a.m., seven days a week, provided that no card access device was used to enter the area during the specified time.
In the event of an alarm, within specified hours and conditions, we want the following to happen: Camera 13 to become visible on the guard's main screen. Camera 1 and 2 are to start recording in real time and trigger onto a secondary screen if there is any activity during the alarm created by camera 13. Additionally, the system should be set up so that the guard can push a single button and have camera 1 and camera 13 swap positions between the master monitor and the secondary monitor. I also want the access system to go into lockout to all but a few specified, emergency, full-clearance cards belonging to my top clearance security and/or management people. We also want all of our exterior cameras, monitoring drives, parking areas, outside walkways, to become sensitive to motion detection and to record any/all such activity (during the alarm) and to notify the guard of such activity through a visual screen. Additionally, I want my perimeter auto-domes to begin auto tracking of any movement in their areas. Here is the catch: we only want the exterior cameras to trigger in a logical order. There is no need for the west end parking lot to suddenly appear on the screen if there was no activity (visually) between this lot and the secured room. I don't want to distract my guard. However, if the person who triggered the alarm walks from the secured room through any visual field in my system, I want the system to automatically track the activity and keep it on the master monitor. This way, the guard can concentrate on responding to the alarm, versus sitting in front of a screen and pushing a lot of buttons.
As you can see, it doesn't take very long for a system to become complicated and hard to track. This is where we start our "Logic Alarm Response Form" (see example above). With our various lists and logic alarm response system, we will be able to trace all activity according to the needs of the application.
Additionally, and more importantly, we will be able to test our entire system, on paper, prior to the purchase of a single unit or piece of equipment.
If nothing else, you should now realize that there is a massive amount of planning needed for today's systems. The difference is that the layout and choice of the equipment mean very little if the forefront isn't done first. With modern-day, digital systems, you will spend 10 times longer designing how you want your system to respond than you will installing it, learning it, or even purchasing it. With the myriad of logical and illogical possibilities available to the average designer and/or security supervisor, is it any wonder that there is so much fear of digital? Well, slow down, use your head. Remember to plan from the beginning and use common sense to organize yourself. If you can imagine it, and can afford it, we can do it. A caveat here about the company you hire as your consultants. Don't ever be afraid to ask to see the organizational sheets and charts. Why not?
If done right, the system has required almost as much time for design as for the actual layout.
Lastly, test your entire system under every possible, conceivable circumstance on paper!
Don't buy a single piece of equipment until you know it works on paper. Write your security response plan as you go (it will save a lot of repeat time later). Ask for demonstrations. And in the very end, good luck.