Welcome to installment No. 3 of Security Sales & Integration’s four-part series — “Digital Video for D.U.M.I.E.S.” (see page A2 of the March issue for Part 1, “Essentials of Digital Video Compression,” and page A2 of the July issue for Part 2, “Essentials of Digital Video Storage”).
This series has been designed to help educate readers on the fast and ever-changing world of digital video as found in the security industry. “D.U.M.I.E.S.” stands for dealers, users, managers, installers, engineers and salespeople.
This month’s chapter — brought to you by Dedicated Micros — explains what comprises a digital video recorder (DVR), and the many features available today in these popular devices.
Converting Analog to Digital
As previously discussed, compression and storage are the two areas most discussed when selecting a DVR. However, there are other important considerations when completing the transition from analog to digitally archived video.
The first, and often forgotten, area of digital processing is the analog-to digital converter (ADC). This device converts continuous signals, or voltages, into digital numbers. The number of discrete values the signal is divided into determines the resolution of the converter.
As an example, an 8-bit system encodes the analog input into 256 discrete values (28 = 256), whereas a 10-bit would offer 1,024 possible output values. The greater the bit structure, the greater the encoded resolution.
The analog video signal is continuous and, therefore, necessary for the ADC to convert this signal many times in order to properly reproduce the original image. The rate of the new values is referred to as the sampling rate of the converter. Just as in the bit structure, the greater the number of times a signal is sampled, the more exact the stored image.
One of the common problems found in the conversion process is what is commonly referred to as line crawl, especially when the video image is viewed in multiscreen displays.
When a video signal is changing faster than the sampling rate of the converter, the stored value will not be a valid number. As the process is reversed and the signals are converted back to analog, spurious signals, or distortions, are generated as a side effect of the digitized analog signal. This distortion is referred to as aliasing.
To help reduce this effect, the sampling rate of the converter should be higher than twice the highest frequency to be processed. There are many different forms of ADCs, all of which have the ability to reproduce quality images as long as their sample rate and bit structure can match that of the system’s inputs.
Managing DVR Resources
Another important aspect for any DVR, and in many cases the last to be recognized, is the operating system (OS) incorporated within the recording unit.
Along with the OS, the processing speed of the central processing unit (CPU) and the quantity of random access memory (RAM) also play important roles in overall system operation. The backbone of any DVR depends on this system to produce a reliable and functional recording system.
Anyone who has ever turned on a computer is aware of the OS — typically a Windows™ application. In fact, it is always the first and last thing anyone sees on the monitor screen. An OS must accomplish two basic things.
First, it must manage the hardware and software resources of the DVR. This includes processor speeds, memory, disk space, etc. Second, it must provide stable and consistent methods to manage all these tasks without anyone having to understand all of the details.
The most common operating systems found in today’s DVRs are Windows 98™, Windows XP™, Windows 2000™, Windows NT™, Linux™ and even a few manufacturer’s own propriety systems. For the most part, all the operating systems manage well, but each one has its own inherent advantages and disadvantages.
It is not this author’s intent to try and compare the pros or cons of each OS; we rely on the computer magazines for that!
Devices Are Memory-Hungry
Each process preformed within a DVR must have enough memory to execute all of its functions. The different levels of memory must be properly used so each process can run effectively.
Using a computing system like a DVR requires much more memory management than a normal computer used for everyday office work. Most computers using an older OS and slower CPU speed have a very difficult time maintaining desired parameters — like recorded images per camera, number of camera inputs, and access speed to stored information — required by the security industry.
Balancing DVR Processes
When applications begin, they are loaded into RAM in block sizes determined by the OS. RAM is considered “random access” because the user can access any memory cell directly if they know the row and column that intersects at that cell. RAM storage is only one of the memory types that must be managed by the OS.
Ranked by order of speed, the types of memory incorporated in a DVR are:
1. Cache memory — rather small in size but fast in speed, accesses data from RAM and stores it for the next operation by the CPU.
2. RAM — measured in megabytes and is the area that stores all the data for system operation.
3. Secondary memory — most often some sort of rotating magnetic or DAT drive storage that keeps applications and data available to be used.
The OS must balance the needs of the various processes with the availability of the different types of memory, moving data in blocks between available memory as the schedule dictates.
Since DVRs move vast amounts of data, many manufacturers have incorporated systems that monitor the available memory in both the cache and RAM sections. This is required because if the allotted memory starts to reach its capacity, the system will lock up. (Anyone who has worked with computers is all too aware of the “Insufficient Amount of Memory” pop-up screen.)
The system component used to monitor the memory is known as “The Watchdog” feature. In some cases, the user has the capability to set the parameters for cleaning the memory, while other machines on the market accomplish the task automatically.
The need to monitor memory is usually associated with older operating systems. As technology changes and more advanced operating systems are incorporated within the digital video marketplace, such system cleansing will no longer be necessary.
