A Digital Video Recorder or DVR is the component of your video
surveillance system that takes care of saving the video that your cameras capture,
storing it digitally and then allowing you to play it back. Digital video recorders
also let you search your saved video, control your PTZ cameras, access your
system offsite, integrate your video with your alarm and access control systems
and so on. If your security cameras are the eyes of your security environment,
the DVR is the brain.
Digital video recorders are very complex pieces of equipment with many, many
internal components which must work well together in order to provide a functional,
seamless, stable CCTV solution. In spite of their complexity, however, a quality
DVR should be easy to use. You should be able to forget about your DVR until
you need it. A quality digital video recorder isn’t something you should
have to worry about on a day-to-day basis.
Because of their complexity and the broad range of prices, quality levels and
feature sets, purchasing a digital video recorder can be an intimidating process.
A knowledgeable CCTV security integrator/dealer can go a long way toward helping
you select a DVR that will fit your needs and your budget.
The intent of this white paper isn’t to make you an expert on digital
video recorders or CCTV system installation. What it will do is help you to
become an informed consumer. Understanding the basics of a CCTV video surveillance
system will allow you to ask the right questions to insure you are getting a
video security system that will work for you.
In this white paper, we’ll look at the modern digital video recorder in
some detail. We’ll cover the DVR from a functional standpoint and then
examine the hardware and the software that make a DVR an easy to use, functional
part of your enterprise security environment.
II - The Digital Video Recorder
A. What Is a DVR?
In it’s most basic form, a digital video recorder is a device which captures
video signals from one or more cameras, stores it in digital format on a hard
drive and allows you to view it and play it back on an attached monitor.
1. The Old Technology
The first video surveillance systems rose out of VCR (video cassette recorder)
technology. It was suddenly possible to inexpensively record video to cheap
video tapes. By today’s standards, these old VCR systems are rather limited
but at the time they were revolutionary.
2. Rise of the DVR
With the advent of computers, it became theoretically possible to capture video
and store it digitally. As processor speed, available RAM, hard drive size and
video compression technology improved, digital video capture and playback became
a reality. Suddenly the video that could be captured was high-quality, high-resolution
and full-motion unlike the low-quality, low-resolution 1 frame per second video
that a VCR based system typically provided. In addition, DVR-based video could
be stored, searched and accessed from anywhere in the world. At the same time,
the price of the technology fell to the point where even small business could
afford to install a powerful, state-of-the-art video surveillance system without
breaking the bank.
Today, digital video recorders are continuing to improve, with new features
being added all the time. With high-bandwidth connections readily available,
modern DVRs can be accessed and controlled from anywhere in the world via the
Internet. Reliability has also greatly improved.
The future of the DVR will see more and more integration with other security
and related information technologies including burglar alarms, fire alarms,
environmental sensors, access control systems, mobile guard stations, transportation
security, production line monitoring as well as asset, personnel and inventory
tracking and management. In fact most of these things can be integrated together
today, though it takes an experienced security company to bring the pieces together
into a cohesive whole.
3. Components of a DVR CCTV System
A CCTV system generally consists of the digital video recorder itself, a monitor,
a keyboard with an (optional) built-in joystick, various types of cameras and
lenses, camera mounting hardware and enclosures, a power panel for the cameras
and video, power and telemetry cable and connectors. Optional components can
include special cabinets for rack mounting, locking DVR cabinets, audio cable,
NVT receivers and transmitters, an uninterruptible power supply (UPS), a CD-ROM
/ DVD for archiving, hot-swappable hard drives, RAID arrays, POS text insertion
devices, relays, speakers and microphones, printers, visible and audible alarms,
conduit, tie wraps, fasteners, other installation components, etc.
For more information on security cameras please see i-ViD’s white paper
entitled Security Cameras.
For more information on cable and connectors used in security installations
please see i-ViD’s white paper entitled CCTV Cables and Connectors.
Additional information, including several other white papers and FAQs can be
found on i-ViD’s web site at www.ivid.co.uk.
4. Components of a DVR CCTV System
The DVR itself is made up of many different components. We’ll briefly
list each component here and go over them in some detail a little later on.
A digital video recorder is, at it’s heart, a computer. As such, it has
all the pieces you’d expect a computer to have: an operating system, a
file system, a CPU, RAM, one or more hard drives, a video card (for display
to a monitor), a motherboard, a PCI bus, a modem, a network card (NIC), etc.
Where a DVR differs from an off-the-shelf PC is in it’s inclusion of a
video capture card (or cards) and the level of engineering that goes into insuring
that all of the components work together to provide high-quality, reliable video
viewing, storage and retrieval.
5. Enterprise vs. Economy DVRs
While it’s possible to buy a cheap computer, install a video capture card
and hook up a few cheap cameras and call it a DVR, it won’t work well,
especially for a system with more than two or three cameras. It will be unreliable
and provide very poor quality video. There are hundreds (maybe even thousands)
of "DVR manufacturers" and this is precisely what many of them do.
They buy a cheap computer, call it a "server", slap in a video capture
card and some canned software and call it a DVR. It’s not.
Professional quality digital video recorders are carefully engineered so that
each hardware and software component works well with every other component.
Poorly designed DVRs run into problems when certain components are asked to
do too much (like the video capture card also being asked to display live video).
If the system is stressed too much, it will slow down or even freeze up. It
only takes one badly designed component to create a bottleneck and ruin system
performance.
You will run into many problems with "economy" DVRs, the worst of
which are reliability (it will crash and/or hang quite often) and video quality
(your video will be poor, erratic or just won’t record at times). Some
of the other problems you will run into include:
Inability to operate with some camera types
Poor PTZ control
Storage times may be much less than you were led to believe
Only a few users can access the system at one time without freezing the DVR
Warranty support is unavailable
Upgrades are unavailable
The system has serious security problems
It’s true that digital video recorders are essentially computers. They
are, but they differ from your garden variety PC in the same way that a top
fuel dragster differs from a passenger car. Every component of a professional
quality DVR is specifically chosen to do one thing and to do it well. You can
find £500 DVRs but you will get what you pay for. We speak from experience.
We could choose to rep these "entry-level" systems but we don’t.
Our reputation means too much to us to offer inferior equipment to our clients.
6. Standard Features of a DVR
First let it be said, there is no standard DVR. Every digital video recorder
is different in features and quality but most DVRs will have some features in
common. What follows are the most common features that should be found in most
any commercial grade DVRs.
Real-time viewing
Multi-user viewing
Multiple viewing modes (1, 4, 9, 16, 32 cameras)
Full-motion (30/25 fps) viewing
Video recording
Selectable recording frame rates
Selectable recording resolution
Selectable recording schedules
Overwrite or halt recording upon full disk
PTZ (pan/tilt/zoom) camera control
Event (motion detection) recording
Time-lapse recording
Video playback
Fast forward / rewind
Video search by event
Video search by time
Remote live video viewing
Printing support
1 channel, simplex audio recording
Password protection
Video loss alarm
7. Extended DVR Features
What follows is a list of just some of the "advanced" features found
in many digital video recorders. It is unlikely that you will find a DVR with
all of these features. It’s important to decide what you need and what
you don’t and then communicate this to your security dealer so that you
get the DVR that’s right for you.
Selectable recording compression rates
Resolution selectable by camera
Frame rate selectable by camera
Hot-swappable hard drives
CD-ROM / DVD recorder
Alarm inputs / outputs
Email / pager / SMS notification
Remote video playback
Remote PTZ control
Automatic video archiving
Redundant storage options (RAID)
Multi-channel duplex audio
POS text insertion
Email video w/ embedded player
PTZ presets
PTZ auto-scan
Video recording directly from client software
Administration and user event logging
Synchronized audio playback
Viewing of multiple cameras from multiple DVRs in the same screen
On-the-fly administration
Built-in web server for complete web browser control
Built-in firewall
B. DVR Hardware
The hardware is one half of what makes a DVR work (software is the other half).
Selecting a DVR whose components work well together is critical if a stable,
reliable system is desired that is capable of providing high-quality video services.
Unfortunately, there’s no way to tell how well a digital video recorder
is designed by looking at it or by reading spec sheets. Relying on the experience
and advice of a reputable security system dealer / integrator is your best bet.
NOTE: Even the best digital video recorder will require the occasional reboot.
In fact, as a preventative measure it’s prudent to set your DVR up to
periodically reboot at a safe time. We recommend once every one or two weeks.
1. Video Capture / Video Display Card
Every digital video recorder includes at least one video capture card which
takes the analog video feeds from the cameras and converts them to digital data
which is stored on the DVR’s hard drive(s). The capture card also retrieves
the digital video data and streams it back out for display. Higher end DVRs
will have multiple video capture cards: one for recording and one for video
display.
Channels & Chipsets
Problems arise when a video capture/display card is asked to do too much. Ideally
a digital video recorder will have a video capture card and a video display
card. Each card will have a separate chipset (the electronics used to process
the video) dedicated to each video channel (camera). This way, no component
is ever asked to do more than it is capable of. With a system like this, you
could record and view full motion video on all of your cameras simultaneously
with no loss of performance. The newest DVRs use Hardware Compression Technology,
this allows for full system performance without using excessive system resources.
Unfortunately, this type of system is very expensive. Video capture cards of
this quality will have hundreds of chips and are difficult to manufacture. Most
DVR manufacturers therefore make a compromise. They realize that you aren’t
going to be recording 30 frames per second on all of your cameras at all times.
Rather, you will probably be recording only 3-5 fps and then only when motion
is present at a particular camera. They reason (and correctly so) that they
can share some of a video capture card’s resources. This enables them
to lower the cost of the DVR and still maintain good performance.
The problem comes when a manufacturer pushes this compromise too far and the
sharing of video resources begins to create a bottleneck which then results
in poor video display, recording and in the worst cases, complete failure of
the system.
"Economy" DVR manufacturers are notorious for using the cheapest video
card they can find, doing little or no testing and then inflating their product
claims. Do not buy an economy DVR unless you are prepared to live with the consequences.
2. Horsepower
Next to the quality of the video capture card, the overall speed of your DVR
is the next most important factor in determining the performance of your digital
video recorder. Speed of a computer (and remember, a DVR is a highly specialized
computer) is affected by three things primarily: CPU speed, amount of memory
(RAM) and PCI bus speed. A well engineered system from a reputable manufacturer
will have a carefully chosen set of components capable of performing their assigned
tasks without adversely affecting overall system performance.
You don’t necessarily need the fastest components to build a great DVR.
In fact, selecting only the best of everything would cost so much that only
the largest installations could afford them. The trick for the manufacturer
is to select the components that fit both budget and performance requirements.
In other words, what separates the professional quality DVRs from their "economy"
cousins is engineering.
CPU
The CPU or "central processing unit" is the part of your computer
that does all of the thinking. CPUs are rated by their maker, model and clock
speed (e.g. Intel Pentium IV 3.6 GHz). In general, the later the model and the
faster the clock speed, the more powerful the CPU.
RAM
If the CPU is the part that does the thinking, then RAM, or "random access
memory", is the computer’s short term memory. RAM is where the computer
stores the things that it’s thinking about (like processing video streams).
RAM is rated as to capacity (in Mb or Gb) and speed as well as type. The general
rule of thumb is "the more the better". Nothing reduces system performance
quite as much as not having enough RAM.
PCI Bus
A PCI Bus (which stands for "peripheral component interconnect bus")
is the part of the computer that transfers data internally from one component
to another like, for instance, between the video capture card and the CPU or
between the CPU and the network interface card (NIC). A slow bus will create
a bottleneck which will result in poor system performance. PCI buses are defined
by their throughput (Mbps) which is directly related to their speed (e.g. 33
MHz, 66 MHz, 100 MHz, etc.). The faster the bus the better.
Cooling Fan
While the cooling fan isn’t directly related to the speed of a computer
system, it is often a weak link. Just like the engine in a car will fail if
it overheats, it’s very easy for a computer to overheat and slow down
(or just simply quit). Modern computers actually use several fans to provide
adequate cooling, along with other techniques such as cooling fins, low-power
chips, etc. A good DVR must have adequate cooling, internal and external. Don’t
lock your DVR up in an unventilated closet without first considering how it
will be kept cool or you will run into problems.
3. Hard Drive(s)
The hard drive is a computer’s long term memory. With respect to a DVR,
which continuously captures huge amounts of data in the form of video and is
then expected to serve it back out for playback, the hard drive is a critical
component, both in terms of size and reliability.
Because a digital video recorder is always writing data to its hard disks as
it captures video, it’s important that the manufacturer engineers the
system in such a way that the data is not always written to the same sectors
on the drive as this will cause the drive platters to wear out prematurely.
Hard drives are rated by size, in Megabytes (Mb), Gigabytes (Gb) and Terabytes
(Tb), as well as by speed in revolutions per minute (RPM) and cache size in
Mb (e.g. 7200 Mb w/ 8Mb cache). Cache size and speed are the two major performance
factors. Hard drive size relates directly to how much video a hard drive can
store.
Standard / Removable / Fault Tolerant
There are several different types of hard drives including internal, external
and hot-swappable. Most DVRs will come with one or more internal hard drives
which are not removable. Some higher-end digital video recorders will come with
hot-swappable hard drives which can be removed from the system while the DVR
is still functioning. Finally, many DVRs can have RAID arrays attached to them.
RAID arrays are external arrays of hard drives set up in such a way that the
failure of any one hard drive will not cause data to be lost. RAID arrays are
ideal when redundancy is important and are often the only way to provide very
large (Terabyte) storage to a DVR.
4. Security I/O Ports
In addition to the standard ports found on most modern computers (which include
USB ports, VGA monitor ports, keyboard and mouse ports, printer ports, etc.)
a DVR will often have specialized ports specific to security applications.
Camera Control
Certain cameras, notably PTZ (pan/tilt/zoom) cameras require the ability for
the user to control the camera via the DVR. This is accomplished in one of several
ways. Some DVRs and PTZ cameras support PTZ control via a special signal sent
over the video cable itself. Most, however, rely on a special type of serial
connection called an RS485 Addressable Device Standard connection.
The RS485 protocol allows up to 32 PTZ devices to be controlled via a common
pair of UTP (unshielded twisted pair) cables. One pair of wires (Tx+ and Tx-)
is reserved for transmission and one pair of wires (Rx+ and Rx-) is reserved
for receiving data signals. Many PTZ security cameras will require only one-way
(simplex) communication and thus only one UTP cable will be required. Some DVRs
and cameras use the common serial port (RS232) for this function. Special adapters
exist to convert RS232 serial ports to RS485 ports.
Alarm Inputs / Outputs
Many digital video recorders are able to handle on/off alarm inputs and outputs.
These can be used to trigger recording and other security functions. DVRs with
this capability will have alarm specific input/output ports on the back of the
console. They will also usually have a port specifically reserved for alarm
system reset.
Specialized Monitor Ports
One of the primary uses of a digital video recorder is the ability to view video
at the "head-end" (the DVR itself). For this to happen, the DVR must
be able to send video to an attached monitor. Most DVRs have a standard BNC
video out monitor port but many (if not most) have other video outputs including,
from lowest to highest quality, RCA, S-Video and Component Video connections.
This allows the best picture quality to be achieved.
Standard computer monitors (with VGA output) can be used with the addition of
a VGA/S-Video modulator.
Audio Input / Output
Most digital video recorders allow audio to be recorded as well as video. For
this to work properly, the audio must be synchronized and saved with the video
channel being recorded. Check the DVR specs to make sure the DVR you’re
interested in performs this synchronization.
Audio can be simplex or duplex. Simplex audio is one-way audio and receives
only. Duplex audio allows two-way audio communication (which is recorded) between
the head-end and the camera/intercom location. Obviously special equipment (e.g.
microphones and speakers) is required to receive, transmit and record audio.
NOTE: recording audio is governed by local, state and federal laws. DO NOT record
audio without first checking the laws that apply to your area and particular
situation. Audio recording is often illegal and can result in severe penalties
including jail time and fines.
5. Peripherals
Peripherals are hardware components that are not essential for the digital video
recorder to function but can be used increase the functionality of the DVR.
Monitor
Surprisingly enough, many digital video recorders do not require a monitor to
function. Running a DVR without a monitor is known as running it "headless"
and is actually a common practice, especially when the DVR is located in a dedicated
server room where the various computers share peripherals via a KVM switch.
(A "server room" is a climate controlled room in a facility that holds
all of a company’s servers and storage hardware.)
It’s possible to run some digital video recorders headless because all
of these DVR’s video viewing, camera control, playback and administrative
functions are available over the network. Not all DVRs have every bit of administrative
functionality, but almost all have viewing and camera control capabilities and
most allow playback.
DVRs which do not have any network access and control are known as "stand-alone"
DVRs.
NOTE: to initially configure a DVR, you will need access to a video monitor
of some kind.
Keyboard
For the same reasons that a monitor is not necessary, a keyboard is not always
necessary either. In the case of a DVR that is being operated from the head-end
(as for instance by a guard or a receptionist), special keyboards, which have
built in joysticks for PTZ control and other specialized controls, are available.
CD-ROM / DVD
Many digital video recorders include built-in (or external add-on via USB) CD
or DVD burners which allow you to save video clips directly to CD or DVD with
a couple clicks of a mouse. While these devices are useful for preserving small
amounts of important video, they aren’t really suitable for video archiving.
When you consider that a moderately sized video surveillance system can lay
down upwards of 8 Gb of data per day, trying to archive all of your video data
to CDs which hold 700 Mb of data or even to DVDs which hold 4.7 Gb of data is
rarely considered practical.
Luckily, it’s rarely necessary to archive every bit of video that you
record. A built-in CD or DVD drive is a nice feature to have in some cases,
but remember that video can also be transferred across the network and saved
to a CD, DVD or hard drive located on another machine.
NOTE: DVRs will show a reduced recording and playback frame rate when archiving
video to CD or DVD.
Printer.
Almost all digital video recorders will have the ability to attach a printer
(via parallel port and/or USB) and print still video frames. This is usually
done by pausing the video on the frame you want to print and then selecting
the print option. Note that some DVRs require PostScript support to print video
stills.
6. Communications Components
Nearly every digital video recorder on the market today has some sort of remote
access available. The ones that don’t are known as "stand-alone"
DVRs and tend to be on the lowest end of the commercial DVR spectrum.
There are two standard types of remote access, Ethernet and dial-up (modem).
Often DVRs will include both types of communication components, though usually
for different reasons.
Network Interface Card (NIC)
Network Interface Cards (or NICs) are Ethernet cards that are used in every
computer that is attached to a local area network (LAN). NICs enable networked
computers to talk to one another. NICs can be either standard or wireless. Most
DVRs use standard NICs to provide remote video viewing, video playback, camera
control and administration functionality.
Note that NICs come in two speeds: 10BaseT (10 Mbps) and 100BaseT (100 Mbps).
Wireless Ethernet access comes in 802.11b (11 Mbps) and 802.11g (54 Mbps) speeds.
The faster speeds for both types of Ethernet connection are usually necessary
for good viewing and playback in multi-camera installations.
NICs are the preferred method to connect the digital video recorder to the Internet
via cable, DSL or other high-speed network connection.
Modem
While video viewing and playback can be accessed via dial-up connection to the
digital video recorder, the low bandwidth will make the experience less than
satisfactory. Modems are useful, however, to enable DVRs to send emails or pager
alerts triggered by certain events when a direct network connection isn’t
available.
7. Power Requirements
DVRs have exactly the same power requirements as a standard desktop PC which
means you just plug them into the wall.
8. Case Form Factor
Digital video recorders usually come in one of three case form factors: either
in a standard PC case, in a case that resembles a VCR or in a rack mountable
case. The first two case types can usually be adapted for rack mounting. Additionally,
special locking enclosures are available to secure your DVR and it’s power
source to prevent unauthorized access or power down.
C. DVR Software
There are many different types of software that a digital video recorder needs
to function and the quality and design of that software is every bit as important
as the hardware found within the DVR case. All digital video recorders need
an operating system, a video database and some sort of viewing, control and
administration software. Additionally, remote viewing, access and control software
is required for accessing the DVR from a remote computer.
Other optional software components include self-diagnostic software, remote-diagnostic
software, a heartbeat monitor (running on a separate PC), network reconnect
software, DNS service (running on a separate PC), web-server software, etc.
1. Operating System
Just like a normal computer, all digital video recorders require an operating
system to run. The operating systems most commonly found on DVRs are Windows
or Linux operating systems. Some manufacturers use "embedded" operating
systems which are proprietary operating systems developed by the manufacturer
specifically for their DVRs. These have the advantage of being optimized specifically
for serving video and for the particular hardware found in the DVR. They also
tend to be a bit more reliable and more immune to external security threats
like viruses and such.
2. Video Database
The reason that video stored in a DVR can be so easily and quickly stored, retrieved
and searched is because the video is kept in a database. Quality digital video
recorders will use quality databases (like Oracle,SQLServer and proprietory
Thumb Print databases) to store the video.
3. Head-End Software
Applications that run on the DVR itself (as opposed to a client PC) are known
as "head-end" applications. These include all the software necessary
to record, playback and search the video database as well as allow local and
remote administration, video access, viewing, playback, camera control, alarm
processing, notification, etc..
Viewing / Control / Administration Software
Every digital video recorder will have an application (or set of applications)
that allow video to be viewed live on an attached monitor, played back, searched,
etc. This application will also allow users to control PTZ cameras, and provide
access to the DVRs administration interface where camera types, resolutions,
frame rates, and recording schedules can be set as well as password access and
other DVR specific functions (like auto-overwrite, network settings, etc.) are
controlled. This software also provides a connection for remote access to some
or all of these features and handles alarm processing (if the DVR is equipped
to do so) and alerts and notifications.
Display Options
A quick note on viewing options. Most DVRs will allow you to set up your head-end
viewing software to display your video in several different ways including:
Sequence.
This will show you a full screen view of each camera for some pre-set (user
configurable) amount of time, effectively cycling through all of your security
cameras.
Picture-In-Picture.
This will allow you to see one camera in a small frame inset into your full
screen camera view. Some DVRs allow you to set the PIP (picture-in-picture)
frame to sequence through your other cameras and bring up the current PIP camera
to the full screen with a simple keystroke or button click.
Multi-Camera.
This allows you to view multiple cameras on screen at the same time. This is
usually in the form of a 3x3, 4x4, or 6x6 grid though some DVRs will allow you
to customize this by making some camera frames larger than others.
Depending on the digital video recorder, some or all of these features may also
be available via the remote (client) viewing and control software.
Security Settings
All digital video recorders will have some form of password protection. This
can range from a simple password upon boot-up to sophisticated multi-user, multi-group
schemes capable of restricting access to viewing, control and administrative
functions on a global or camera-by-camera basis.
Self-Diagnostic Software
Some high-end digital video recorders have a "watchdog" application
that continually monitors software and hardware status and can report any problems
to an administrator via email, pager, etc.
Network Reconnect Software
Many high-end digital video recorders have a program that will attempt to reconnect
the DVR to the network in the case of a network failure. This is only necessary
when the network requires a login. Networks that don’t require a login
will be automatically reconnected when they come back up.
Web Server Software
Some high-end DVRs have a built in web-server which allows some or all of the
digital video recorder’s functionality (including viewing, playback, camera
control and DVR administration) to be accessed via any standard web browser
from anywhere on the Internet. Note that this remote functionality is more often
provided by an application (discussed below) that must be installed on the client
computer. A built-in web-server allows you to dispense with this client application
altogether. This can be useful for accessing your DVR from computers where you
don’t have the correct permissions to install the client software (such
as from a computer in an Internet cafe).
4. Client Software
Almost all digital video recorders can be accessed remotely over a network,
including from anywhere in the world via the Internet. In most cases this is
done by installing an application provided by the DVR manufacturer on the remote
client computer, though with some DVRs all you need for access is a standard
web browser (see above).
This client software will allow you to view live video remotely. Most of these
client applications will allow you to playback video and control your PTZ cameras
as well and some will allow you full control, including administrative access
to your DVR over the network. Check the specifications carefully to see what
remote activities the DVR you are looking at provides.
NOTE: in almost all cases, you will need a broadband connection on both the
DVR and the client end for acceptable video viewing and playback. You will not
be able to get full-motion (30/25 fps) live video viewing over a dial-up connection.
In fact, you’ll be lucky to get 2 fps on a single camera over a dial-up
connection.
5. Other Software
Other software is available to enhance your DVR-based security system. Some
of the more important options are listed in this section.
Remote Diagnostic Software
Some high-end digital video recorders come with software which allows you (or
a technician) to diagnose DVR problems remotely over the network. This can sometimes
save the time and expense of a site visit.
DNS Service
For your client software to be able to find your DVR over the Internet, your
client software needs to know its IP address. (An IP address is a numerical
address in the form of AAA.BBB.CCC.DDD (eg. 012.132.034.125) which uniquely
identifies your computer on the Internet.)
Ideally, you want to have an IP address that never changes. This is known as
a "fixed" IP address. Unfortunately, it’s not always possible
to have a fixed IP address. Instead, you’re DVR may be assigned a "dynamic"
IP address by your ISP. Dynamic IP addresses are assigned from a pool of IP
addresses that your ISP owns and they can (and do) change from time to time.
The solution to this is to use a special computer called a DNS server. A DNS
server translates a network name, like "dvr1.ivid.co.uk" into an IP
address. Instead of specifying the IP address of your DVR in your client software,
you specify the network name (i.e. dvr1.ivid.co.uk) and the IP address of your
DNS server. The client software then asks the DNS server what the current IP
address of your DVR is (in our example this is "dvr1") and then uses
that IP address to connect. This allows you continuous access to your DVR even
if its IP address changes.
DNS servers can be maintained locally on your network or a public DNS server
can be used. In all cases, a DNS server is separate from your DVR.
Heartbeat Monitor
If a problem arises on your digital video recorder and it is running self-diagnostic
software, it will be able to tell you if there is a problem. But what happens
if your DVR has a problem so severe that it’s unable to report the situation
to you (like a power failure)? How will you know your security system is off-line?
The answer is: "By using a heartbeat monitor."
A heartbeat monitor is simply another computer on the network which "pings"
your DVR periodically (once a minute perhaps) and reports to you if it doesn’t
get a response. It’s a simple way for you to insure the continuity of
your security environment. Heartbeat services are available from many third-party
providers. Your security dealer / integrator can help you find a provider or
even set up your own internal heartbeat monitor – an arrangement that
will save you money in the long run.
D. Recording Time
The most commonly asked question regarding digital video recorders is also the
hardest question to answer with certainty. Namely, "How many days of video
will my DVR record?" There are many reasons this question is hard to answer.
We’ll take a look at the different factors that affect recording time
in the next few sections.
Storage time is a thorny question because not only does it depend on many different
factors, such as resolution, frame rate of each camera, etc. but it also depends
on the compression technology used by the manufacturer to store the video and
on the amount of time per day the camera is actually recording.
1. Calculating Storage Time
In general storage time can be calculated using the following formula:
T = [(11574.07 * S) / (H * V * F * C * P)]Cam1 + [(11574.07
* S) / (H * V * F * C * P)]Cam2 + ...
T = storage time in days
S = storage capacity in Gigabytes
P = percent of time camera is recording video
H = horizontal resolution in pixels
V = vertical resolution in pixels
F = frame rate in frames per second
C = average video compression rate as a decimal
The difficult parts of this calculation (aside from the fact that different
cameras may be recording at different frame rates and resolutions) are: 1) determining
the average compression rate (C) and 2) determining the percent of time the
camera is actually recording (P).
Compression Ratio (C)
The compression ratio is difficult to calculate because video is not compressed
at a constant rate. The amount of compression achieved depends heavily on how
much one frame changes from the next as compression technology relies on the
ability to record only the (hopefully) small amount of information that changes
from one scene to the next. If very little changes, compression rates will be
high. If everything in the scene changes, there will be little to no compression.
Certain things, like fluorescent lights which constantly flicker, fluttering
leaves or changing light levels can cause large portions of a scene to be constantly
changing and thereby drastically reduce the amount the video is compressed.
In practice it’s very difficult to exactly predict the average compression
ratio for many camera installations, though experience is a helpful guide.
Percent of Time Camera is Recording (P)
Most digital video recorders will be set up to record only when motion is detected
in the video frame. Activity levels can vary widely from one camera to another
and the same things which cause compression rates to fall also cause motion
activated recording to start, including flickering lights, fluttering leaves
and changing light levels. A camera that is constantly recording will eat up
hard drive space very quickly.
There are generally three things you can do to reduce the amount of time a camera’s
motion detection recording is activated: camera placement, sensitivity adjustment
and motion blocking.
Camera Placement
The first and easiest way to reduce recording time is to place your camera intelligently.
For instance, if you are recording a scene near a busy street, try to orient
the camera so that it doesn’t capture the cars moving by on the street.
Sensitivity Adjustment
The second way to reduce recording time is to adjust the motion sensitivity
of the DVR. Most DVRs will allow you to adjust the motion detection sensitivity
of each camera. This can help keep camera recording from being turned on for
subtle changes in the monitored scene.
Motion Blocking
Many DVRs also allow you to specify the area or areas of a scene where motion
will activate recording. For instance, if you have a camera pointed at a door,
you might want to set the camera to record only when motion is detected in and
around the door and to ignore motion everywhere else in the scene.
DVR Calculators
Manufacturers will all provide DVR storage calculators and give statements regarding
the storage time their units will provide but keep in mind a couple of things:
one, the times the manufacturer states assumes "normal" conditions,
whatever those are and two, manufacturers are in competition with each other
which causes the statistics they publish to be "optimistic".
Estimating Storage Time
An experienced CCTV solutions provider will be able to help you estimate the
storage time you will be able to achieve given a particular DVR and your particular
situation. Since the digital video recorder is usually the most expensive component
of your system and since it’s relatively difficult (though certainly not
impossible) to upgrade the DVR hard drive after installation, it’s usually
better to overbuy on storage space a little bit rather than be locked in to
solution that doesn’t meet your needs. Moving up to the next storage level
will only add between 3% and 5% to the overall project cost. We think it’s
an extra expense well worth making.
2. Factors Which Affect Recording Time
Hard Drive Size
As should be obvious, hard drive size directly correlates to the amount of recording
time you will realize. After a DVR video surveillance system has been optimized
(by using motion blocking, camera placement, sensitivity adjustments, adjusting
the camera frame rates, setting the appropriate resolution, etc.) the only way
to get more recording time is to increase the size of the digital video recorder’s
hard drive. Recording time varies linearly with hard drive size. In other words,
if you double the size of your hard drive, you’ll double your recording
time.
Frame Rate
Frame rate refers to how many images are displayed in a video clip every second.
30 frames per second (fps) is considered full motion video. Frame rate is also
used to describe how many video frames a DVR is capturing and writing to disk
every second.
FPS / PPS / IPS
There are several terms, including "frames per second" (fps), "pictures
per second" (pps) and "images per second" (ips) that are used
interchangeably in the industry even though they have different meanings. This
can lead to considerable confusion when discussing video display and recording
frame rates.
The terms "pictures per second" and "images per second"
generally refer to different methods of expressing video viewing and storage
rates and methods. "Frames per second" generally refers to the number
of video frames that are captured or displayed and is the most generic of the
three terms.
MJPEG vs. MPEG
Video is written to disk in two primary ways, MJPEG and MPEG with the term "pictures
per second" generally reserved for MJPEG video encoding and the term "images
per second" reserved for MPEG video encoding.
MJPEG (PPS) video is referred to as uncompressed video while MPEG (IPS) video
is compressed video. MPEG video takes up much less storage space than MJPEG
video. See the section entitled Compression & Average File Size later on
in this document for more information on video compression.
Full Motion Video
Full motion video (the equivalent of European television broadcast and cable
signals) is defined to be 25 frames per second at 4CIF, which is roughly equivalent
to a digital resolution of 704x576 pixels.
Please see i-ViD’s white paper entitled CCTV Cables & Connectors for
more information.
Global Frame Rate
When frame rate is specified for a digital video recorder, it’s usually
the "global" frame rate that’s indicated. Global frame rate
refers to the total frame rate available across all video channels, including
recording and playback, that the DVR is capable of. This is best illustrated
by way of an example.
Let’s say we have a DVR that specifies it’s frame rate as "60
fps global simplex". This means that for a 10 camera system, you could
record each camera at 6 fps (or 5 cameras at 3 fps and 5 cameras at 9 fps, etc.).
However, if you wanted to playback video, you would have to use some of your
available global frame rate for that purpose (i.e. you would no longer be operating
in "simplex" mode but now operating in "duplex" mode). Therefore
if you wanted to play back a video clip that was stored at 10 fps, you would
lose 1 fps per channel of recording capability while the video was playing back.
All Higher-End DVRs use Triplex mode which allows you to view, record and playback
video without any frame rate loss.
This is just one example of the way a typical DVR works. Depending on the way
the digital video recorder is designed, there may be differences in the way
frame rate is parsed out. For example, some DVRs can play back remotely (but
not from the head-end) without affecting the recording frame rate while other
DVRs record at the maximum global frame rate all the time so if, in a 30 fps
global system, only one camera is recording it will record at 30 fps but if
3 cameras are recording simultaneously, each one will record at 10 fps. This
has the advantage of maximizing the recorded video quality at all times but
has the disadvantage of requiring a lot of hard drive space to record the high
frame rate video.
Optimum Frame Rate
To maximize recording time we want to set the frame rates we record at intelligently.
Full motion video (30 fps) is rarely necessary for video surveillance. In most
cases a frame rate of between 3 fps and 5 fps is sufficient to capture all action
in a scene. Some situations may require a higher frame rate (at a cash register,
for example) while many situations can get by with a frame rate of 1 fps. It’s
extremely rare for CCTV security video to be recorded at more than 10 fps.
Compression & Average File Size
To save disk space, video is usually compressed before it is stored to a hard
drive using an algorithm called a "codec" (which stands for "COmpression
/ DECompression"). There are many different codecs on the market used to
store video but they all work essentially the same way. Many DVR manufacturers
will claim to have "proprietary" compression technologies but they
are usually just a standard compression technology that they have licensed and
then optimized to work with their systems.
MJPEG and MPEG
There are two primary types of video storage that you need to concern yourself
with when evaluating a DVR: MJPEG and MPEG Your choice here makes a big difference
on the length of recording time you’ll realize given a particular sized
hard drive. The basic difference between the two is that MJPEG stores the entire
frame to disk while MPEG stores only the changes from one frame to another (more
or less). MPEG-2, which is the most common MPEG compression standard, stores
approximately five to seven times more video than MJPEG.
How Video is Compressed
MJPEG
With MJPEG each frame is stored as a JPEG picture. JPEG is a compression standard
used for still pictures where similar features and colors are combined into
mathematical representations allowing the picture file size to be significantly
reduced. JPEG allows the user to select the amount of compression desired, from
completely uncompressed to compression rates of 100:1 or more. The more a JPEG
image is compressed, the greater the loss of detail and color information. Most
DVR manufacturers will select a JPEG compression of between 10:1 and 20:1 as
this is seen as a good compromise between compression and loss of detail. MJPEG
video compressed at these rates will be hard to tell from uncompressed video.
(Some DVRs allow the user to specify this compression ratio.)
While MJPEG compresses each frame, it doesn’t compress video as well as
MPEG because it doesn’t consider the previous and subsequent frames when
compressing the video.
MPEG
When recording video there will be significant portions of the scene that remain
unchanged from one frame to another. Imagine a video camera focused on a wall
that includes a window with a busy street visible through it. The portion of
the image that represents the window may be constantly changing but the rest
of the scene may not change at all. MPEG-type compression codecs take advantage
of this phenomenon, which is present in virtually all recorded video to some
degree, to greatly reduce the file size of each image.
The way this is accomplished is that every few frames the codec records a reference
frame (known as an "intra-frame") which is a full frame that doesn’t
have to reference any other frames to recreate itself. Then, for the next several
frames, only the differences from the intra-frame and the current frame are
recorded. These are known as "predicate frames". There are also "bidirectionally-interpolated
(B) frames" which reference future frames as well as past frames. These
different frame types are combined to produce a compression technique that results
in very small file sizes compared to MJPEG, up to 50X smaller for some codecs
of this type.
Where this type of compression breaks down a bit is in situations where large
portions of the scene are changing all at once. This might happen, for instance,
at sunset or sunrise when light levels are changing. The flicker of fluorescent
lights can also create a similar problem, though most DVRs have sensitivity
settings in their motion control decision algorithms which will prevent recording
altogether due to light flicker alone which can help mitigate the problem. When
recording is happening, however, light flicker can cause relatively large image
file sizes with some DVRs.
The Various Codecs
There are several different "MPEG-type" compression codecs in use
in digital video recorders today. (Though not all of them are governed by the
MPEG standards committee, they all work in a similar way and so we’re
looking at them together.) The following chart summarizes the most popular codecs
used today, comparing them to the MJPEG codec.
NOTE: the "average file sizes" and the "compression ratios"
shown above are just approximations based on a "normal" scene. The
chart above is intended to show the relative differences between the various
compression schemes and should not be used to calculate storage space without
factoring in the other factors which can affect video compression rates.
Resolution
Resolution, or picture quality, also affects storage space to a very high degree.
The greater the resolution, the more storage space is required to save the video
to disk so it’s important to select the lowest resolution necessary to
get the job done.
Resolution refers to the level of detail present in an image and is most easily
understood when expressed in pixels. (Other ways to express resolution are in
"TVL" and in "CIF". See i-ViD’s white paper entitled
CCTV Security Cameras for a complete description of these terms and how they
relate to one another.)
In the CCTV industry, "standard" PAL resolution is 352x288 pixels
while "high" resolution is 704x576 pixels. Note that television "resolution"
is roughly equivalent to 704x576 pixels at 25 frames per second. CCTV security
video is usually recorded at standard resolution and 3-5 frames per second,
therefore most CCTV video is of considerably lower quality than the video displayed
by your TV. In most cases this isn’t a problem as you don’t need
high resolution or full-motion video and you are more concerned about storage
space. When you do need to capture a better quality picture, you can use a high
resolution camera and capture the video at a higher frame rate and resolution.
The important thing to note here is that high resolution video takes up four
times as much storage space as standard resolution video. In other words, video
from four standard resolution cameras will take up the same amount of storage
space as video from one high resolution camera.
For more information on resolution as it relates to video surveillance systems,
please see i-ViD’s white paper entitled CCTV Security Cameras and also
i-ViD’s CCTV Security Camera FAQ.
E. Alarm, Access Control and CCTV Integration &
Notification Options
Intrusion alarms, environmental alarms, access control systems and video surveillance
systems all work together to create a comprehensive security environment. These
systems are commonly installed as stand-alone systems, incapable of communicating
with one another. This results in redundant hardware, inconvenient access procedures,
false alarms, and myriad control and administration points.
Within the past year or so it’s become possible to completely integrate
these disparate systems into one cohesive whole. The key to making this happen
is for one component to act as the central integration point for your security
environment. In a complete system, this role is most often served by the access
control system. For integrated security systems that do not have an access control
component, however, the digital video recorder (DVR) can serve in this capacity.
Powerful integrated security environments provide strong event associations,
remote and local single-point control / administration and comprehensive notification
feature sets. Please see i-ViD’s white paper entitled CCTV Total Integration™
Security Systems for more information on this exciting topic.
III - About i-ViD Ltd
For more information regarding any information contained in this document, please
feel free to contact us.
We’d be happy to answer any questions you might have, provide you with
further information or help you design a security system that meets your exact
needs.
IV - Notices
DISCLAIMER
IMPORTANT INFORMATION - READ THIS!
This information is provided as-is and is intended solely to assist our customers
to make informed buying decisions. It may contain errors and/or omissions of
important facts and is not intended to be used to install, operate or maintain
any equipment or for any other purpose. Under no circumstance will iViD Ltd
be held liable for the information presented herein.
