The IP Evolution
Security over the network comes with challenges and accolades
- By Brian Simmons
- Nov 01, 2017
Up until about 2004, analog-based security
cameras were the default choice for many
enterprise and data center customers.
But by 2010, IP-based cameras began to
overtake those analog-based designs. The
market transition occurred because of several
advantages of IP-based systems including, lower cost
of ownership, one cable plant supports both and connect,
control, converge and power by moving to everything IP.
All of the above-mentioned advantages mentioned are similar
to the adoption of VoIP technology. By using the IP network
for both voice and data, the user not only eliminates the need
for two disparate systems, but they also eliminate the need for
two separate support teams. Additionally, like with VoIP, the IPbased
video surveillance system can also be powered via Power
over Ethernet (PoE) technology. This eliminates the need for a
separate dedicated power circuit to support each device—a huge
cost savings.
A few other advantages with IP-based systems worth mentioning
are remote access anytime, anyplace; digital images do
not degrade over time; real time analytic capability; and content
distributed electronically quickly and easily.
However, even with the numerous advantages that IP-based
systems offer, there are some new concerns that must be accounted
for that weren’t relevant with analog-based systems. Among
them is protecting the connected IP network from remote hackers
located anywhere in the world who are able to gain unauthorized
access to the system. A robust network security software that is
continually updated to protect against new threats is a must.
Another is bandwidth capacity consumed. As the resolution
of cameras continues to increase, more network bandwidth is
consumed to support each camera. This also leads to an increase
in the available storage capacity required. Both concerns of storage
capacity and available network bandwidth can be eliminated
with proper planning and investing in the appropriate equipment.
But increasing bandwidth also demands more from the network
infrastructure.
The focus of this article will be to illustrate what new HD and
UHD technologies incorporated into IP surveillance video cameras
mean to the network infrastructure. It will also discuss recommended
solutions to support advanced video surveillance systems.
Network Bandwidth and
IP Video Surveillance
There are several factors that affect bandwidth requirements
from the network.
Resolution. The resolution of IP surveillance cameras varies
widely. There are VGA resolutions (600x480 lines of resolution)
up to 8.3 MP cameras which can capture 4k video quality. The
higher the resolution, the more bandwidth each camera will require
from the network and the greater the stress placed upon the
network infrastructure.
Frame rate. The human eye cannot discern above 24 fps
(frames per second) approximately. For reference, Ultra High
Definition (UHD) movies are broadcast at 30 fps. Typically, in
a video surveillance application, cameras are set to capture anywhere
between 8 fps and 20 fps depending on several factors. If
the camera will be recording high speed motion, a higher frame
count is needed. If the camera will be operating in low light environments,
a higher frame count is also needed. The higher the
frames per second, the more bandwidth required.
Streaming mode. There are generally two options with today’s
IP surveillance cameras. They are Constant Bit Rate (CBR) and a
Variable Bit Rate (VBR). For CBR, this streaming mode streams
at a constant rate. This mode is primarily used when monitoring
a relatively stable environment, with limited motion and activity.
Conversely, VBR is used when monitoring a highly active environment,
and the bit rate increases as the activity level increases.
This mode requires access to much more bandwidth during periods
of high activity than that of a CBR streaming mode.
Compression. Compression is done by a device or software
called a CODEC (COder/DECoder or COmpression/DECompression).
In IP video surveillance applications, the CODEC is inside the camera. There are several compression
technologies available, and they each
have their advantages and disadvantages.
Some compression technologies use what’s
called a predictive frame approach. This
means that if nothing changes from frame
to frame, the CODEC only transmits what
has changed in order to minimize the bandwidth
required. The other main type of
CODEC uses a transformative approach.
In this approach, the CODEC chops the
images into manageable chunks before
actually compressing it. Some of the most
popular compression technologies include
MPEG-1, MPEG-2, MPEG-4, and H.264,
with H.264 being the most popular for IP
video surveillance applications.
It is beyond the scope of this article to
elaborate in detail about each CODEC.
However, it is worth noting that the CODEC
used can have a big effect on the
bandwidth required. Changing the CODEC
from MPEG to H.264, for example,
can reduce bandwidth requirements more
than 50 percent. Incorporating compression
technologies over IP networks can reduce
bandwidth requirements, but it also
means that each frame is critically important.
Frame errors can have devastating
consequences for an application like video
surveillance, so a robust network infrastructure
is very important.
Network protocols. When transmitting
information over IP networks, there is always
a balance to be struck between latency
and reliability. In applications that are not
as time-sensitive, like data transmission
for instance, typically employ a protocol
like Transmission Control Protocol (TCO/
IP). With TCP/IP, if the receiving device
determines there are packet errors, it will
buffer the data and request a resend from
the transmitting device. Upon successful
delivery of the resent packets, the receiving
device will reconstruct the packet transmission
and display the information.
Therefore, the TCP/ IP protocol prioritizes
reliable delivery over latency. In other
applications, like VoIP and video, where
there is extreme time sensitivity, then protocols
like Real Time Protocol (RTP), or
User Datagram Protocol (UDP) are used.
When using RTP or UDP, packet errors
are simply dropped until error free packets
are delivered. The user will notice the
dropped packets as the image or audio will
be distorted or turn off temporarily until
error free packets are received again. RTP
and UDP prioritize latency over reliability.
The compression technology H.264 for
example is usually transmitted with UDP
or RTP protocols. Again, a high- quality
network infrastructure is critical to video
surveillance application and can contribute
to minimizing packet errors.
IP Video Application Testing
The TEK Center at Berk-Tek conducted
testing to determine how well different
structured cabling infrastructures supported
high definition and ultra-high definition
video in various conditions. TEK
Center engineers prepared bundles of
several different cables: LANmark-XTP
(Cat 6A), LANmark-10G FTP (Cat 6A),
LANmark 10G2 (Cat 6A), and Generic
Cat 6 cables.
The bundles were set up in a 6-around-1
configuration, as illustrated in Figure 1.
Each bundle was made up of seven cables,
and the cable in the middle (called the victim
cable) experiences the worst case alien
crosstalk possible. The surrounding six
cables are called the “disturbers.” The cable
bundles were then configured into 100
meter, four-connector channels. Please see
Every cable of every bundle was first
tested with a Fluke DSX-5000 field test
unit to ensure each met the Cat 6 or Cat 6A
performance accordingly. After successful
performance testing, the TEK Center began
testing by energizing only the victim
cable while transmitting in either 1080p or
4K UHD. Each test was repeated 12 times
for statistical confirmation. Then, the test
was repeated, but the six disturbers were
also energized with 10GBASE-T traffic
(similar frequency to UHD) to find what
happens when alien crosstalk develops
within the 6-around-1 cable bundle. This
testing was also repeated 12 times. Table 1
illustrates the test results.
The TEK Center used Quantum Data’s
780C Multi-Interface Interoperability Tester
(seen in Figure 2), which was originally
released in June 2014. The test unit provides
“Results of Confidence” which is a
near-instantaneous measurement used to
determine if the channels under test will
provide satisfactory performance. Both
1080p and 4KUHD transmissions were
tested. Please note that Quantum Data
states that the 780C has a maximum reach
of 75 meters when measuring 4K transmissions.
Therefore, the 100m 4K test results
are not considered relevant because
they are beyond the distance capabilities
of the test unit at the time of the testing
(February 2016).
In order to compensate for the test unit,
the TEK Center shortened the channel to
75 meters, and two-connector channels
were used for 4K transmissions. The results
of the 75-meter, two-connector channel
testing can be seen in Table 2 below.
Then, the TEK Center turned up the
heat. In environments like hot plenum
spaces (think Arizona summertime), combined
with up to 100W of PoE, the cables
toward the center of the bundle especially
will get hot. Heat creates higher attenuation
that will negatively impact performance.
The TEK Center tested all bundles
in the two-connector 75 meter channel test
configuration at 75-degree C (167-degree
F), to which all Berk-Tek cables are listed.
The results can be seen in Table 3.
Two solutions, LANmark-XTP and
LANmark-10G FTP, are listed to 90-degree
C (194-degree C), so the TEK center
tested both to that temperature as well.
The results are shown here in Table 4.
Structured Cabling
Recommendation
An important point to note is that, while
Category 6 and 5e are installed today to
support IP video surveillance, the probability
for success decreases significantly (as demonstrated by the TEK Center’s testing) with any of the
following.
- When multiple cables are bundled and transmitting 1080 HD
or especially 4K UHD.
- The closer to the maximum length of 100 meters.
- Any electromagnetic noise sources close to the structured cabling
(fluorescent light ballasts, power cables).
- Operating environments where elevated temperatures occur.
When transmitting high definition, ultra-high definition or beyond,
alien crosstalk (cable-to-cable noise) starts to become a significant
factor. The structured cabling system has to be designed
to account for alien crosstalk, especially when cables are bundled,
and unfortunately Category 6 products were not designed for this.
Additionally, the structured cabling system will most likely also
be carrying power (up to 100W) to energize the display monitor
eliminating the need for a separate outlet. This will elevate the
temperature (especially in a cable bundle), further stressing the
network infrastructure.
This article originally appeared in the November 2017 issue of Security Today.