In the July 2008 issue of Security Technology & Design magazine, managing editor Paul Rothman began a roundtable to survey some of the top video surveillance manufacturers and integrators on adoption of IP video. Half of the roundtable appeared in the July issue of ST&D (see first part of this roundtable) and the other half appears here on SecurityInfoWatch.com and IPSecurityWatch.com.
The Participants:
• Guy Apple, vice president of marketing and sales for Network Video Technologies (NVT)
• Phil Aronson, president of Aronson Security Group
• Dr. Bob Banerjee, product marketing manager of IP Video products for Bosch Security Systems Inc.
• Jean-Pierre Forest, CPP, director of security solutions for Avigilon
• Eli Gorovici, president and CEO of DVTel, Inc.
• Duncan Havlin, vice president of product management for Samsung GVI Security
• Mickey Lavery, system specialist for I2C Technologies LLC
• Fredrik Nilsson, general manager of Axis Communications Inc.
• Mark S. Provinsal, vice president of marketing and product strategy for Dedicated Micros Inc.
• Joel Rakow, Ed.D., president of Ollivier Corporation
• Moti Shabtai, Executive VP, Strategy and Products, NICE Systems
• Steve Surfaro, Group Manager, Strategic Technical Liaison, Panasonic Security Systems
QUESTION: What’s the best or most efficient way for end-users to migrate from analog to an IP-based video surveillance solution?
Bob Banerjee/Bosch: A hybrid system is an excellent way to meet the needs of customers looking to leverage existing analog camera investments when upgrading or expanding video surveillance systems. Using IP encoders as the bridge between analog cameras and the network, video can be viewed using video management software or a Web browser, recorded to NVRs or centralized storage area networks that offer high-density fault-tolerant storage devices instead of a single hard drive
Recording at the edge, or storing video at the edge of the network instead of transporting it to a centralized NVR, is a bandwidth friendly option in today’s IP video systems. This decentralized approach only uses the network to replay video at a workstation. This renders recording independent of other network conditions, such as congestion and downtime. Many users also have analog switches with attached tapes or DVRs for recording. Or, recording can also be achieved with multi-channel encoders with direct-attached iSCSI RAIDs, as a direct and more flexible direct substitute.
Duncan Havlin/Samsung-GVI Security: Use either a DVR to connect their existing analog cameras to the network or an encoder. Install video management software that talks to their DVR and IP cameras. In addition, adding IP ready products like our new Video Plus UTP power hub that enable analog coaxial video signals and power to be sent over Cat-5 cable, is ideal for new projects that have pre-wired Cat-5 cabling but that are not yet ready to install an IP-based system.
Guy Apple/NVT: Stop installing coax! UTP-based hybrid analog, without a doubt, is the best way for end-users to set themselves up for possible future migration. The future is clearly a mix of analog and digital video. One of the most popular and fastest growing architectures is that of a hybrid system. A DVR-based system is the simplest example, where video is digitized within the DVR and made available for viewing locally or over the Ethernet network or via the internet. Analog cameras, UTP transmission and IP digitizers with hybrid network video recorders at the control room provide the next step in the evolution, where the IP backbone is comfortably confined (for high bandwidth) within the control room. The key is the deployment of a UTP-based structured cabling transmission system. It allows the use of cost-effective analog cameras, and eliminates the need for a full IP-based distribution system. It also provides for migration to a full IP architecture down the road.
Eli Gorovici/DVTel: Through integration, analog-based systems can make their way to IP-based video surveillance solutions and then as cameras are brought on-line, you just add them directly into the system. As channels fail, just bypass the integration and go directly into the new system. As a manufacturer, we see equal number of installations between retrofits and “greenfieldâ€. When migrating from analog, it is not necessary to throw away or discontinue using perfectly good equipment -- another tremendous benefit of IP-based video surveillance.
Phil Aronson/ASG: Follow the money. Digital recording has a clear cost advantage over the VCR when total costs are evaluated. Use the end-of-life of a system’s infrastructure to phase in digital replacements, consolidating equipment as facilitated by the IP infrastructure. Once the core of a system is IP-based, all new cameras and system extensions should also be IP. Finally, coordinate with other facility updates to extend the IP network to camera locations so that cameras can be upgraded to IP according to their lifecycle.
Jean-Pierre Forest/Avigilon: The most efficient way to migrate from analog to IP-based solutions is to replace the existing recording infrastructure with a solution that supports both analog cameras and HD (high definition) IP cameras. This allows an end-user to continue to use the existing analog cameras that are still useful, and HD IP cameras can be used everywhere new cameras are going to be deployed.
Fredrik Nilsson/Axis Communications: That depends on what the existing system looks like, but there are many technologies for bridging an analog system into network video. Video encoders with built-in intelligence, PoE and multi-stream capabilities are widely available, either as standalone devices to be installed close to the camera or a video encoder rack with blades, providing very high density for replacement of DVRs. Additionally there are technologies for using existing coax cables for IP transmission, a benefit when an analog camera is being replaced with a network camera.
Mickey Lavery/I2C: The easiest way to migrate analog investments to IP technology is the use of video encoders to convert the analog video stream to a digital stream. This can be done at the head end by plugging the coaxial cabling directly into an encoder. The encoder is the connected by a network cable to a server or PC. Many companies make the transition to IP as their current DVR systems get older or need replaced.
Moti Shabtai/NICE Systems: The best way is to adopt the evolutional approach and choose a platform that will support a migration path by supporting on the same unified front-end both analog and IP solutions, eliminating the need for fork lifting the analog platform.
Joel Rakow/Ollivier Corporation: One scenario is to deploy signal converters to existing analogs when it is time to replace the DVR with an NVR and then add IP cameras as needed.
Steve Surfaro/Panasonic: The use of low-latency control platforms, dedicated, embedded network disk recorders, high definition hardware decoders and video encoders supporting over-the-coax control all together form the “best practice†in analog-to-IP Video migration and deployment.
QUESTION: Are megapixel and HD technology driving increased migration to IP-based video?
Surfaro: Yes, these identification rich technologies are a key driver for IP video in certain applications. As with any IP video deployment, careful consideration to accommodate storage, maintain system cost and display High Definition imaging sources is of great importance.
Nilsson: Any technology that improves image quality will drive the technology shift. Megapixel is doing it in the same way as progressive scan is. Megapixel has many advantages, with higher resolution as the obvious one, but also drawbacks such a lower light sensitivity and higher storage needs. The sensors and light sensitivity are improving and with H.264 compression the storage concerns are being addressed.
Gorovici: Yes. Think about it: even your TVs will be forced to be HD-ready by February 2009 and just about everyone has network access in their homes. These two developments combined show the power of technology and how it affects our everyday life. With TV programs like CSI and Vegas showing enhanced video as commonplace, end users and integrators alike are eager to deploy technology that can make their system more efficient. Megapixel cameras are becoming a more mature product and are at costs competitive with higher level IP and analog cameras. Through improved engineering, some models require only a portion of the bandwidth and storage compared with when they were first introduced. Like networked-based video surveillance, edge devices are also maturing and becoming more commonplace, effective solutions.
Forest: Absolutely, end users are rapidly adopting the newest generation of HD surveillance solutions because they make it possible to offer better protection with fewer cameras and at a lower total cost. Unlike a couple of years ago when HD solutions only made sense in critical infrastructure applications, we are now seeing HD surveillance solutions being implemented in applications across the board including retail, commercial, transportation as well as school campuses.
Aronson: Megapixel and HD are examples of camera features that are natural extensions of the IP infrastructure. Technology innovators can quickly deliver these features within IP networks relative to adapting them to an analog infrastructure that was not designed to support these capabilities.
Havlin: Megapixel cameras are just beginning to make progress in the security and surveillance markets. HD is still in the future. The driver for increases in the IP video is due to the ability to better manage storage, forwarding of video events to others and central command and control of operations in major corporate and government users
Banerjee: Megapixel technology delivers superior picture quality and is especially attractive in LAN environments where the appropriate amount of bandwidth exists. It’s important to note that megapixel and HD technology are not the only answers when a user seeks increased picture quality; the majority of the mass market does not even use NTSC IP cameras to their full capability (30 FPS at 4CIF). This should be a warning sign as to the levels of expectation of the market and the user’s readiness to jump to a 40 Mbps stream per camera, which will increase storage needs by 20 to 40 times the current capacity.
Lavery: I believe that many security directors and IT departments know that these technologies will eventually be the standard, so they are making the switch to IP now. As prices for storage go down, sales of megapixel cameras will go up. In my opinion, the main reason we are seeing increased migration to IP systems is the use of standard PC-based servers for recordings instead of the DVR. As IT departments converge with security departments, we will continue to see the switch from DVRs to standard business class servers or PCs.
Provinsal: Yes. These technologies provide capabilities that cannot be achieved using traditional analog cameras. Unfortunately, the cost of bandwidth and storage limit the adoption rate.
Rakow: No. There are not enough systems that allow recording of megapixel images.
Shabtai: Potentially yes and we are starting to see the signs. Megapixel cameras not only offer a change in the video format from analog to digital, but also a much better video quality (actually for the first time better than pure analog), and better foot coverage of an area with less cameras.
QUESTION: Does a lack of operating standards impact the effectiveness of IP-based video?
Lavery: A lack of operating standards does impact the effectiveness of IP-based video, and this is why it’s important to find the right software vendor and camera manufacturer. Recently, Axis Communications, Bosch Security Systems and Sony Corporation announced that they will be working together to create an open forum to develop a standard for the interface of network video products. The goal of this new standard is to allow the integration of various brands of network video equipment. This will also help manufacturers and software developers ensure product interoperability.
Apple: IP cameras from different manufactures are often not compatible with one another, and IP video management software applications from different manufactures are often not compatible with one another. It is a choice, cost and flexibility issue.
Forest: Unfortunately, many early IP-based video solutions have suffered from poor image quality because of a lack of operating standards which directly impacted their effectiveness and tarnished the reputation of IP-video. The newest IP-based solutions, however, have addressed these issues with operating standards such as High Definition Stream Management (HDSM) and are able to deliver the high image quality and availability that end-users now expect from a surveillance system.
Nilsson: The success of the IT market was built with a strong focus on standards and the convergence of physical security products to IT will be no different. Already many of the IP enabled products follow many standards such as 802.3af for PoE and H.264 for compression. There is however more that can be done standardizing the discovery of devices and interaction of network cameras with video management software, an effort addressed by an industry forum being formed by Axis, Sony and Bosch, three of the world’s leading security camera manufactures, which was announced recently.
Rakow: No, not in my experience.
Shabtai: Yes. The fact that an IP camera requires a special interface to the system while all analog cameras are interchangeable, delays the adoption of IP solutions as customers are reluctant to depend on single-source solutions with high levels and sometimes complex levels of integration.
Surfaro: There are ANSI-approved Interoperability Standards in place now for the security industry. These standards have been developed by the Security Industry Association.
Provinsal: Yes. The customer has to validate that any IP camera that they want to utilize in their security system has been integrated into their software. Analog cameras are based on a standard. An IP video standard adopted by the industry would remove another obstacle from the adoption rate.
Aronson: There are sufficient standards for IP-based video to deliver feature parity with analog systems. Development of standards for the next generation of IP video is a market exercise, such as we have seen with networking and Internet standards.
Banerjee: Yes. For this reason, Bosch, Axis and Sony are now working on a global standard that will define how information should be exchanged between components of a network video system. An open standard for communication between devices, such as cameras, encoders and video management systems, means that end users and systems integrators will have greater flexibility in using products from multiple vendors in the same project. By creating a standardized interface, we are supporting the increasing demand for and penetration of network video equipment.
Gorovici: When IP-based video standards are finally agreed upon, it will enhance the interoperability of different systems to one another, but IP-based video is still a very effective and fast-growing solution today. Software development kits are readily available to integrate IP cameras and encoders to the software-based solutions and most manufacturers today surmount any problems of standards by implementing integration techniques.
Havlin: To a certain extent it does, but it can prevent commoditization and innovation of some products. The user must make a solid choice for selection of a partner that can support them now and into the future. A poor choice may force a user to replace product that they recently purchased or have an isolated system that cannot communicate with their video management software.
Another problem that affects operating standards is a lack of a clear definition of terms. For example, everyone knows MPEG4 is an industry standard. However, there are 20 different profiles of MPEG4 compression, the latest being H.264, which is also known as MPEG4 version 10. So, a developer can stay inside a standard, but not limit their creativity. This can be both positive and negative, depending on the consumer’s preferences of features and benefits.
QUESTION: How much longer will analog remain a viable technology choice for end-users, or, when do you see the security industry migrating to IP-based systems exclusively?
Surfaro: Both technologies will continue to be sustained and deployed in a variety of applications. The use of these technologies will be driven by the user’s own application needs and functions, regardless of the video source type. Of course, infrastructure will be the biggest influencer for the video market, and even the correct deployment of power solutions for the security industry has yet to be mastered by our integrators. The wonderful thing about these related industries are the credentialing opportunities that exist to encourage quality and success. Examples of these credentials are the CPP, PSP, RCDD and CISSP.
Gorovici: The security industry will eventually migrate to IP-based systems exclusively. With price becoming less of an issue and the current functionality and future potential of IP-based systems so much greater, it is hard to imagine that analog will remain viable after about five more years.
Provinsal: The use of analog tape-based recorders should become extinct in the next few years. The evidence is the disappearance of manufacturers these units. There are two basic hardware components of the IP-based solution: cameras and recorders. The economics of rewiring an existing installation in order to switch out analog cameras for IP cameras do not make sense. For this reason, analog cameras will continue to be viable for the next 5-10 years. The recording platforms for IP-based surveillance systems are categorized as DVRs or NVRs. However, DVRs should be categorized as encoders with built in storage. Many DVRs are now also recording IP camera video sources. The use of DVRs and NVRs are predominant in our market today. The two types of recording products will continue to co-exist until analog cameras are no longer installed.
Apple: For how many years have you heard that IP would totally take over analog systems in five years? This is a moving target. The fact of the matter is that, coming from some really small camera counts, the growth numbers for IP cameras are pretty impressive, but I have to say that the bulk of the market will be using analog cameras on coax and analog cameras in hybrid UTP beyond 2015. Installing IP cameras on UTP is just another choice.
Banerjee: Because of its solid foundation in the industry, analog will remain a viable technology for many years to come, simply because for most people it is “good enoughâ€. In fact, for the majority it is “way more than enoughâ€. However, larger applications are switching to IP video faster because they are willing to accept a more technically complex solution because of the flexibility it offers.
Aronson: The advantages of digital recording and IP distribution are too great to ignore. Currently there are few new all-analog systems being installed; most companies are implementing “hybrid†systems, i.e., analog cameras and digital storage. It could be as much as 15 years before video is exclusively IP. For some applications, however, analog cameras can remain a viable choice for many years to come. The interesting metric is when the market “tips†and 90 percent of new systems are IP. At that point, it no longer matters how long analog remains viable in niches. If we look at what has happened to telephone systems as a model, that point could be in as little as 5 years.
Havlin: Several studies have shown that IP cameras sales will eclipse that of analog, but nobody can say for sure when this will happen. It is true that analog camera sales will decrease over time, however, analog will never completely disappear. In fact, the trend may actually be slowing down somewhat as previously forecasted. Most industry experts would agree that IP-based systems have not taken on the market as fast as everyone had anticipated. Nevertheless, IP-based systems will take over the more sophisticated systems and IP products will shift more into the mainstream as the market matures.
Forest: I believe that the industry will rapidly migrate to IP-based recording solutions but that many analog cameras will remain in the field until they expire in the next 5-8 years since they can still deliver value in certain limited applications.
Lavery: Analog cameras should remain an option for the next 10-20 years, but on a much smaller scale. As computer chip and IT equipment prices continue to drop, more and more analog camera manufactures will make the transition to network cameras. A much more computer savvy workforce will be in place by then, and they will not be intimidated by the new technology, as is the case with many people today.
Nilsson: Analog systems will certainly still be around 10 years from now, if nothing else for the replacement market of existing analog systems. Just like the VCRs are hanging around longer than expected, so will the analog cameras and the DVRs, especially in the entry-level market with smaller systems of 4 to 8 cameras. However, I do not expect any of the major players to be focusing on the analog market 5 to 7 years from now.
Rakow: It is only a viable choice for the integrator if the integrator does not have the technical expertise to design and install IP. It has not been a viable choice for customers for about a year.
Shabtai: Analog will remain as long as analog solutions are cheaper for replacing existing analog infrastructure, and as long as megapixel cameras remain marginal in their market share.
Learn More:
Go back and read the other half of this IP vs. analog video roundtable; it appeared in the July 2008 issue of Security Technology & Design magazine.
