CCTV cameras that record in Full HD have become the norm, and 4K is not going to be far behind. The level of visual fidelity they provide has obvious advantages, such as being able to clearly identify objects, facial features, and generally enhance the capability of all the fancy analytics software running in the background.
But capturing that footage is not enough: the data generated has to be transported across a local network, and in many cases externally to remote control rooms and data centres. Here are some statistics that illustrate the scale of the challenge: 4K cameras require about 110 GB of storage per hour if used at their maximum capacity, as opposed to around 12,7 GB; per hour for standard definition cameras. Furthermore, research expects the enterprise surveillance market to post a CAGR (compound annual growth rate) of 42% through 2021. It is predicted that video will account for 15,1 zettabytes (1 zettabyte = 1 trillion gigabytes) of data annually, which is more than any other IoT application.
With the proliferation and convenience of wireless networks, this begs the questions: are they up to the job of handling all this data, when the reliability of surveillance networks is so crucial? Hi-Tech Security Solutions asked three local experts for their views:
Teresa Huysamen, wireless business unit executive, Duxbury Networking
Point-to-point wireless is the best topology to use for backhauling from the cameras’ collection point or base stations to the control room, since a high-capacity link can be employed and bandwidth from multiple cameras is carried on a single transmission medium. Its capacity is designed to cater for the total bandwidth required.
In terms of delay and jitter control, wireless products, while excellent, cannot match wired performance. However, given that voice and video applications only require latency under 50 ms and jitter below 3 ms; for seamless video feeds, wireless networks can easily meet these specifications if designed properly.
This is not the best technology to use in the case of consecutive cameras where each additional camera in the chain would cause data to accumulate to the point where every consecutive link must carry the additional data from the previous camera. This would cause a bottleneck and result in reduced quality.
Point-to-multipoint wireless is the ideal topology to use when the data from several cameras from different locations need to link to a central point. Each camera, or group of cameras via a switch, will connect to a subscriber module which will then connect wirelessly to a base station. This is not always practical due to line-of-sight restrictions from each camera site to the central point. The total bandwidth requirement must always be kept in mind and not exceed the capability of the base station unit.
Coping with weather and physical obstructions
Moisture such as fog, rain, and snow (depending on its water content) adds attenuation to the signal’s path. Rain depends on the amount of rainfall (measured in mm/h) and the size of the raindrops. The heavier the raindrops and the higher the velocity of the rainfall, the higher the attenuation. Typical rainfall produces roughly5,5 dB of attenuation.
It depends on the amount of rain coming down and the frequency being used. Radios are designed to operate with a certain level of ‘fade margin’ that allows the system to operate at a predictable reliability, so if a system is designed and installed properly, a wireless backhaul system can still produce reliability of 99,999% (less than 5 minutes’ predictable yearly outage).
Generally, radio links are never designed with an obstruction in the path. It is possible to have non-line-of-sight or near-line-of-sight links, but for reliable transmission this is normally avoided. All links are designed with clear line of sight and first Fresnel zone clearance to avoid signal degradation resulting from reflections on the path.
Wireless interference is the biggest challenge faced by wireless networks. Modern radios like those from Cambium have several interference mitigators. These include an active filter which adjusts to the operating frequency of the link and electronically filters out interference from nearby frequencies in the same band. GPS synchronisation eliminates self-interference and directional antennas with excellent front-to-back ratio allow frequency re-use to help minimise RF frequency requirements.
To completely avoid interference, licensed frequency links can be employed, which gives the user exclusive use of the frequency and cannot be interfered with by another link using the same frequency.
Understand the environment
Even though some environments, such as long distances over rough terrain, are probably best suited for wired communication, wireless can be comfortably deployed in commercial buildings, malls, campuses, residential estates, etc., provided that the general guidelines for the use of wireless technology are respected.
A wireless signal must overcome certain obstacles which cause signal attenuation, like walls, trees and conductive materials which are often common in the urban environment. If correctly designed and the obstacles are avoided as much as possible, this is a workable solution. The biggest enemy of wireless in the unlicensed frequency arena is radio frequency interference. The major players today have advanced interference mitigation techniques which can still offer great performance despite these challenges.
Fit for enterprise use
The technologies used for enterprise applications generally are based on the standard chipsets available, but they are modified to add proprietary protocols to counter the snags that can crop up using pure Wi-Fi technology, such as the hidden node problem.
GPS synchronisation and high-specification antennas with excellent front-to-back ratio, air fairness, active filtering and excellent spectral efficiency are all used in the unlicensed frequency bands to ensure that these bands can still be used effectively.
The other options are the licensed frequencies in the 6–38 GHz frequency bands which require an annual licence fee but guarantee interference-free performance. The 17 GHz band also offers licence-free connectivity and has a lower likelihood of interference, but this comes at a price. The millimetre-wave frequency bands like 60 and 80 GHz are also very practical for short-range links and very high capacity up to around 10 Gbps. 60 GHz and 5G are also being developed for point-to-multipoint, high-bandwidth applications for surveillance in the urban environment.
Some vendors are standards agnostic and demonstrate a readiness to adapt to new technologies as new wireless networking options become available and business requirements evolve. This allows them to exploit the possibilities offered by technologies like 5G in future.
Security should never be an afterthought and it should start at the design phase. Open standards for edge devices increase the speed of innovation but also increase the vulnerability. A layered security architecture across all the devices within the network should be followed so that there is no single point of failure. It should include physical, management, data and process security.
Nick Ehrke, country manager – South Africa, RADWIN
Any topology needs to support the fundamental customer requirement for availability of the surveillance network, the resiliency, serviceability and security of the network. Any network deployed should always be ‘fit for purpose’, so it must meet the current requirements, but have the scalability to meet future network growth.
The topology used at the outset should take this into account. So as an example, by deploying a wireless point-to-point (PtP) ‘ring’ topology for the backbone, the network can sustain a link failure without losing connectivity to the surveillance network.
A point-to-multipoint (PtMP) network has a number of cost and scalability advantages for an enterprise surveillance network. Deploying the base station in a secure location, with redundant power, etc., provides the availability required, and a loss of a subscriber does not affect any other subscriber in the network.
The type of wireless equipment used should be carrier grade in terms of robustness, and be able to work in harsh environmental conditions, but also be secure, i.e., the wireless network should not be able to broadcast an SSID.
A mesh network from RADWIN operating in the 60 GHz band will be able to operate as either PtP, PtMP, or multipoint-to-multipoint, and have the advantage of being able to cascade, one node to the next. Furthermore, it will introduce self-organising and self-healing to the network, for optimal link robustness and uptime.
Environmental capabilities and planning
Wireless networks in sub-6 GHz are less susceptible to environmental conditions like rain, sleet, snow, sandstorms, etc. Yet still, the equipment selected should have technology that specifically focuses on link robustness and uptime. Companies should look for specific technologies like forward error correction (FEC) and advanced automatic repeat request (ARQ) mechanism.
Avoid wireless technologies that support carrier-sense multiple access with collision avoidance (CSMA-CA) protocol (typical Wi-Fi based products). Wireless network synchronisation is critical to avoid self-interference, while the most reliable PtMP networks have smart beamforming in the uplink and downlink. This has the advantage of mitigating interference from competing radios, as well as overcoming near-line-of-sight or no-line-of-sight challenges.
The wireless network selected should always be planned, designed and implemented with due regard for the environment it is being deployed into. By planning correctly, any and all obstacles can be avoided or overcome. Even though a wireless link can be established over a distance of 1 or 120 km, can it deliver the high capacity and low, consistent latency and jitter a surveillance network requires? This is why network planning is key to all successful deployments.
Wireless leads the way
The wireless ecosystem is continually advancing, and is in fact leading the advance for high-capacity connectivity both on the consumer side (Wi-Fi) and on the enterprise side (Enterprise Wi-Fi, fixed wireless and mobile wireless like LTE and 5G), and all these technologies, rather than being seen as competing, should be seen as complementary.
Consider the case of an emergency situation, where the mobile operator networks are able to provide connectivity to employees who can view live surveillance feeds on their mobile devices wherever they are in the world. The backhaul to the mobile device is over an LTE network, carried from the hub sites using fibre backhaul. Any system should take into account the end-to-end architecture and be designed accordingly, so that the network underpins the business objectives.
Riaan Graham, sales director, Ruckus Networks sub-Saharan Africa
In this day and age, we want as clear a picture as possible, so we’re talking about Full HD or even 4K cameras. So the first thing is you need to be able to at least support the latest standards across the Wi-Fi infrastructure.
The second thing we need to think of is that you would not necessarily have to build an independent security surveillance infrastructure. Let’s use the case of a metro, for instance – the metro would be able to use its traditional Wi-Fi network that it uses for its offices or the hotspot connectivity within the metro to piggyback the CCTV or the surveillance systems off that existing infrastructure. There’s therefore a compelling argument that the cost need not be as excessive as a lot of people think.
It’s also important to note that Wi-Fi extends far beyond the typical consumer’s use of it as a convenience for Internet connectivity. Globally, 70% of all data that gets transferred across any network is done via Wi-Fi, be it video, audio, etc. So the technology has proven itself to be very robust and reliable for data transfer, regardless of what it is that you want to transfer. From a quality perspective, or a reliability perspective, there is no doubt in my mind about Wi-Fi being the ideal technology to utilise.
Wi-Fi’s suitability for surveillance
Wi-Fi technology has grown and developed in leaps and bounds. In the case of a surveillance network, you can very easily set up independent VLANs (virtual local area networks) that get transferred across the Wi-Fi network. In that way you would set up a dedicated service for video surveillance specifically, you can create dedicated services for IoT, for monitoring services, etc. And they all live happily within the Wi-Fi environment – this is something we can pretty much do with our eyes closed.
Further to this point, at the end of the day a surveillance network is typically an IT LAN deployment, so we engage with the security people to facilitate the translation of their ‘language’ into IT lingo so to speak, to make sure that the IT guys understand exactly what the security guys’ needs are, and vice versa. This is important in order to avoid issues in the deployment.
The key message is that you needn’t build independent surveillance networks, you can easily use existing infrastructure. Again, I emphasise the metro environment as a reference because they normally have quite a substantial CCTV and surveillance network. Utilising Wi-Fi thus reduces cost, it is very easy to prioritise services via VLANs, it is secure, and it caters for future growth such as the use of IoT services that might be introduced. This is something we’ve been doing for a long time, it’s not something new. It’s been done, proven and tested.
Products fit for purpose
Ruckus Networks has got a number of technology sets, including for indoor access and outdoors, which is critical for your outdoor hotspots and is typically where your surveillance requirements would be. For instance, our technology is well suited to be deployed at intersections. Over and above CCTV, you are able to connect IoT services and devices onto the access point (AP) as well. Its devices come IoT-ready, which means that you can add functions such as sound monitoring and traffic flow management, in conjunction with the surveillance.
“On some of our AP ranges, we’ve got PoE (Power-over-Ethernet) in and PoE out, catering for ease of installation of surveillance cameras – you just plug the camera into the access point, it gets powered up so you don’t have to pull your own cabling, and it’s on the network. Again, this represents a very easy way of either utilising existing infrastructure, or if you need to deploy rapidly, Wi-Fi is probably the best technology to use in conjunction with surveillance.”
For more information contact:
• Duxbury Networking, +27 11 351 9800, [email protected], www.duxbury.co.za
• RADWIN, +27 74 114 2805, [email protected], www.radwin.com
• Ruckus Networks, 0800 999 336, [email protected], [email protected]
© Technews Publishing (Pty) Ltd. | All Rights Reserved.
printer friendly version