LATE last month, the Little Rock Police Department in Arkansas publically acknowledged that its $US620,000 53-camera surveillance system is choked to a viewing and recording rate of just 10 frames per second because of insufficient bandwidth.

Next-door neighbour, the North Little Rock Police Department, is also having the same trouble, with its networked solution getting images back to the monitoring centre at only 15 frames per second. These numbers compare to the U.S. NTSC TV standard of 30 frames per second. 

The LRPD told its local TV station KATV that the problem stemmed from the fact local network providers could not deliver the bandwidth necessary to support its WAN-based cameras. At the same time they’ve gone public with the problems they’re having, both the LRPD and NLRPD point out that their CCTV systems have significantly reduced crime in both cities, as well as helping detectives solve multiple cases. 

Integrators and security managers will not be surprised to hear that at least part of the reason for the failings of these 2 systems is crimped costs. Spokespeople for each city hall say it’s too expensive to pay for the necessary bandwidth, pointing out they’re spending thousands on bandwidth as it is. 

While these particular problems are being faced by public surveillance systems in the U.S. they mirror issues we see here in Australia – not just in WAN-based but in LAN-based solutions as well. The fact is, networks are now an integral component of most new CCTV solutions and they are failing us. 

Whether it’s too little money for adequate storage for the gold standard 30 days retention, too little money for adequate network bandwidth; or non-existent bandwidth where it’s needed in the case of public surveillance solutions; the failure of networks undermines the argument for use of our best video surveillance cameras. It’s tempting to hope that H.265 is the answer to 30ips 1080-p but that answer is at least 2 years away, much longer for a pervasive presence in the market. 

There’s no doubt at all that many CCTV systems in Australia deploying the best current CCTV cameras – offering the highest resolutions and the highest frame rates – are fundamentally hamstrung by the inability of networks to move and store the video streams they produce.

What’s clear from my observation is that if you want a high quality CCTV solution you have to have everyone on board – most of all you need funding for the network side. The best example of a high quality CCTV system I’ve seen in recent years is Bankstown Sports Club, installed by ARA Security, but sanctioned and carried through thanks to the commitment of the club itself. This was a LAN-based solution with a fibre backbone. 

When it comes to WAN-based systems, especially public surveillance solutions, things are far more challenging. Running my mind across the public surveillance systems we’ve covered over the last couple of years it’s very clear that unless councils get a shoulder up from friendly utilities prepared to offer dark fibre in existing trenches, the best comms solution in public is a wireless mesh. 

The Hills Pacom Fluidmesh solution integrated by Smith & Co Security for Wellington Council in western NSW springs to mind as an example of a large wireless mesh offering high quality at an exceptional price. Certainly in these public surveillance applications, there’s a duty to ensure adequate performance. 

It’s worth recalling that last year the NSW Administrative Decisions Tribunal as part of a privacy case found Shoalhaven Council had failed in its duty to ensure “CCTV information…collected (was) relevant to purpose…(was) accurate…and (was) complete.” Essentially, the ADT argued the Nowra street surveillance system was not doing its job properly. 

Nowra’s was a legacy CCTV solution installed many years ago. In 2014, there’s no excuse for installing an IP video surveillance solution that fails to fulfil these core functions, especially if that failure is caused by a refusal to comprehend that any CCTV system is only as capable as its network’s narrowest point.