“ Comprehensive quality monitoring at every delivery stage can be prohibitively expensive for budgetconscious providers .” - Interra
from neighbouring networks and devices , or sending a customer a push notification to advise moving an extender for better coverage and performance . It ’ s not inconceivable that at some point in the future , networks will become completely autonomous , with physical connectivity issues being fixed by the deployments of robots or drones . AI can be positively deployed to create better experiences for subscribers and save time and money for service providers . We ’ ve coined the term ‘ Intuitive Network ’. A network that
leverages learning-based intelligence to use machine-based intuition to predict and create the best possible outcome for consumers . We believe WiFi should just work , learning and self-optimising according to individual and changeable needs and usage patterns . Interra : Many companies are proactively capturing multiple data points and analysing them alongside historical data to identify patterns and take corrective actions when necessary . For example , in OTT service monitoring , effectively capturing data from across the entire workflow , end-to-end , including sources , encoders , origin servers , edge servers , and end devices , not only enhances QoE monitoring but also enables insights and analysis of usage patterns , the diagnosis of infrastructure bottlenecks , and even the prediction of potential equipment failures . Torque : Self-diagnosis , yes quite likely . Guiding troubleshooting is already common for many consumer technologies , but the question is how effectively they can be applied to the engineering side of TV . Perhaps
12 EUROMEDIA contrary to popular belief , I do not think broadcast networks and technologies have gotten more complex over the past 10-15 years .
There was a massive step-impulse change in the move from analogue transmission to digital , and perhaps less of a change at the introduction of chunked ABR services , but since then , I think the overall complexity of broadcast operations has either stayed the same or perhaps gotten a bit simpler . ( The only possible exception might be the routing and topology mess created with the deployment of SMPTE-2110 ).
So within that context , if AI-assisted diagnostic tools can help the engineers and operators resolve problems more quickly , then that is a great thing . However , the engineers already know how to resolve those problems . An AI tool might bring it to their attention
sooner , but how to fix it is by now ‘ old school ’ knowledge .
Self-healing is an entirely different matter . Today ’ s networks are already designed with multiple levels of redundancy and multiple transmission paths . Like telecom equipment from 20 years ago , they also have rudimentary fail-over mechanisms to switch over to a backup circuit or device if something goes wrong .
The problem is that each fail-over point operates independently and isolated from the others . This can lead to dangerous ‘ cascade failures ’. There are numerous examples of such cascade failures in the telephony world . Mitigating those kind of catastrophic failures by ceding responsibility to an AI is one idea . I wonder if anyone is ready to actually do it ? VeEX : Yes , large CSPs like Comcast are already making large investments in their home-grown ecosystem , in order to handle diagnostics at all levels and increase selfhealing wherever is possible . This is easier for CSPs who have tight control and ownership over all the elements in their ecosystem .
Has there been an increase in virtualised / Cloud-based services ? Aprecomm : We see a massive trend of service providers moving telecom services into the cloud . By embracing virtualisation , operators can enhance service delivery , reduce operational costs , and increase their network ’ s agility , scalability , and security . Virtualisation enables telecom networks to quickly scale resources ( compute , storage , networking ) up or down based on traffic demands without overhauling physical hardware . It allows for the faster deployment of new services , making it easier to respond to customer needs or market changes . Virtualisation will enable services to be run closer to the user at the network edge , improving latency and performance and supporting the large volumes of data generated by IoT devices and intelligent applications . Virtualisation can reduce energy consumption by optimising hardware usage and consolidating workloads , a benefit currently at the top of the service provider ’ s list . Running virtualised functions on fewer physical servers can reduce energy consumption and cooling costs , contributing to a greener , more sustainable network . Virtualised telecom services provide better network monitoring and visibility , as the virtualised infrastructure can be easily monitored , controlled , and analysed for performance and faults . Bridge : At Bridge we have spent the past few years bringing functionality from all of our probes into the Cloud . But this is as a supplement to our appliance , embedded and software solutions , not a replacement . As with AI , the Cloud is an incredibly useful concept that has opened up a lot of avenues for a lot of operations , but it isn ’ t a one-size-fits-all solution . Broadcasters need to consider their operational context before implementing Cloud-based services : usage and demand , geographical dispersion of operations , budget , reliability , etc , and so at Bridge , we remain committed to helping them understand their own needs and then giving them a full range of choices to meet them . Interra : There has been a significant surge in demand for cloud-based QC , monitoring , and video analysis software . Cloud solutions offer unparalleled simplicity and flexibility over traditional on-prem systems , with web-based interfaces that allow access from any device , anywhere . This shift provides those overseeing critical media operations with immense benefits , enabling them to maintain and manage their infrastructure more efficiently and remotely . Torque : Cloud-based broadcast