What challenges arise when your service provider customer is packaging a vast range of content and services from multiple providers ?: Aprecomm : This doesn ’ t matter as we optimise delivery over broadband , regardless of the service and which third party is delivering it . For example , our deep packet inspection and optimisation software recognises every application and optimises the traffic based on an actual need . The key is understanding that every application behaves differently , whether you ’ re streaming a Netflix movie , making a Zoom video conference call , or writing a simple email on Outlook . With the massive proliferation of
“ Service providers face major challenges when packaging third-party content and services .” – Witbe
connected devices and applications available to businesses and homes , service providers must employ intelligent systems to bring the best possible experience . Bridge : The first and most apparent answer is the sheer scale that comes with broadcasters ’ expansion of services ; it creates a lot more moving parts which demand a lot more overall bandwidth and processing power from monitoring solutions . At Bridge we have constantly expanded the capacity of our probes , allowing for a greater number of services to be monitored in parallel , and an even greater number using targeted and round robin strategies so that broadcasters can keep eyes on operations regardless of the scale of operation or diversity of sources .
But there ’ s also the issue of diversity in relation to formats and standards , and the knock-on impact this has on both interoperability and viewer consistency . Our approach has not been to pin our flag to a single mast , but to integrate monitoring for as many standards , formats and protocols as possible , so that broadcasters have full flexibility and choice in the way they structure their content ecosystem . Comprehensive monitoring from end-to-end then helps to ensure that regardless of the content source , the viewer experience is consistent . In the field of production particularly , we ’ ve created unique tools to allow for production specialists to work with multiple standards on the same platform ; for instance , by previewing HDR content even on SDR monitors , or by facilitating immersive audio monitoring in a stereo downmix or through single channel isolation . Similarly , our recent integration of Chromorama ’ s colour management system means that different camera sources can all be matched seamlessly , remotely and in real time . Interra : Creating a robust video processing infrastructure that ensures a high-quality experience for users is a significant challenge , especially given the large content volumes and variety today . Video service providers often prioritise high compression efficiency , while content providers must conduct comprehensive quality control ( QC ) to meet compliance and quality standards . This includes verifying video quality , metadata accuracy such as closed captioning , subtitles , and language tracks , and ensuring content readiness .
Beyond QC , an effective encoding and transcoding strategy must account for diverse content types , devices , and network conditions . This requires significant investment in time , resources , and user-centric data analysis to optimise QoE . Delivery adds another layer of complexity where proactive monitoring is essential to detect and resolve issues like packet loss , stream failures , or quality degradation , sometimes down to individual segments . For live streaming , real-time error detection and resolution across multiple sources and delivery paths are critical for seamless performance .
For quality control the challenge is in maintaining consistent QoE across diverse content and services as each provider may have different standards and performance levels . Complexity in integration is also an issue . Combining different technologies can be technically challenging , requiring significant effort to ensure seamless integration and compatibility . Managing the legal and contractual aspects of content licensing can be complex , especially when dealing with multiple providers with varying terms and conditions . And finally , scalability . With the growth of subscribers , the infrastructure must be scalable to handle increased demand without compromising performance . If a service is dependent on third parties , everyone involved should be in sync to take care of demand fluctuations . Torque : Quality of Experience monitoring presents a challenge for budget-conscious video service providers , who often face an apparent choice between comprehensive testing that ’ s cost-prohibitive or minimal oversight that risks poor performance and dissatisfied customers . Field measurements demonstrate that testing ten to fifteen critical user journeys hundreds of times daily , rather than sporadically testing thousands of rarely used features , reduces costs and improves issue detection . While it involves reducing overall test coverage , this focused , high-frequency approach provides superior results . VeEX : Without industry standards , the access to information about behaviour , performance issues , and test records becomes a challenge . From a test , monitoring and measurement perspective , these services can be viewed as simple data streams that are aggregated , thus explaining the industry ’ s focus on link ‘ speed ’ or throughput . However , advancing toward the integration of ML / AI components into proactive / predictive network monitoring requires an industry-wide standard or agreement . Such a framework would define how analysis , contextual interpretation and resolution information is stored , enabling effective training of AI models .
We often hear about AI ’ s advancements and potential in the medical field . This is largely due to the requirement not only to store test results , but also to document what was found , how it was treated and the outcomes . The expertise in this documentation provides context to ML / AI , and is largely absent from the communications service environment . Where information does exist , it is often stored in proprietary formats or classified as confidential . As the number of services grow , their interactions will extend beyond the shared bandwidth requirements . In order for this to work , content , service providers and network operators must openly share detailed information in all directions .
An alternative approach is to disregard existing historical data and knowledge , allowing ML / AI to start from scratch and learn solely by monitoring the network and services . However , this would be a slower process , and would still require knowledgeable humans to provide context , interpretation , and feedback for the ML / AI to learn effectively . For example , ML / AI doesn ’ t have access to the physical world ( e . g ., someone needs to log in what condition the fibre was found , what may have caused the damage and how it was fixed at X location ).
Another challenge , or concern , we are seeing today is the overuse of the AI moniker for mere marketing purposes . “ Just add some simple ML function , attach an AI logo to the brochure and make some noise ”. That trivialisation could create a trust backlash for AI in our industry . Witbe : Service providers face major challenges when packaging third-party content and services . When integrating OTT apps like Netflix or Peacock into their platforms , providers must ensure seamless functionality of all services on their devices . Poor performance will drive customers to use native Smart TV apps instead , reducing platform engagement . For providers offering direct content partnerships like Canal + with Paramount , the challenge lies in efficiently testing vast , constantly changing content libraries where manual verification becomes cost-prohibitive . In both cases , customers hold the platform accountable for their entire Quality of Experience , regardless of content origin . This necessitates automated , scalable testing approaches that can verify app functionality and content availability across the entire offering to prevent customer churn .
EUROMEDIA 15