Video analytics strategies at the edge of network

• Autor(a) principal: Mendes, Bruno Miguel Fonseca
• Data de Publicação: 2021
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10773/31295
• topic: SDN; 5G networks; Edge datacenter; Video; Analytics; QoS

Several technologies are being presented with 5G networks’ evolution, technologies such as Network Functions Virtualization (NFV) and Software Defined Networking (SDN). With this evolution, we are getting closer and closer to the appearance of Smart Cities. Smart Cities will enjoy enough of the evolution of this new type of network. Cities like these are characterized by high numbers of users and a high number of equipment connected to the network (sensors, cameras). All these decrease the quality of service, while latency increases, giving a bad experience for the users. The triggering factor for this increased latency and decreased quality of service could be the existing video processing. With all the video data crossing the network, it makes the connections busier, thus worsening the other users’ quality of service. It is here where this Thesis comes in, placing this processing closer to the equipment on Edge Datacenters so that this data does not occupy the central ones. Using SDN for traffic control and Video analytics applications to process data sent to these locations. A study was carried out to study the impact of this SDN technology on a network characterized by video traffic. The results show that SDN does not impose any negative impact on this network, nor an added computational weight, and we concluded that performance had not been lost compared to traditional networks. We do gain network flexibility and programmability compared to traditional networks.