Cloud computing with its services models (SaaS, PaaS, and IaaS) and its characteristics such as (on demand services, rapid elastics, and scalability) still has a number of problems that prevent some applications from the advantages of this computing paradigm. The long distance between the end-users and the cloud server might be the biggest issue for some real time applications, such as online gaming and content delivery, in term of latency. Fog computing is an emerging paradigm used to solve some issues in cloud computing. It empowers deploying the resources and the services at the edge of the network that is close to end users. The use of streaming services for sensor networking applications is a very important model in machine-to- machine paradigm of communication. In this paper we propose a system for content delivery services for mobile users using fog computing. The fog nodes are distributed at the edge of the network, in the middle between the cloud and mobile users. These nodes receive requests from the users and respond to each request either immediately, in case the service is already downloaded in the fog server, or by bringing the servers either from the cloud server or other fog nodes. The fog nodes are represented as server using COAP protocol. As expected, the results show fog computing provides low response time against the cloud computing. The results also, investigate the power consumption of mobile users for the four dierent modes, Low power mode, Transmission mode (Tx), Receiving mode (Rx), and CPU when they connect directly to fog servers.