DOCSIS 3.1 upstream OFDMA channel optimization

点击翻译按钮获取中文摘要

People in urban areas are usually connected to the internet via physical plant consisting of coaxial cable optical fibre and electronic components such as amplifiers, diplexers and directional couplers/taps arranged in such a way as to allow one coaxial cable to be used for both inbound and outbound traffic. This plant connects cable modems and TVs in a subscriber's residence to a cable modem termination system located on the service provider's site. The entire system, including the software that controls it, is referred to as the CATV network. A standard known as DOCSIS, the latest version of which is 3.1., controls the structure of both the software and hardware of the CATV network and specifies the functionality of all the components used in the network. As every internet user can imagine, there has been continuous pressure to increase the data-carrying capacity of CATV networks. Service providers have responded by making improvements to their physical plant and are now planning to update the data transmission format to OFDM/OFDMA, which is a state-of-the-art format that is now allowed under DOCSIS 3.1. Choosing the parameters for OFDM/OFDMA transmission that maximize the data through-put is not a simple task. The optimum parameters are complicated functions of the size and shape of the physical plant, the impairments in the physical plant as well as the statistics of the traffic. To sustain maximum through-put the CATV plants must be continually maintained in nearly new condition. The difficulty in a proactive network management strategy is to identify the network components, while still in operation, that are beginning to degrade, but have not yet degraded to a point where they cause significant loss of through-put. The project outlined in this application has two objectives: 1) To develop algorithms whose function is to determine the optimum settings for a DOCSIS 3.1 OFDM/OFDMA system and 2) To investigate the characteristics of CATV components that predict when the components are beginning to degrade.