Media access protocols for cable modems
| dc.contributor.advisor | Kumar, S. | |
| dc.creator | Sandhu, Navjeet Singh | |
| dc.date.accessioned | 2017-10-05T18:26:13Z | |
| dc.date.available | 2017-10-05T18:26:13Z | |
| dc.date.issued | 1997-11 | |
| dc.date.submitted | November 1997 | en_US |
| dc.description.abstract | Cable modems offer qualitatively better performance than either dial-up modems or basic rate ISDN terminal adapters. The cable TV network employs coaxial cable in which the downstream consumes most of the bandwidth from 54 MHz up to the maximum frequency from 600 to 1000 MHz. The upstream is typically assigned to the range from 5 to 42 MHz. One drawback of cable modems is the lack of standards permitting any cable modem to work with any headend equipment. Work toward a multimedia cable network system (MCNS) specification was initiated by Cablelabs, a consortium of Time Warner, Comcast, Cox and Continental Cablevision. This specification defines the physical layer, media access control (MAC) layer, and operational support capabilities. In this thesis, MAC protocols are described for the upstream path of the hybrid fiber coax network system. These are Frequency Division Multiple Access (FDMA), and Slotted ALOHA which are standard methods, and Dynamic Reservation without Contention (DRNC), Dynamic Reservation with Contention (DRC), Dynamic Reservation with Limited Contention (DRLC), and Modified Dynamic Reservation with Limited Contention (M-DRLC) which are proposed in this work. Detailed mathematical analysis is done and throughput and performance are compared for all six MAC protocols when the users of the network have a variety of data rates. Simulations are done to corroborate the analysis, and simulation results are presented. The effect of network delay on the performance is also analysed. Other phenomena, such as white noise and impulse noise also affect the performance. Simulations are presented to examine the effects of noise on the cable system performance. From the analysis and simulation results, it is found that FDMA is a very inefficient protocol and it wastes lots of bandwidth. Slotted ALOHA is a simple way to allocate bandwidth, but it is efficient only at very low channel traffic. DRNC gives higher frame delay because it wastes bandwidth by polling inactive users. DRLC is efficient only at very high channel traffic. DRC is efficient if the number of users is very large and user traffic is bursty. M-DRLC is a combination of the DRC and DRLC protocols and it gives low frame delay even at high channel traffic. Ingress is a big problem in HFC network systems. Even narrowband ingress can make more than 15% of the upstream spectrum useless. Large round trip delay also degrades the cable system performance. Most of the errors generated by white noise and impulse noise can be controlled by using forward error correcting codes. It was found that system performance does not degrade substantially at 10-4 random bit error rate. | en_US |
| dc.identifier.uri | http://hdl.handle.net/10388/8175 | |
| dc.title | Media access protocols for cable modems | en_US |
| dc.type.genre | Thesis | en_US |
| thesis.degree.department | Electrical and Computer Engineering | en_US |
| thesis.degree.discipline | Electrical Engineering | en_US |
| thesis.degree.grantor | University of Saskatchewan | en_US |
| thesis.degree.level | Masters | en_US |
| thesis.degree.name | Master of Science (M.Sc.) | en_US |