A novel 4G MAC protocol guaranteeing high QoS and capacity in CBR communications networks by continuous group divisions and resource re-allocations to considerably reduce contention mini-slot collisions

The authors previously showed that high QoS and capacity in CBR communications networks can be achieved by the user-group-division method. The novel MAC protocol based on this method and fundamental theories and postulates (supported by the CBR network environment) required to develop the protocol are presented here. No change in hardware or standards is required to implement this protocol and the protocol can be executed on the mobile stations (ms) or cable modems (cm) from the base station (BS) or head-end (HE). Initially, the protocol divides up the network into groups formed by one contention mini-slot (cms) per region and a group of users (with the group size equals the avg. number of users per cms in the network) allocated to that particular cms. Afterwards, the cms bandwidth is re-allocated inter-regionally (regrouped) and the total channel bandwidth is re-distributed between the cms and ds (data slot) based on the feedback received on the collision and available resource (bandwidth) conditions of the network during the previous round-trip time (rtt). The protocol is executed in the network at each rtt by running the feedback mechanism, regrouping, and re-distributing the bandwidth. The advanced network simulator (ANS-1) written in MATLAB using my novel abstractions and the novel Nasrin traffic generation and distribution model are used to evaluate the protocol in a flexible and reliable network simulation environment. The results showed high fair utilization and low waste of the required cms and data bandwidth with high collision reduction. The proposed approach accomplishes a record 80% cms collision reduction and 85% cms and 100% data bandwidth utilization (guaranteeing high QoS and capacity) while providing high controllability to the BS/HE for reducing the probability of collisions. The protocol achieves these improvements by distributing the cms to small groups of users (reducing the probabilities of uncontrollable and high collisions), repeating its execution (thinning the accumulation and repetitions of collisions), and managing the ds and cms resource requirements with feedback to allocate the resources to the needy regions (reducing the probabilities of buildup and recurrences of collisions).