binding adaptation method in cellular systems optimized to differentiate between high and low variability

摘要

"Binding adaptation method in cellular systems optimized to differentiate between high and low variability". The invention relates to a method for effecting link adaptation on the radio interfaces of an optimized packet cellular data network that handles various modulation and coding (mcs) schemes to maximize data production. In a preliminary off-line step, the system behavior in terms of net production of the various available mcss is simulated for different c / i conditions. From the simulation, two sets of tables are obtained, each including ascending and descending limits expressed in terms of block error rate (bler). thresholds correspond to switching points from one mcs to the two available mcss that have the lowest or highest immediate protection. Both sets of tables refer to rf environments of greater or less diversity, and further specialize to take into account type ii egprs hybrid files, namely incremental redundancy (ir). During transmission, the transmitted blocks are checked for fec, and the results are sent to the network. The network continually updates the bler using exponential leveling. In order to obtain the correct time response, regardless of whether or not rlc blocks are received, a reliability filter is provided whose output is used to decide the weight between new and old measurements to take the filter pulse response. of exponentially decreasing bler over time. ir efficiency is tested for each incoming block, and an indicative ir_status variable is filtered using the same approach as for the bler. each actual bler limit to be used in binding adaptation is obtained by a linear interpolation between the non-ir and perfect ir table bound, both weighted with filtered ir_status. the filtered bler is then compared against said interpolated thresholds to test for the arrival of a mcs switching condition. The power control aims to keep the peak production of qos per time interval constant (figure 16).