Real time adaption of a latency critical application in a wireless network

A system, comprising:

• at least one edge computing system which is located nearby a respective base station of a wireless network and deployed and managed by a network provider, wherein the at least one edge computing system provisions at least one latency critical application which is to be provided to at least one end user device in a cell via a base station serving the cell and located nearby the at least one edge computing system;
• and/or
• at least one end user device which is located in the cell and which provisions the at least one latency critical application;
• wherein the system further comprises:
• a scheduler associated with the base station serving the cell;
• and
• the at least one latency critical application;
• wherein the at least one edge computing system and/or the at least one end user device are configured, respectively, to communicate with the scheduler associated with the base station which serves the cell;
• wherein a reference point value is assigned to the at least one latency critical application, wherein the reference point value is calculated via a reference point value function PWR(L, D), wherein the reference point value function PWR(L, D) assigns a tuple (LAV, DAV) consisting of mean latency LAV and mean throughput DAV of the at least one latency critical application to the reference point value PWR(LAV, DAV), and wherein the reference point value represents an allocation of resource utilization of the at least one latency critical application;
• wherein at least one of the scheduler, the at least one latency critical application, the at least one end user device and the at least one edge computing system are-is configured to:
• 1. calculate via a time variable point value function PW(L, D, T) an actual point value PW for the at least one latency critical application based on latency and throughput currently required by the at least one latency critical application at a point in time T_N-u within a time interval [T_N-u,T_N], wherein the point value function PW(L, D, T) assigns at a point in time T each tuple (L, D) consisting of latency L and throughput D to a distinct point value PW, and wherein each point value represents an actual allocation of resource utilization of the respective application;
• 2. compare the reference point value with the actual point value;
• 3. determine a difference value between the reference point value and the actual point value;
• 4. repeat steps 1) to 3) for a subset of points in time T_v within the time interval [T_N-u,T_N], wherein N, u, v are positive integer values, and u≤N and N-u≤v≤N;
• 5. sum up all difference values determined in step 3) for all points in time T_v of the subset, thus forming to form a sum SDP;
• and
• 6. provide the at least one latency critical application with the sum SDP;
• wherein the at least one latency critical application is configured to continuously adapt in real time to current transmission conditions, particularly on the cell and/or at the at least one end user device, by considering the sum SDP.

A non-transitory computer-readable storage medium comprising instructions that when executed, cause one or more processors of a wireless network to:

• A) provision at least one latency critical application which is to be provided to at least one end user device in a cell via a base station serving the cell;
• B) determine for each latency critical application of the at least one latency critical application mean latency and mean throughput which are required by the respective latency critical application, and assign the determined mean latency LAV and mean throughput DAV to a reference point value, wherein the reference point value is calculated via a reference point value function PWR(L, D), wherein the reference point value function PWR(L, D) assigns the tuple (LAV, DAV) consisting of mean latency LAV and mean throughput DAV to the reference point value PWR(LAV, DAV), and wherein the reference point value represents an allocation of resource utilization of the respective latency critical application;
• C) allocate the calculated reference point value to the respective latency critical application;
• D) calculate via a time variable point value function PW(L, D, T) an actual point value PW for the respective latency critical application based on latency and throughput currently required by the respective latency critical application at a point in time T_N-u within a time interval [T_N-u,T_N], wherein the point value function PW(L, D, T) assigns at a point in time T each tuple (L, D) consisting of latency L and throughput D to a distinct point value PW(L, D), and wherein each point value represents an actual allocation of resource utilization of the respective latency critical application;
• E) compare the reference point value with the actual point value;
• F) determine a difference value between the reference point value and the actual point value;
• G) repeat steps D) to F) for a subset of points in time T_v within the time interval [T_N-u,T_N], wherein N, u, v are positive integer values, and u≤N and N-u≤v≤N;
• H) sum up all difference values determined in step F) for all points in time T_v of the subset to form a sum SDP;
• and
• I) continuously optimizing current use of resources in the cell by considering the sum SDP of step H) for adapting the respective latency critical application in real time to current conditions on the cell.