Predicting wireless quality of service (QoS) for connected vehicles
A method of a route prediction system utilizing real-time mobile communication network data. The method includes receiving a route request originating from a connected vehicle, the route request identifying a route, determining segments for the route, localizing mobile communication network resources to the segments, determining key performance indicators for the segments based on at least a current offline model, and sending predicted service level indicators (SLIs) for the segments to the connected vehicle.
1 . A method of a route prediction system utilizing real-time data obtained for a mobile communication network, the method comprising:
receiving, via a cloud platform, the real-time data and a route request originating from an autonomous vehicle connected to the mobile communication network, the route request identifying a starting location and a destination for the autonomous vehicle, wherein the cloud platform pre-processes the real-time data and identifies a route for the autonomous vehicle from the starting location to the destination;
determining segments of the route corresponding to areas covered by the mobile communication network of a mobile network operator, wherein the segments correlate to mobile communication network resources to be utilized by the autonomous vehicle with particular segments along the route;
localizing the mobile communication network resources for use in communicating with the autonomous vehicle for each of the segments;
determining key performance indicators (KPIs) of the mobile communication network for each of the segments based on at least an offline model to derive a set of KPIs that rate mobile communication network coverage for each of the segments;
adjusting the KPIs of the offline model based on the real-time data;
aggregating the KPIs into predicted service level indicator (SLI) levels;
aggregating the predicted SLI levels for the mobile communication network resources for each segment; and
sending the predicted SLI levels for the segments to the autonomous vehicle via the cloud platform.
2 . The method of claim 1 , wherein the real-time data includes any one or more of coverage area, required quality of service (QoS), current subscriber level, and cell level cellular wireless QoS, and wherein the cell level cellular wireless QoS includes any one or more of antenna, alarms, past performance, and handover settings.
3 . The method of claim 1 , wherein the offline model uses at least one of a Digital Terrain Model (DTM), a clutter map, or both the DTM and the clutter map.
4 . The method of claim 1 , further comprising:
adjusting the KPIs based on experience data received with the route request, wherein the experience data indicates an actual quality of service (QoS) for a given location and a mobile network operator.
5 . The method of claim 1 , wherein the localizing the mobile communication network resources includes correlating cell towers or geographical grids of network resources with the segments.
6 . The method of claim 1 , wherein the offline model is selected from a plurality of propagation models based on performance.
7 . The method of claim 1 , wherein an experience forecast is generated for each cell tower coverage zone that covers the route, where an approximate probability distribution is made for each cell, and where the experience forecast combines probability distributions as a weighted average.
8 . The method of claim 1 , wherein the KPIs are linked to the route by location from a set of latitude longitude pairs (LLPs) defining the route and expected vehicle speed and distance.
9 . The method of claim 4 , wherein when the experience data includes reports of low quality of service (QoS) for a current segment and selected mobile communication network, a respective KPI is downgraded or similarly adjusted to reflect the experience data.
10 . A network device to implement a route prediction system utilizing real-time data obtained for a mobile communication network, the network device comprising:
a non-transitory machine-readable storage medium having stored therein a route prediction block; and
a processor coupled to the non-transitory machine-readable storage medium, the processor to execute the route prediction block to:
receive, via a cloud platform, the real-time data and a route request originating from an autonomous vehicle connected to the mobile communication network, the route request identifying a starting location and a destination for the autonomous vehicle, wherein the cloud platform pre-processes the real-time data and identifies a route for the autonomous vehicle from the starting location to the destination;
determine segments of the route corresponding to areas covered by the mobile communication network of a mobile network operator, wherein the segments correlate to mobile communication network resources to be utilized by the autonomous vehicle with particular segments along the route;
localize the mobile communication network resources for use in communicating with the autonomous vehicle for each of the segments;
determine key performance indicators (KPIs) of the mobile communication network for each of the segments based on at least an offline model to derive a set of KPIs that rate mobile communication network coverage for each of the segments;
adjust the KPIs of the offline model based on the real-time data;
aggregate the KPIs into predicted service level indicator (SLI) levels;
aggregate the predicted SLI levels for the mobile communication network resources for each segment; and
send the predicted SLI levels for the segments to the autonomous vehicle via the cloud platform.
11 . The network device of claim 10 , wherein the real-time data includes any one or more of coverage area, required quality of service (QoS), current subscriber level, and cell level cellular wireless QoS, and wherein the cell level cellular wireless Qos includes any one or more of antenna, alarms, past performance, and handover settings.
12 . The network device of claim 10 , wherein the offline model uses at least one of a Digital Terrain Model (DTM), a clutter map, or both the DTM and the clutter map.
13 . The network device of claim 10 , wherein the route prediction block is further to adjust the KPIs based on experience data received with the route request, wherein the experience data indicates an actual quality of service (QoS) for a given location and a mobile network operator.