Blank 01

MassWare Event Model
Situational contexts are very important information in vehicle applications for either drivers or autonomous control systems to avoid traffic accidents, especially in complicated environments like cross streets and crowd highway lanes etc. Existing context-aware middleware only supports simple context monitoring or the combination of simple contexts. To improve the expressive ability and comprehensiveness for context evaluation, a binary tree based hierarchical event notification model is proposed (see the Figure), in which several events can be organized and integrated as a binary tree structure to construct an event sensor to monitor and process interested contexts and trigger subscribed actuators at runtime for reconfiguration when the conditions of the sensor are satisfied. MassWare event model supports both context expressions and user-defined functions by applying awarefunc components. To enhance the efficiency of system reconfiguration, MassWare contains multiple event sensors for supporting multiple actuators, so that each actuator can subscribe to dedicated event sensor. In the constructed event tree, each event node contains a conditioner, a left hand side (LHS), and a right hand side (RHS). There are two types of conditioners: the compare conditioner and the Boolean conditioner that do compare and Boolean operations on LHS and RHS respectively. The LHS and RHS can subscribe to the conditioner of a lower-layer event node or an event source, and notify the conditioner when the lower-layer contexts or operation results are changed. The event source can be a constant value, single context awareness, or an awareness expression that is also built on a binary tree structure, in which each node has an operator, a LHS, and a RHS. Therefore, all the contexts are organized in a hierarchical way to form a sensor. The event nodes in sensors communicate based on the publish/subscribe model. An upper-layer conditioner or an actuator can subscribe to an event node or a sensor as a listener. Thus the sensor can monitor and process the awareness results according to the event tree, and eventually report interested events to the subscribed actuator and trigger the application reconfiguration. To improve the efficiency of sensors, the hierarchical event tree is constructed according to the Modified Directed Acyclic Graph (MDAG). That is, before creating a new event node, check whether an identical node or an inverse node already exists. Event node a is defined as the inverse node of b if a and b have the same event source and comparison value, but inverse comparison operator, e.g. the inverse event of “Min(AVI_CPU, 10) < 1.0” is “Min(AVI_CPU, 10) 1.0”.

MassWare Event Model

Situational contexts are very important information in vehicle applications for either drivers or autonomous control systems to avoid traffic accidents, especially in complicated environments like cross streets and crowd highway lanes etc. Existing context-aware middleware only supports simple context monitoring or the combination of simple contexts. To improve the expressive ability and comprehensiveness for context evaluation, a binary tree based hierarchical event notification model is proposed (see the Figure), in which several events can be organized and integrated as a binary tree structure to construct an event sensor to monitor and process interested contexts and trigger subscribed actuators at runtime for reconfiguration when the conditions of the sensor are satisfied. MassWare event model supports both context expressions and user-defined functions by applying awarefunc components. To enhance the efficiency of system reconfiguration, MassWare contains multiple event sensors for supporting multiple actuators, so that each actuator can subscribe to dedicated event sensor.

In the constructed event tree, each event node contains a conditioner, a left hand side (LHS), and a right hand side (RHS). There are two types of conditioners: the compare conditioner and the Boolean conditioner that do compare and Boolean operations on LHS and RHS respectively. The LHS and RHS can subscribe to the conditioner of a lower-layer event node or an event source, and notify the conditioner when the lower-layer contexts or operation results are changed. The event source can be a constant value, single context awareness, or an awareness expression that is also built on a binary tree structure, in which each node has an operator, a LHS, and a RHS. Therefore, all the contexts are organized in a hierarchical way to form a sensor. The event nodes in sensors communicate based on the publish/subscribe model. An upper-layer conditioner or an actuator can subscribe to an event node or a sensor as a listener. Thus the sensor can monitor and process the awareness results according to the event tree, and eventually report interested events to the subscribed actuator and trigger the application reconfiguration.

To improve the efficiency of sensors, the hierarchical event tree is constructed according to the Modified Directed Acyclic Graph (MDAG). That is, before creating a new event node, check whether an identical node or an inverse node already exists. Event node a is defined as the inverse node of b if a and b have the same event source and comparison value, but inverse comparison operator, e.g. the inverse event of “Min(AVI_CPU, 10) < 1.0” is “Min(AVI_CPU, 10) 1.0”.