Optimization-Algorithms

Goal

Compare simulated annealing and quantum annealing algorithm in node placement problem

Paper Details

Title: A quantum annealing based approach to optimize the deployment cost of a multi-sink multi-contoller WSN

Conference Name: 16th International Conference on Mobile Systems and Pervasive Computing

Implementation Details

Quantum annealing algorithm implementation is available in java and python

Simulated annealing algorithm implementation is available in java

How To Execute Python Version

Initialize graph (nodes and edges). Then select a candidate list for controllers and sinks. Tune parameters (sink and controller max distance, max sink and controller coverage, etc.). Then execute an instance of TestQuantumAnnealingAlgorithm. A sample of this procedure exists here.

qaTest = TestQuantumAnnealingAlgorithm(
        graph,
        main.candidate_sinks_list,
        main.candidate_controllers_list,
        main.sink_Y_spin_variables_2d_arr,
        main.controller_Y_spin_variables_2d_arr,
        main.SENSOR_SINK_MAX_DISTANCE,
        main.SENSOR_CONTROLLER_MAX_DISTANCE,
        main.MAX_SINK_COVERAGE,
        main.MAX_CONTROLLER_COVERAGE,
        main.MAX_SINK_LOAD,
        main.MAX_CONTROLLER_LOAD,
        main.COST_SINK,
        main.COST_CONTROLLER
    )
    
 qaTest.execute()

How To Execute Java Version

The procedure is exactly the same as python version. The sample exists here.

 TestQuantumAnnealingAlgorithm qaTest = new TestQuantumAnnealingAlgorithm(
                graph,
                m.candidateSinks,
                m.candidateControllers,
                m.sinkYSpinVariables,
                m.controllerYSpinVariables,
                SENSOR_SINK_MAX_DISTANCE,
                SENSOR_CONTROLLER_MAX_DISTANCE,
                MAX_SINK_COVERAGE,
                MAX_CONTROLLER_COVERAGE,
                MAX_SINK_LOAD,
                MAX_CONTROLLER_LOAD,
                COST_SINK,
                COST_CONTROLLER
        );
        
 qaTest.execute();