Intelligent Signal Processing Léonard Studer IPHE-UNIL 2017/11/10 1 Intelligent Signal Processing ? Sensors … Electronics … DSP … ISP High Machine Intelligence Quotient Human-like Information Processing 2017/11/10 2 Components of ISP Artificial works (ANN) Adaptability, robustness, data-oriented Fuzzy Logic (FL) Interface btw language and numeric puting (EC) (aka. ic algo/programming …) « invent » unforeseen solution, (sub)-optimization [ Other Related Concepts ] Support Vector Machines, puting, AI… 2017/11/10 3 Léonard Studer, IPHE-UNIL Which for What ? ANN EC FL Learning Capability Optimizing Capability Representing Capability bi is possible and used: Goal is to realize processing systems with greater intelligence 2017/11/10 4 Léonard Studer, IPHE-UNIL Artificial works Biologically work of simple processing elements Distributed function Inputs Outputs : Neuron : weighted link 2017/11/10 5 Léonard Studer, IPHE-UNIL Some real ANN usages Recognition of hand-written letters Predicting on-line the quality of welding spots Identifying relevant documents in corpus Visualizing high-dimensional space Tracking on-line the position of robot arms … etc 2017/11/10 6 Léonard Studer, IPHE-UNIL ANN a good choice if: Data-rich / model-deficient problem Failure of classical mathematical modeling Nonlinear, multidim input/output mapping Failure of classical linear methods (try it first) Enough time to design the final ANN Hours to days to get a ~m sec cycle ANN 2017/11/10 7 Léonard Studer, IPHE-UNIL Preliminary steps for ANN Get a lot of data : inputs and outputs Analyze data on the PC Relevant inputs ? Linear relations ? Transform and scale variables Other useful preprocessing ? Divide in 3 data sets: Learning set Test set Validation set 2017/11/10 8 Léonard Studer, IPHE-UNIL First design step for ANN Set the ANN architecture (PC or board) MLP, RBF, TDNN, Kohonen, GNG ? Number of inputs, outputs ? Number of hidden layers N
Intelligent Signal Processing 来自淘豆网m.daumloan.com转载请标明出处.
