Signal processing has become a critical tool in optimizing vehicle noise. This seminar will help you to understand the foundation common to all NVH data acquisition equipment including digitizing, windows, aliasing, averaging techniques, and common analysis functions such as the power spectrum, transfer function and coherence. Fundamental concepts such as filtering, modulation, convolution, and correlation, as well as specialized techniques used in rotating machinery such as adaptive re-sampling and order tracking, will be covered. The seminar will also cover multi-input multi-output (MIMO) signal processing, array based solutions for force identification, source and path characterization and data visualization. Brief introductions to emerging concepts will also be explored and computer demonstrations, physical experiments and case studies will be used to illustrate applied, real-world problems.

• Explain the fundamental controls typical in modern spectrum analysis tools
• Interpret NVH data and judge its relevance to physical phenomena
• Extract new types of useful information from NVH data
• Implement new signal processing techniques

• Properties of the FFT
  • Sampling and digitizing
  • Aliasing and filters
  • Leakage and windows
  • Averaging techniques
  • Autopower, crosspower and coherence
  • Transmissibility and isolation
  • Measuring and interpreting the transfer function
• Rotating Machinery Basics
  • What is an order?
  • Rotation synchronous data acquisition methods
  • AM and FM modulation effects
  • FIR, IIR and re-sampling filters
  • Up-sampling down-sampling and adaptive re-sampling
• Time Frequency Methods
  • Short time Fourier transform
  • Gabor expansion and Gabor transform
  • Orthogonality, invertability and the dual function relationship
  • Gabor order tracking
  • Introduction to wavelets
• Fundamentals of Multi-Input-Multi-Output (MIMO) System Analysis
  • Review of Single-Input-Single-Output (SISO) systems
  • Introduction to Single-Input-Multiple-Output (SIMO) systems
  • Partial correlation concepts
  • Coherent output power
  • Statistical errors in basic estimates
  • Conditioned spectral analysis
• Forces and Sources in MIMO Systems
  • Least squares solution techniques
  • Force estimation technique Conditioned Source Analysis (CSA)
  • Case history: transfer path analysis
  • Case history: model correlation and updating
• Introduction to Data Classification and Pattern Recognition
  • Techniques for building and analyzing feature vectors
  • Recognition engines: neural networks and hidden Markov models
  • Applications: machine noise recognition, vision based gear mesh quality

Michael F. Albright is co-founder and General Manager of Signal.X Technologies LLC, a firm specializing in NVH engineering and custom test & measurement application development. His past positions include Business Development Manager for the NVH consulting group of LMS North America, Management and Project Engineering positions at the Roush Anatrol Division of Roush Industries, Inc, and Civilian engineering for the U.S. Navy. With experience predominantly in the automotive industry, Mr. Albright has addressed a very diverse range of noise and vibration control issues including powertrain NVH, vehicle NVH, brake noise, engine accessory noise, driveline NVH, test procedure and facility development, test/CAE hybrid simulation methods, as well as a host of manufacturing quality issues. Mr. Albright holds a B.S. in mechanical engineering from University of Cincinnati and an M.S. in mechanical engineering from Purdue University.