Real-Time Signal Acquisition, High Speed Processing and Frequency Analysis in Modern Air Data Measurement Instruments

T. Katsibas
T. Semertzidis
X. Lacondemine
N. Grammalidis
ISBN: 978-953-307-002-5, InTech, Croatia, pp. 459-480, Dec 2009


There is no doubt that the signal processing task is a very critical issue in the majority of new technological inventions and challenges in the area of electrical engineering. Despite the fact that during the last thirty years numerous algorithms and techniques have been developed in the field and classical methods of DSP technology were established, it is still an extremely skilled work to properly design, build and implement an effective signal processing unit able to meet the requirements of the increasingly demanding and sophisticated modern applications. This is especially true when it is necessary to deal with real-time applications of huge data rates and computational loads. It is the objective of this book chapter to present all the successive steps of such a processs in the framework of an ambitious plan to develop a functional and reliable lidar-based air data measurement instrument, launched by the FP6 Aeronautics and Space NESLIE project. Designed to perform in real-time (onboard the plane) and to provide vital flight parameters to the cockpit of the aircraft, this remote sensing equipment dictates the implementation of a robust, flexible and very fast signal processing unit. This chapter starts by introducing the characteristics of the expected signal as well as the overall requirements of the system. Taking into account the very low signal-to-noise ratio (SNR) of a signal resulting from aerosol particles light backscattering and the different flight levels in a non-uniform and unstable environment, at which the measurement unit has to operate, we specify the appropriate algorithms and develop the necessary techniques for an effective signal processing scheme. The high speeds of modern aircrafts imply the use of a correspondingly impressive sampling rate which inevitably necessitates the use of parallelism of the computations and the optimum usage of all available resources. Very powerful FPGA boards have been selected for that purpose, making the fast processing implementation feasible. A ''slow signal processing'' module, performed in the CPUs and using input from these fast FPGA boards, produce the final measurements at a slower data rate. This software integrates the entire system by estimating the air data parameters, presenting them to a suitable MMI and synchronizing all the modules of the signal processing unit. Many simulations and real flight test results are also shown in order to prove the ability of the developed unit to perform real-time measurements of the vital flight parameters in a very effective and reliable way.