The course is intended not just to lead students to know and properly describe the electronic techniques and instrumentation developed for recovering sensor signals from noise, but rather to gain a good insight in the problems faced and the approaches developed to overcome them. This implies to critically evaluate the solutions, avoiding the attitude where sensors and electronics are designed and employed just according to the established rules and standards. It is instead necessary to clarify the reasons of the choices and decisions in the light of the physics of phenomena involved, of the basic principles of signal and noise processing and of the actual performance of the available devices. It is necessary to clearly distinguish the intrinsic limitations set by physical laws from the current limitations set by the state of the art, which can be overcome by the technological progress. In essence, to gain insight means to progress at the pace of the technology evolution and be able to contribute to it.

Knowledge and understanding (Dublin Descriptor 1)

At the end of the course and after successfully passing the exam, the student will learn how to:

• Identify the theoretical best filters
• Understand the working principle of the most common filters
• Understand and manage the issues related to the 1/f noise
• Have a basic knowledge of the physics of photodiodes, phototube, strain gauges and temperature sensors

Applying knowledge and understanding (Dublin Descriptor 2)

At the end of the course and after successfully passing the exam, the student will be able to:

• Apply appropriate techniques in the design of different types of filter to meet given specification
• Select sensors and front-end electronics suitable for the specific application
• Perform the calculation of the obtainable signal to noise ratio
• Analyze and comment on specific architectural choices
• Apply the theory to assess the availability of a best filter

Signals and noise. Introduction to measurements, errors and statistical distributions. Mathematical treatment of signals and noise in the time and in the frequency domain. Signal-to-Noise ratio (S/N). Autocorrelation functions, energy and power spectra. Noise sources in electronic circuits and sensors. Main types of noise spectra. Noise interpretation and modeling with statistical pulse sequences.

Extracting signals from noise. Linear filters with constant parameters and with time-variant parameters, action on signals and noise and resultant S/N. Pulse-signals and constant-parameter low-pass filters; Gated Integrator (GI); Boxcar Integrator (BI); Sample-and-Hold (S&H) and fast samplers; discrete filtering by sampling and weighted average of samples. Optimum filtering for pulse-amplitude measurements, significance and practical usefulness. Noise with 1/f spectrum: characteristic features and ensuing problems, filtering approach. Constant-parameter high-pass filters; correlated double sampling (CDS) and further developments; Baseline Restorer (BLR). Periodic signals and constant-parameter resonant filters; modulation of signals and noise; Lock-in Amplifier (LIA), analog and digital implementations of LIAs.

Sensors are treated by discussing the physical principles of their operation; the device structure and technology; characteristic features and electrical parameters; output signals and information content; equivalent electric circuit; internal noise. Photodetectors: vacuum tube and semiconductor photodiodes; photoconductors; Photomultiplier tubes (PMT), avalanche photodiodes (APD) and single-photon avalanche diodes (SPAD); analog and digital detection, single-photon counting (SPC) and time-correlated single-photon counting (TCSPC). Temperature Sensors: thermo resistances. Strain and Force Sensors: strain gauges and piezoelectric sensors.

Foundations of electronic circuits and semiconductor devices. Foundations of signals and transmission. Basic knowledge of probability and statistics. General background in mathematics and physics.

Assessment will be based on a final exam consisting of a 2.5 hours, closed-book written exam, typically including two problems related to practical applications and theoretical questions.

Students who pass the written exam with a score above 26/30 may take an optional oral exam to adjust the final grade. The oral exam has no positive or negative limits in the increment of the written exam result.

After the exam, the problem solution will be published and the student will have one day to decide whether to request (by email) that his or her task is corrected.

Written exam (Dublin Descriptor 1-2) will be based on:

Solution of numerical problems

• Computation of signal to noise ratio in the particular application
• Minimum detectable signal estimation
• Sensor performancescalculation

Exercises focusing on design aspects

• Selection and design of filters for the specific applications
• Definition of minimum required performance of specific kind of sensor

Open-answer theoretical/practical questions on any topic of the course.

Oral exam (Dublin Descriptor 1-2) will be based on:

Theoretical/practical questions drawn from any topic of the course.

Complete set of slides employed in the lectures; Text and explanation of problems given in the written tests carried out in previous years; Papers, presentations, technical documentation, suggested references and websites dealing with signal recovery, sensors and measurement instrumentation

The text covers all the course topics and can be downloaded fo free.