Noise in RF microelectronics
Download as PPTX, PDF1 like666 views
This document discusses noise in RF microelectronics. It introduces noise and nonlinearity as the two main performance limitations. Noise is random and expressed as average power. Device noise sources include thermal noise in resistors and MOSFETs, as well as flicker noise in MOSFETs. Noise is represented using voltage and current sources and metrics like noise figure are used to describe noise added by circuits. Sensitivity, dynamic range, and spurious-free dynamic range relate to a system's noise performance. Impedance transformation and matching networks can change impedances. Scattering parameters describe input and output matching as well as gain and reverse gain.
![Device Noise
Noise of MOSFETs
Thermal, flicker.
Flicker noise 1/f. modeled by a
voltage source in series with the gate.
The choice of the lower frequency
of the system should be larger than the
Corner frequency fc.
flicker noise+ nonlinearity+ (time-
variance) = translation of flicker noise
into the band of interest [Mixers, OSC]
(*)K is a process-
dependent constant](https://image.slidesharecdn.com/jjzzl2qnr5a5tziwl6yg-signature-3fabea3da5a579b1782aa9a32edb9cd99371a3313279a19531c4ac5a1a6bf461-poli-180119085649/85/Noise-in-RF-microelectronics-5-320.jpg)
Noise in RF microelectronics
- 1. RF MicroelectronicsLECTURE -2 : NOISE BY: AHMED SAKR. [email protected] SUPERVISED BY: PROF. HESHAM HAMED DR. MAHMOUD A. ABDELGHANY.
- 2. Introduction Performance limitations The two main performance limitations in RF design are noise and nonlinearity. Noise is RANDOM, which means that the instantaneous value of noise can not be predicted. Noise is expressed in term of its average power Pn.
- 3. Device Noise Thermal noise of resistors Resistors noise is modeled by an ideal resistance in series with voltage source with Vn 2 or in parallel with a current source of In 2. model should be chosen to simplify calculations. The polarity of the sources is unimportant but should be same throughout the entire calculations.
- 4. Device Noise Noise of MOSFETs Thermal, flicker. γ is the Excess noise coefficient. 2/3 for long channel MOS, >= 2 for short channel MOSs. Thermal noise may be modeled by series voltage source with the gate or parallel current source between the drain and source. flicker noise 1/f.
- 5. Device Noise Noise of MOSFETs Thermal, flicker. Flicker noise 1/f. modeled by a voltage source in series with the gate. The choice of the lower frequency of the system should be larger than the Corner frequency fc. flicker noise+ nonlinearity+ (time- variance) = translation of flicker noise into the band of interest [Mixers, OSC] (*)K is a process- dependent constant
- 6. Representation of noise in circuits. Input-referred noise Noise is modeled by series voltage source & parallel current source. To calculate Vn 2 we short the input port and measure output noise, To calculate In 2 we do the same with input port opened.
- 7. Representation of noise in circuits. Noise figure SNR (signal-to-noise ratio) = 𝑃 𝑠𝑖𝑔𝑛𝑎𝑙 𝑃 𝑛𝑜𝑖𝑠𝑒 NF describes the added noise from the circuit into the signal.
- 8. Sensitivity & dynamic range Sensitivity is the minimum signal level that a receiver can detect with acceptable quality. Noise floor The total integrated noise of the system Dynamic range : 𝑡ℎ𝑒 𝑚𝑎𝑥𝑖𝑚𝑢𝑚 𝑑𝑒𝑠𝑖𝑟𝑒𝑑 𝑖𝑛𝑝𝑢𝑡 𝑙𝑒𝑣𝑒𝑙 𝑎 𝑟𝑒𝑐𝑒𝑖𝑣𝑒𝑟 𝑐𝑎𝑛 𝑡𝑜𝑙𝑜𝑟𝑎𝑡𝑒 𝑡ℎ𝑒 𝑚𝑖𝑛𝑖𝑚𝑢𝑚 𝑖𝑛𝑝𝑢𝑡 𝑙𝑒𝑣𝑒𝑙 𝑖𝑡 𝑐𝑎𝑛 𝑑𝑒𝑡𝑒𝑐𝑡 (𝑆𝑒𝑛𝑠𝑖𝑡𝑖𝑣𝑖𝑡𝑦)
- 9. Sensitivity & dynamic range Dynamic range: 𝑡ℎ𝑒 𝑚𝑎𝑥𝑖𝑚𝑢𝑚 𝑑𝑒𝑠𝑖𝑟𝑒𝑑 𝑖𝑛𝑝𝑢𝑡 𝑙𝑒𝑣𝑒𝑙 𝑎 𝑟𝑒𝑐𝑒𝑖𝑣𝑒𝑟 𝑐𝑎𝑛 𝑡𝑜𝑙𝑜𝑟𝑎𝑡𝑒 𝑡ℎ𝑒 𝑚𝑖𝑛𝑖𝑚𝑢𝑚 𝑖𝑛𝑝𝑢𝑡 𝑙𝑒𝑣𝑒𝑙 𝑖𝑡 𝑐𝑎𝑛 𝑑𝑒𝑡𝑒𝑐𝑡 (𝑆𝑒𝑛𝑠𝑖𝑡𝑖𝑣𝑖𝑡𝑦) SFDR (spurious-free dynamic range): 𝑀𝑎𝑥𝑖𝑚𝑢𝑚 𝑖𝑛𝑝𝑢𝑡 𝑙𝑒𝑣𝑒𝑙 𝑖𝑛 𝑡𝑤𝑜−𝑡𝑜𝑛𝑒 𝑡𝑒𝑠𝑡 𝑤𝑖𝑡ℎ 𝐈𝐌 𝐥𝐞𝐯𝐞𝐥≤𝐧𝐨𝐢𝐬𝐞 𝐟𝐥𝐨𝐨𝐫 (𝑆𝑒𝑛𝑠𝑖𝑡𝑖𝑣𝑖𝑡𝑦)
- 10. Impedance transformation Quality Factor Quality factor Q indicates how close to ideal an energy-storing device (Inductor, Capacitor) is.
- 11. Impedance transformation Parallel to series conversion Quality factor Q indicates how close to ideal an energy-storing device (Inductor, Capacitor) is. For Q2>>1 (true for a finite frequency range)
- 12. Impedance transformation L sections used for matching Sometimes the load resistance is needed to become larger or smaller, that can be achieved by matching networks as following
- 14. Scattering parameters. Input matching Reverse gain Output matching Gain
- 15. Any Questions?