Preprint. This is the submitted version of a paper published in Electronic environment.

Similar documents
ECEN 5014, Spring 2013 Special Topics: Active Microwave Circuits and MMICs Zoya Popovic, University of Colorado, Boulder

Complex RF Mixers, Zero-IF Architecture, and Advanced Algorithms: The Black Magic in Next-Generation SDR Transceivers

A technique for noise measurement optimization with spectrum analyzers

6.976 High Speed Communication Circuits and Systems Lecture 16 Noise in Integer-N Frequency Synthesizers

( ) D. An information signal x( t) = 5cos( 1000πt) LSSB modulates a carrier with amplitude A c

Amplifiers. Department of Computer Science and Engineering

3.6 Intersymbol interference. 1 Your site here

Frequency Hopped Spread Spectrum

Analog and Telecommunication Electronics

Lousy Processing Increases Energy Efficiency in Massive MIMO Systems

Recognition of User Activity for User Interface on a Mobile Device

An image rejection re-configurable multi-carrier 3G base-station transmitter

Image Enhancement II: Neighborhood Operations

ICT 5305 Mobile Communications. Lecture - 3 April Dr. Hossen Asiful Mustafa

Wireless Channel Modeling (Modeling, Simulation, and Mitigation)

DKAN0008A PIC18 Software UART Timing Requirements

Consumers are looking to wireless

Outline. Wireless Networks (PHY): Design for Diversity. Admin. Outline. Page 1. Recap: Impact of Channel on Decisions. [hg(t) + w(t)]g(t)dt.

High Speed Communication Circuits and Systems Lecture 10 Mixers

The way of PIM3 to -150dBc

ELEC3106 Electronics. Lecture notes: non-linearity and noise. Objective. Non-linearity. Non-linearity measures

Page 1. Telecommunication Electronics TLCE - A1 03/05/ DDC 1. Politecnico di Torino ICT School. Lesson A1

Measuring the Speed of Light

On the Impact of Fading and Inter-piconet Interference on Bluetooth Performance

Chapter 25: Transmitters and Receivers

PLANNING AND DESIGN OF FRONT-END FILTERS

Thinking Outside the Band: Absorptive Filtering Matthew A. Morgan

ECE5984 Orthogonal Frequency Division Multiplexing and Related Technologies Fall Mohamed Essam Khedr. Channel Estimation

Frequency-Foldback Technique Optimizes PFC Efficiency Over The Full Load Range

Chapter 6: Introduction to Digital Communication

Lock-In Amplifiers SR510 and SR530 Analog lock-in amplifiers

ATLCE - B5 07/03/2016. Analog and Telecommunication Electronics 2016 DDC 1. Politecnico di Torino - ICT School. Lesson B5: multipliers and mixers

Bode Plot based Auto-Tuning Enhanced Solution for High Performance Servo Drives

Intermodulation Distortion in RF Connectors

Detection and direction-finding of spread spectrum signals using correlation and narrowband interference rejection

Software Defined Radio Forum Contribution

APPLICATION NOTE #1. Phase NoiseTheory and Measurement 1 INTRODUCTION

SAW STABILIZED MICROWAVE GENERATOR ELABORATION

A Physical Sine-to-Square Converter Noise Model

Comparison of Optical Sparse Aperture Image Restoration with Experimental PSF and Designed PSF Zhiwei Zhou, Dayong Wang

ISSUE: April Fig. 1. Simplified block diagram of power supply voltage loop.

Digital Image Processing Chapter 3: Image Enhancement in the Spatial Domain

McGill University. Department. of Electrical and Computer Engineering. Communications systems A

Traditional Analog Modulation Techniques

Analog ó Digital Conversion Sampled Data Acquisition Systems Discrete Sampling and Nyquist Digital to Analog Conversion Analog to Digital Conversion

Spectrum allocation with beamforming antenna in heterogeneous overlaying networks

Chapter 3. System Theory and Technologies. 3.1 Physical Layer. ... How to transport digital symbols...?

Indoor GPS Technology Frank van Diggelen and Charles Abraham Global Locate, Inc.

Prof. Paolo Colantonio a.a

Outline. Wireless PHY: Modulation and Demodulation. Admin. Page 1. g(t)e j2πk t dt. G[k] = 1 T. G[k] = = k L. ) = g L (t)e j2π f k t dt.

Predicting the performance of a photodetector

Intermodulation Distortion in RF Connectors

Outline. Wireless PHY: Modulation and Demodulation. Admin. Page 1. G[k] = 1 T. g(t)e j2πk t dt. G[k] = = k L. ) = g L (t)e j2π f k t dt.

High Precision Wireless Measurement of Temperature by Using Surface Acoustic Waves Sensors

Announcements. Focus! Thin Lens Models. New Topic. Intensity Image Formation. Bi-directional: two focal points! Thin Lens Model

Analog and Telecommunication Electronics

Validation of a crystal detector model for the calibration of the Large Signal Network Analyzer.

Signals and Systems II

Architectural benefits of wide bandgap RF power transistors for frequency agile basestation systems Fischer G., Member IEEE

Simulation of Radio Frequency Integrated Circuits

Finding Loop Gain in Circuits with Embedded Loops

Fundamentals of Spectrum Analysis. Christoph Rauscher

Spread-Spectrum Technique in Sigma-Delta Modulators

Sinusoidal signal. Arbitrary signal. Periodic rectangular pulse. Sampling function. Sampled sinusoidal signal. Sampled arbitrary signal

Potentiostat stability mystery explained

Chapter 2 Introduction to Logic Circuits

Estimation and Compensation of IQ-Imbalances in Direct Down Converters

Performance of LTE Linear MIMO Detectors: Achievable Data Rates and Complexity

Signal Sampling. Sampling. Sampling. Sampling. Sampling. Sampling

EEE 311: Digital Signal Processing I

Study Guide for the First Exam

287. The Transient behavior of rails used in electromagnetic railguns: numerical investigations at constant loading velocities

1. Motivation. 2. Periodic non-gaussian noise

A MATLAB Model of Hybrid Active Filter Based on SVPWM Technique

Introduction to OFDM. Characteristics of OFDM (Orthogonal Frequency Division Multiplexing)

Experiment 7: Frequency Modulation and Phase Locked Loops Fall 2009

On the Optimality of Single-Carrier Transmission in Large-Scale Antenna Systems

The lunar laser communication demonstration time-offlight measurement system: overview, on-orbit performance, and ranging analysis

A Program to Analyse the Origin of Noise in Ultra- Stable Quartz Crystal Resonators

EXPLOITING RMS TIME-FREQUENCY STRUCTURE FOR DATA COMPRESSION IN EMITTER LOCATION SYSTEMS

Improved Random Load Processing Algorithm Based on Kalman Filter

Instantaneous frequency Up to now, we have defined the frequency as the speed of rotation of a phasor (constant frequency phasor) φ( t) = A exp

Optimizing Reception Performance of new UWB Pulse shape over Multipath Channel using MMSE Adaptive Algorithm

1. Introduction. Kenichi Higuchi, Noriyuki Maeda, Hiroyuki Kawai and Mamoru Sawahashi

Noise Removal from ECG Signal and Performance Analysis Using Different Filter

Simulation Results for Permutation Trellis Codes using M-ary FSK

Sequence-based Rendezvous for Dynamic Spectrum Access

Study on 3D CFBG Vibration Sensor and Its Application

CX On the Direct Conversion Receiver. Abstract. Traditional Reception Techniques. Introduction

INTERFERENCE effects of wind turbines on communication

Multiple access techniques

New metallic mesh designing with high electromagnetic shielding

White Paper. DEVIALET 10 place Vendôme PARIS Téléphone:

Noise. Interference Noise

A UHF CMOS Variable Gain LNA with Wideband Input Impedance Matching and GSM Interoperability

On the Impact of Carrier Frequency Offsets in OFDM/SDMA Systems

Development of an FHMA-based Underwater Acoustic Communications System for Multiple Underwater Vehicles

state the transfer function of the op-amp show that, in the ideal op-amp, the two inputs will be equal if the output is to be finite

D E V I A L E T PA R I S

Effects of Bit Allocation on Non-contiguous Multicarrier-based Cognitive Radio Transceivers

Transcription:

http://www.diva-portal.org Preprint This is the submitted version o a paper published in Electronic environment. Citation or the original published paper (version o record): Stranneb, D. (0) A Primer on PIM. Electronic environment, (): 0- Access to the published version ma require subscription. N.B. When citing this work, cite the original published paper. Permanent link to this version: http://urn.kb.se/resolve?urn=urn:nbn:se:oru:diva-0050

A Primer on PIM Background PIM is an acronm or Passive InterModulation, an intererence problem in wireless sstems. The problem is not new, but has been known since long time back b designers o, or instance cell phone sstems, space probes, connectors, coaxial cables, antennas and ilters. The problem most requentl occurs, when dealing with high RF-currents in conined spaces. In this text, the basic theories behind PIM will be briel discussed. Intermodulation Intermodulation is a process where two or more signals, having requenc components,, etc, are mixed in such a wa, that new requenc components, not belonging to the initial set o requencies, are created. In some applications, or instance ampliiers, intermodulation causes distortion o the signal, and is not a desired propert. In mixers, modulators and demodulators however, intermodulation is used to shit signals rom one requenc band to another. In this case, it is a desired eature. For intermodulation to occur, two or more signals need to be multiplied. A common wa to achieve this is to add the signals, and to pass the sum through a nonlinear device. To illustrate the properties o a nonlinear device, let us start with the concept o linearit. Assume we have or instance an ampliier, with input signal ( a linear case, be expressed as: ( kx( x and output signal (. The relation between the two signals can, in =, where k is the gain, which is a constant. The linear ampliier is simpl a perect proportionalit, as shown in igure. A straight line, passing thru the origin, where k is the slope o the line. Figure, the transer curve o a linear device Now, assume the input signal x ( to the device consists o two ine unctions, with requenc and respectivel: ( = ( π + ( π x then, the output signal ( will be: ( = kx( = k ( π + k ( π

The output signal onl contains requencies Perect!,,, the same requencies as in the input signal. Now, or the nonlinear device, sa or example that the transer unction o the ampliier is: ( x ( igure. =, i.e. the output signal is the input signal raised to three. An example can be ound in Figure, example o transer curve o a nonlinear device This is certainl NOT a straight line, and is thereore a nonlinear transer unction. The ormal requirements o a linear transer unction are: d = k and = 0 at x = 0. dx I the same input signal as above is applied to this nonlinear device, the output signal will be: ( = x ( = ( ( π + ( π ) = ( π + ( π ( π + ( π ( π + ( π The our terms can be rewritten as: ( π = ( ( π + ( π ) ( π ( π = ( π + ( π ( ) + ( π ( + ) ( π ( π = ( π + ( π ( ) + ( π ( + ) ( π = ( ( π + ( π ) Here it is obvious that a number o new requenc components have been created. I we examine the expression above, we ind:, are the original requenc components, no problem. The requenc components +, +,, are quite high requencies, compared to the desired ones, and can oten be iltered out easil. But and pose problems, since the are close to the desired requencies, and cannot be removed using ilters.

In this case, the nonlinearit had the exponent, and the requenc o the problematic intermodulation products were and. I the exponent had been 5, the problematic requencies would have been and. For exponent 7, and would be created. In the general case, this group o intermodulation products, can be expressed as IM = m n, where m and n are integers. The order o the intermodulation products are obtained as m + n. So ar, onl odd exponents, i.e. intermodulation orders have been studied. It is an interesting act, that even numbered orders never produce requenc components close to the desired, original requencies, and thereore, in most cases do not present a problem. Nonlinearities The nonlinear mechanisms considered in this context are passive. Passive means that the device does not have a power suppl. Examples: connectors, cables, antennas etc. It ma even suggest that mechanical, non-electric parts, e.g. cable clamps, handles and bolts can act as passive nonlinearities. Nonlinearit oten gets more pronounced at higher signal levels, i.e. strong RF-currents. There are basicall two situations, where a part ma carr strong RF-currents. The irst case is conducted current. Current originating rom e.g. a strong radio transmitter. A tpical situation is RF-current lowing in cables, connectors, cable joints and antennas. I, or instance, a connector act nonlinearl, intermodulation products ma be created. The second case is current induced b radiation. Metallic parts in the vicinit o a transmitter antenna, will pick up RF-power rom the electromagnetic ield and convert it into a RF-current in the part. I the part has a nonlinear behavior, intermodulation products ma occur, which will then be reradiated as wireless intererence. But wh do metallic parts have nonlinear current behavior? There are mainl two mechanisms involved. The irst one is the properties o the conducting material itsel. For example, some magnetic materials ma exhibit a nonlinear perormance, due to the act that the current causes magnetic ields, and that magnetization curves are nonlinear or strong currents. In other cases, the nonlinearit o a material ma be caused b polarization issues. The second mechanism is caused b surace eects in the interace between two materials. For example, oxidation on a contact surace ma contribute a MIM-diode. MIM stands or Metal- Insulator-Metal. This is a nonlinear quantum eect. MIM-diodes suer rom poor mechanical stabilit, but the can operate at ver high requencies, above 0 THz. The have also been considered or solar energ conversion. So, to avoid PIM problems, select materials careull, and keep our connector suraces clean rom oxide and dirt. PIM in practice We will use a tpical GSM base station to illustrate the practical problems with PIM. The GSM sstem is a ull duplex sstem, which implies that all transmitters and receivers will operate simultaneousl. To avoid intererence, dierent requencies are used. In or instance the P-GSM-900 band, an uplink band is deined between 890.0-95.0 MHz, and a downlink band between 95.0-960.0 MHz. The uplink is rom the handset to the base station, and the downlink the other wa round.

Tpicall, we will ind a number o prett strong radio transmitters transmitting in the downlink band, while a number o sensitive radio receivers operate in the uplink band. Obviousl, we do not want an signals rom the downlink to interere with the delicate signals in the uplink. As long as everbod stas on their allocated requenc, all will work ine. However, intermodulation has the nast eect o creating new requencies that are not expected... The downlink band harbors channels. I we transmit on p channels, there ma occur p ( p ) problematic intermodulation products. This means that in worst-case, 55 intermodulation products will be generated. This is a simpliied example, in realit there are man more requenc components to take into account (due to the modulation o the signals). Oten, the intermodulation products ma not be experienced as discrete requencies, but rather as a general increase in the noise loor. Where in the requenc band ma the intermodulation products show up then? Doing some calculations, varing the transmitting requencies and rom the lower to the upper limit o the downlink, we can ind the location o the intermodulation products. For third order products the requenc o the intermodulation products are given b and, or the ith order b and and or the seventh order b and. As can be seen in igure, the intermodulation products occur pair wise and smmetricall round the downlink band. Yellow is the downlink. Purple is the third order, blue ith order and pink seventh order intermodulation products. (Onl intermodulation products outside the downlink band shown). Figure, requenc ranges o uplink, downlink and intermodulation products outside the downlink The green band in igure is the uplink. It is clear that the requenc range o the intermodulation products overlap considerabl, and that the risk or intererence is imminent. Raising the noise loor, in the sensitive uplink, degrades the perormance o the radio links and the base station, thus reducing the revenue. So, at last, a word o wisdom: Sta linear! Dag Stranneb Dag is perorming PIM studies at Campus Alred Nobel, Örebro Universit, in collaboration with Nolato AB in Hallsberg, Sweden. More inormation can be ound at: http://vimeo.com/dstranneb/pim