state_of_rf_03_07_23.lyx

This a framework to test new ideas in transmission technology. Actual development is a LDPC-coder in

#LyX 1.3 created this file. For more info see http://www.lyx.org/
\lyxformat 221
\textclass article
\language english
\inputencoding auto
\fontscheme default
\graphics default
\paperfontsize default
\spacing single 
\papersize Default
\paperpackage widemarginsa4
\use_geometry 0
\use_amsmath 0
\use_natbib 0
\use_numerical_citations 0
\paperorientation portrait
\secnumdepth 3
\tocdepth 3
\paragraph_separation indent
\defskip medskip
\quotes_language english
\quotes_times 2
\papercolumns 1
\papersides 1
\paperpagestyle default

\layout Title

State of the different RF-cards
\layout Date

\SpecialChar ~

\layout Abstract

Created 2003-07-17
\layout Section

Existing RF-cards from STMicroelectronics
\layout Standard

The RF-cards from STMicroelectronics have been built in 2001.
 Since then they haven't been touched.
\layout Subsection

Architecture
\layout Standard

The RF-cards have the A/D and D/A on-board and communicate through a digital
 interface with the digital aquistion card.
 There are four boards:
\layout Enumerate

PCI aquisition card
\layout Enumerate

A/D and D/A card
\layout Enumerate

RF-frontend
\layout Enumerate

Clock generator
\layout Standard

In order to work, these cards need another external signal-generator to
 give the carrier-frequency.
\layout Subsection

Proprieties
\layout List
\labelwidthstring 00.00.0000

Carrier 2.1 GHz
\layout List
\labelwidthstring 00.00.0000

Bandwith 3 MHz
\layout Subsection

State of the card
\layout Standard

These cards are in a working state, but the following problems exist:
\layout Itemize

High 
\begin_inset Formula $SNR_{IIS}$
\end_inset 

due to hardware-limitation
\layout Itemize

Difficult (if not impossible) to calibrate
\layout Section

Philips-card
\layout Standard

In Eurecom, a re-build of the previous card is in progress.
\layout Subsection

Architecture
\layout Standard

These cards also have the A/D and D/A on-board.
 But all clock-generators are on board.
 So only two cards are needed:
\layout Enumerate

PCI-aquisition card
\layout Enumerate

RF-frontend
\layout Standard

This, together with an antenna and a PC, gives a working system.
\layout Subsection

Proprieties
\layout List
\labelwidthstring 00.00.0000

Carrier 1.9GHz
\layout List
\labelwidthstring 00.00.0000

Bandwith 10MHz
\layout Subsection

State of the card
\layout Subsubsection

RF-frontend
\layout Standard

Prototye has been confirmed to work.
 Sub-contractor has been found that builds the card and does testing/debugging
 of all the components.
 Cost per card: 6000 Euros.
\layout Subsubsection

Aquisition card
\layout Standard

Prototype has been done here at EPFL.
 Some signals still need changes.
 Time to work and debug: 1 week.
 Time to build: 1 month.
 Cost per card: 1000 Euros
\layout Section

EPFL-development
\layout Standard

In order to do MIMO measurements and development, an own effort from EPFL
 has been started.
\layout Subsection

Architecture
\layout Standard

Aquisition and data-generation cards that operate on an IF of 70MHz, whose
 signal is up-converted to the carrier-frequency with an RF-frontend.
 For 4 full-duplex antennas, the following cards are needed:
\layout Enumerate

PCI 4xA/D card
\layout Enumerate

PCI 4xD/A card
\layout Enumerate

Four RF-frontends
\layout Subsection

Proprieties
\layout List
\labelwidthstring 00.00.0000

Carrier 2.4GHz - 2.48GHz
\layout List
\labelwidthstring 00.00.0000

Bandwith 10MHz - 20MHz
\layout Subsection

State of the card
\layout Subsubsection

Aquisition card
\layout Standard

The cards exist as ready-to-buy cards from ICS.
 Demand of order has been sent to SIC.
 Jose Gonzalez is in charge of it, but he's on vacation till 5th of August.
 Cost: 20000 Euros for A/D and D/A cards together.
\layout Subsubsection

RF-Frontend
\layout Standard

Brecken Uhl from Radio Innovation did a preliminary design study including
 logical layout and component list for the RF-frontend.
 Expected arriving time: end 2003.
 Cost: 5000$ for preliminary study (nearly finished), 20000$ (to be confirmed,
 might be higher) for building and testing and re-building the cards.
 These developement cost exceed the product cost.
 One RF-frontend is thought to cost 200$ for material.
\layout Subsection

Alternatives for RF-frontend
\layout Standard

At ETHZ Daniel Baum did a very simple FDD(?) setup for translating the IF-freque
ncy to the carrier-frequency.
 Cost is about 1000$ per RF-frontend.
\the_end