<p>FIR (Finite-length Impulse Response) filters are the corner stone of many signalprocessing devices. A lot of research has gone into their development as wellas their effective implementation. With recent research focusing a lot on powerconsumption reduction specially with regards to FPGAs, it was found necessaryto explore FIR filters mapping on FPGAs.</p><p>Time multiplexed FIR filters are also a good candidate for examination withrespect to power consumption and resource utilization, for example when implementedin Field Programmable Gate Arrays (FPGAs). This is motivated by thefact that the usable clock frequency often is higher compared to the required datarate. Current implementations by, e.g., Xilinx FIR Compiler suffer from highpower consumption when the time multiplexing factor is low. Further, it needs tobe investigated how exploiting coefficient symmetry, scaling the coefficients andincreasing the time-multiplexing factor influences the performance.</p>

<p>Nowadays, FPGAs are large enough to host entire system-on-chip designs, wherein a soft core processor is often an integral part. High performance of the processor is always desirable, so there is an interest in finding faster solutions.This report aims to describe the work and results performed by Karl Bengtson and Olof Andersson at ISY. The task was to continue the development of a soft core microprocessor, originally created by Andreas Ehliar. The first step was to decide a more widely adopted instruction set for the processor. The choice fell upon the MIPS32 instruction set. The main work of the project has been focused on implementing support for MIPS32, allowing the processor to execute MIPS assembly language programs. The development has been done with speed optimization in mind. For every new function, the effects on the maximum frequency has been considered, and solutions not satisfying the speed requirements has been abandoned or revised.The performance has been measured by running a benchmark program—Coremark. Comparison has also been made to the main competitors among soft core processors. The results were positive, and reported a higher Coremark score than the other processors inthe study. The processor described herein still lacks many essential features. Nevertheless, the conclusion is that it may be possible to create a competitive alternative to established soft processors.</p> / <p>FPGAer används idag ofta för stora inbyggda system, i vilka en mjuk processor ofta spelar en viktig roll. Hög prestanda hos processorn är alltid önskvärt, så det finns ett intresse i att hitta snabbare lösningar. Denna rapport skall beskriva det arbete och de resultat som uppnåtts av Karl Bengtson och Olof Andersson på ISY. Uppgiften var att fortsätta utvecklandet av en mjuk processor, som ursprungligen skapats av Andreas Ehliar. Första steget var att välja ut en mer allmänt använd instruktionsuppsättning för processorn. Valet föll på instruktionsuppsättningsarkitekturen MIPS32. Projektets huvutarbete har varit fokuserat på att implementera stöd för MIPS32, vilket ger processorn möjlighet att köra assemblerprogram för MIPS.Utvecklingen har gjorts med hastighetsoptimering i beaktning. För varje ny funktion har dess effekter på maxfrekvensen undersökts,och lösningar som inte uppfyllt hastighetskraven har förkastats eller reviderats. Prestandan har mätts med programmet Coremark. Det har också gjorts jämförelser med huvudkonkurrenterna bland mjuka processorer. Resultaten var positiva, och rapporterade ett högre Coremarkpoäng än de andra processorerna i studien. Slutsatsen är att det ärmöjligt att skapa ett alternativ till de etablerade mjuka processorerna, men att denna processor fortfarande saknar väsentliga funktioner som behövs för att utgöra en mogen produkt.</p>

Ett digitalt vågfilter av Richardstyp har implementerats i en FPGA på ett utvecklingskort. Sampel kan skickas till filtret och mottas från filtret via serieporten på en dator. Metoden som användes är att en modell av filtret konstruerades i Simulink. Filtret har modifierats med avseende på skalning, brus och stabilitet. VHDL-koden till filtret genererades i Simulink genom att bygga modellen av Xilinx Blockset. Ytterligare VHDL-kod konstruerades för att kunna skicka sampel mellan filter och minnet på utvecklingskortet. För kommunikation mellan minnet på utvecklingskortet och dator utnyttjades färdiga lösningar. / Filtrets funktion efter implementeringen var samma som modellens byggd i Simulink. A Richards’ structure wave digital filter has been implemented on an evaluation board in an FPGA. Samples can be sent to the filter and received from the filter using the serial port of a computer. The method used is that a modell of the filter has been created in Simulink. The filter has been modified with respect to scaling, noise and stability. VHDL for the filter has been generated in Simulink by using Xilinx blockset to build the modell. Also, VHDL has been constructed to be able to send samples between the filter and the memory on the evaluationboard. For communication between the memory on the evaluationboard and the computer, existing solutions have been used. The functionality of the filter after implementation was the same as in the modell built in Simulink.

The aim of this thesis was to configure an FPGA with high speed ports to capture data from a prototype 4 bit ΣΔ analogue-to-digital converter sending data at a rate of 2.4 Gbps in four channels and to develop a protocol for transferring the data to a PC for analysis. Data arriving in the four channels should be sorted into 4 bit words with one bit taken successively from each of the channels. A requirement on the data transfer was that the data in the four channels should arrive synchronously to the FPGA. A Virtex-5 FPGA on a LT110X platform was used with RocketTMIO GPT transceivers tightly integrated with the FPGA logic. Since the actual DUT (Device Under Test) was not in place during the work, the transceivers of the FPGA were used for both sending and receiving data. The transmission was shown to be successful for both eight and ten bit data widths. At this stage a small skew between the data in the four channels was observed. This was solved by storing the information in separate memories, one for each of the channels, to make possible to later form the 4 bit words in the PC (MatLab). The memories were two port FIFOs writing in data at 240 MHz (10 bit data width) or 300 MHz (8 bit data width) and read out at 50 MHz. / Syftet med examensarbetet var att konfigurera en FPGA med höghastighetsportar så att data från en prototyp av en 4 bitars ΣΔ analog-till-digital omvandlare kan samlas in med en hastighet av 2.4 Gbps i var och en av fyra kanaler och att utveckla ett protokoll för överföring av dessa data från FPGAn till en PC för analys. Insamlade data ska sorteras i 4 bitars ord med en bit successivt tagen från var och en av kanalerna. Ett krav på dataöverföringen var att data i de fyra kanalerna skulle anlända synkront till FPGAn. En Virtex-5 FPGA på en LT110X plattfrom användes med GTP transceivrar tätt integrerade med FPGA logiken. Då utrustningen som skulle testas inte var tillgänglig under tiden arbetet utfördes användes FPGAns transceivrar till att både sända och ta emot data. Överföring av data med både 8 och 10 bitars datavidd uppnåddes framgångsrikt. Data i de fyra kanalerna visade sig dock inte anlända synkront till mottagaren. Detta problem löstes genom att lagra informationen i separata minnen, ett för varje kanal, överföra data från minnena till PCn och där med hjälp av MatLab sortera dem till 4 bitars ord. Som minnen användes tvåportars FIFOn där data skrivs in med en hastighet av 240 MHz (10 bitars datavidd) eller 300 MHZ (8 bitars datavidd) och läses ut med en hastighet av 50 MHz.

FIR (Finite-length Impulse Response) filters are the corner stone of many signalprocessing devices. A lot of research has gone into their development as wellas their effective implementation. With recent research focusing a lot on powerconsumption reduction specially with regards to FPGAs, it was found necessaryto explore FIR filters mapping on FPGAs. Time multiplexed FIR filters are also a good candidate for examination withrespect to power consumption and resource utilization, for example when implementedin Field Programmable Gate Arrays (FPGAs). This is motivated by thefact that the usable clock frequency often is higher compared to the required datarate. Current implementations by, e.g., Xilinx FIR Compiler suffer from highpower consumption when the time multiplexing factor is low. Further, it needs tobe investigated how exploiting coefficient symmetry, scaling the coefficients andincreasing the time-multiplexing factor influences the performance.

The development in electronics increases the demand for good design methods and design tools in the field of electrical engeneering. To improve their design methods Ericsson Microwave Systems AB is interested in using computer tools to create a link between the specification and the implementation of a digital system in a FPGA. Xilinx System Generator for DSP is a tool for implementing a model of a digital signalprocessing algorithm in a Xilinx FPGA. To evaluate Xilinx System Generator two testcases has been designed. The testcases are selected to represent the FPGA designs made at Ericsson Microwave Systems. The testcases show that Xilinx System Generator can be used to effectivly implement a model made in Simulink in a FPGA from Xilinx. The result of the implementation is comparable to the implementation of VHDL code written by hand. The use of tools for implementation of a model in hardware cause change in the design methods used at Ericsson Microwave Systems. The higher level of abstraction introduced by System Generator results in the design decisions made at system level having a higher impact on the final realization.

Today, wireless communications is used more then ever before. Wired systems are replaced with wireless versions. New methods and transmission standards are developed and tested. The purpose of this thesis is development of a flexible high-performance System Test Platformfor test of the SoCTRix Wireless LAN Transceiver. The result is a Xilinx Virtex-II FPGA based System Test Platform board with CardBus PC Card interface to a computer. The hardware achieved has the following features: - 8-layer PCB - PCMCIA CardBus PC Card interface, enabling 133 MB/s data throughput - 1M Gate Virtex-II FPGA with reprogrammable configuration memory - Debugging via LEDs and Logic Analyzer connectors - 2x SPI EEPROM - 40 MHz system clock - Easy connection of two daughter-boards Specially designed for wireless transmitter development, can also be used for other computer related highperformance applications.

Digital Radio Frequency Memory (DRFM) is a technique widely used by the defense industry in, for example, electronic countermeasure equipment for generating false radar targets. The purpose of the DRFM technique is to use high-speed sampling to digitally store and recreate radio frequency and microwave signals. At Saab Bofors Dynamics AB the technique is used, among others, in the Electronic Warfare Simulator (ELSI). The DRFM technique is implemented in a full-custom ASIC circuit that has been mounted on circuit boards in ELSI. Today, the progress in the programmable hardware field has made it possible to implement the DRFM design in a Field Programmable Gate Array (FPGA). The FPGA technology has many advantages over a full custom ASIC design. Hence, the purpose of this master's thesis has been to develop a new DRFM design that can be implemented in an FPGA, using a hardware description language called VHDL. The method for this master's thesis has been to first establish a time plan and a requirement specification. After that, a design specification has been worked out based on the requirement specification. The two specifications have served as a basis for the development of the DRFM circuit. One of the requirements on the design was that the circuit should be able to communicate through an external Ethernet interface. A part of the work has, thus, been to review available external Ethernet modules on the market. The result is a DRFM design that has been tested through simulations. The tests shows that the design works as described in the design specification.

In this thesis, we introduce a methodology to design and implement a Homeplug1.0 channel decoder that is completely conforming to Homeplug 1.0 specifications definedin HomePlug Power-line Alliance Standard (HPA) including Reed-solomon decoding,Viterbi decoding, punctured ,and de-interleaving technologies. Further, by using MATLAB/Simullink software, Xilinx System Generator, Xilinx Alliance tools, XilinxISE and Modelsim SE software, we build up a transceiver platform to simulate and analyze the performance of the power-line channel decoder based on FPGA hardware implementation. The hardware can be used directly in practical Homeplug 1.0 systems.

The rapid growth of communication technology with the success of internet, has brought huge profits and great convenience to our daily life. Computer networks can be built using either wired or wireless technologies. It will be an important issue that how to select a medium for the transmission. Wired Ethernet has been the traditional choice in most of the networks. However, it has to deploy the Ethernet wires. For the wired internet networks, the power line communication (PLC) technology will be an alternative choice. In this wire-line communication system, the power line network is used as the transmission medium. Therefore, computer networks can work on the existing power line system. No extra new transmission infrastructure is needed. So far, several PLC standards are available, shch as X-10, CEBus(Consumer Electronic Bus), Echonet and Homeplug. This thesis studies the Homeplug AV specification developed by the Homeplug powerline Alliance. By employing MATLAB/Simulink, we build up a PLC baseband transmission model and simulation platform. We carry out the Homeplug AV baseband transmission performance in system level on this platform. The Homeplug AV adopts 3072-point FFT which is not the power of two. It will be a challenge to design the 3072-point FFT processor. Here, we use Xilinx System Generator to design and implement the 3072-point FFT processor. The function verification of the implemented 3072-point FFT processor for Homeplug AV system is carried out by simulation.