Lecture 9: Digital Signal Processors: Applications and ArchitecturesProcessor ApplicationsProcessor MarketsThe Processor Design SpaceMarket for DSP ProductsDSP ApplicationsAnother Look at DSP ApplicationsServing a range of applicationsWorld’s Cellular SubscribersCELLULAR TELEPHONE SYSTEMHW/SW/IC PARTITIONINGMapping onto a system on a chipExample Wireless Phone OrganizationMultimedia I/O ArchitectureMultimedia System on a ChipRequirements of the Embedded ProcessorsArea of processor cores = CostAnother figure of merit Computation per unit areaCode sizeBENCHMARKS - DSPstoneEvolution of GP and DSPEmbedded Systems vs. General Purpose Computing - 1Embedded Systems vs. General Purpose Computing - 2DSP vs. General Purpose MPUSlide 25TYPES OF DSP PROCESSORSNote of Caution on DSP ArchitecturesArchitectural Features of DSPsDSP Data Path: ArithmeticDSP Data Path: PrecisionDSP Data Path: Overflow?DSP Data Path: MultiplierDSP Data Path: AccumulatorDSP Data Path: RoundingData Path320C54x DSP Functional Block DiagramFIR Filtering: A Motivating ProblemBENCHMARKS - FIR FILTERMicro-architectural impact - MACMAC Eg. - 320C54x DSP Functional Block DiagramDSP MemoryConventional ``Von Neumann’’ memoryHARVARD ARCHITECTURE in DSPMemory ArchitectureEg. TMS320C3x MEMORY BLOCK DIAGRAM - Harvard ArchitectureEg. 320C62x/67x DSPDSP AddressingDSP Addressing: FFTBIT REVERSED ADDRESSINGDSP Addressing: BuffersCIRCULAR BUFFERSAddressingAddress calculation unit for DSPDSP Instructions and ExecutionADSP 2100: ZERO-OVERHEAD LOOPInstruction SetSpecialized Peripherals for DSPsSpecialized peripheralsTMS320C203/LC203 BLOCK DIAGRAM DSP Core Approach - 1995Summary of Architectural Features of DSPs1Kurt KeutzerLecture 9: Digital Signal Processors:Applications and ArchitecturesPrepared by: Professor Kurt KeutzerComputer Science 252, Spring 2000With contributions from:Dr. Jeff Bier, BDTI; Dr. Brock Barton, TI; Prof. Bob Brodersen, Prof. David Patterson2Kurt KeutzerProcessor ApplicationsGeneral Purpose - high performancePentiums, Alpha’s, SPARCUsed for general purpose software Heavy weight OS - UNIX, NTWorkstations, PC’sEmbedded processors and processor coresARM, 486SX, Hitachi SH7000, NEC V800Single programLightweight, often realtime OSDSP supportCellular phones, consumer electronics (e.g. CD players) Microcontrollers Extremely cost sensitiveSmall word size - 8 bit commonHighest volume processors by farAutomobiles, toasters, thermostats, ... IncreasingCostIncreasingvolume3Kurt KeutzerProcessor Markets$30B$9.3B/31%$5.7B/19%$10B/33%8-bitmicro16-bitmicroDSP32-bitmicro$5.2B/17%$1.2B/4%32 bit DSP4Kurt KeutzerThe Processor Design SpaceCostPerformanceMicroprocessorsPerformance iseverything& Software rulesEmbeddedprocessorsMicrocontrollersCost is everythingApplication specific architecturesfor performance5Kurt KeutzerMarket for DSP ProductsMixed/SignalAnalogDSPDSP is the fastest growing segment of the semiconductor market6Kurt KeutzerDSP Applications Audio applications• MPEG Audio• Portable audioDigital camerasWireless• Cellular telephones• Base stationNetworking• Cable modems• ADSL• VDSL7Kurt KeutzerAnother Look at DSP ApplicationsHigh-endWireless Base Station - TMS320C6000Cable modem gatewaysMid-endCellular phone - TMS320C540Fax/ voice serverLow endStorage products - TMS320C27Digital camera - TMS320C5000Portable phonesWireless headsetsConsumer audioAutomobiles, toasters, thermostats, ... IncreasingCostIncreasingvolume8Kurt KeutzerServing a range of applications9Kurt KeutzerWorld’s Cellular Subscribers01002003004005006007001993 1994 1995 1996 1997 1998 1999 2000 2001MillionsYearDigitalAnalogSource: Ericsson Radio Systems, Inc.Will providea ubiquitousinfrastructurefor wirelessdata as wellas voice10Kurt KeutzerCELLULAR TELEPHONE SYSTEMPHYSICALLAYERPROCESSINGRF MODEMCONTROLLER 1 2 3 4 5 67 8 90415-555-1212SPEECHDECODESPEECHENCODEA/DBASEBANDCONVERTERDAC11Kurt KeutzerHW/SW/IC PARTITIONINGPHYSICALLAYERPROCESSINGRF MODEMCONTROLLER 1 2 3 4 5 67 8 90415-555-1212SPEECHDECODESPEECHENCODEA/DBASEBANDCONVERTERDACANALOG ICDSPASICMICROCONTROLLER12Kurt KeutzerMapping onto a system on a chip RAMµCRAMDSPCOREASICLOGICS/PDMAphonebookprotocolkeypadintfccontrolS/PDMAspeechqualityenhancmentde-intl &decodervoicerecognitionRPE-LTPspeech decoder demodulatorand synchronizer Viterbi equalizer13Kurt KeutzerExample Wireless Phone OrganizationC540ARM714Kurt KeutzerMultimedia I/O Architecture Low Power BusRadioModemEmbedded ProcessorFifoVideoDecompVideoAudioFBFifoGraphicsPenSched ECC PactInterfaceDataFlowSRAM15Kurt KeutzerMultimedia System on a ChipFuture chips will be a mix of processors, memory and dedicated hardware for specific algorithms and I/OµPDSPComsVideo UnitcustomMemoryUplink RadioDownlink RadioGraphics OutVideo I/OVoice I/OPen InE.g. Multimedia terminal electronics16Kurt KeutzerRequirements of the Embedded ProcessorsOptimized for a single program - code often in on-chip ROM or off chip EPROMMinimum code size (one of the motivations initially for Java)Performance obtained by optimizing datapathLow costLowest possible areaTechnology behind the leading edgeHigh level of integration of peripherals (reduces system cost)Fast time to marketCompatible architectures (e.g. ARM) allows reuseable codeCustomizable coreLow power if application requires portability17Kurt KeutzerArea of processor cores = CostNintendo processorCellular phones18Kurt KeutzerAnother figure of meritComputation per unit areaNintendo processorCellular phones???19Kurt KeutzerCode sizeIf a majority of the chip is the program stored in ROM, then code size is a critical issueThe Piranha has 3 sized instructions - basic 2 byte, and 2 byte plus 16 or 32 bit immediate20Kurt KeutzerBENCHMARKS - DSPstoneZIVOJNOVIC, VERLADE, SCHLAGER: UNIVERSITY OF AACHENAPPLICATION BENCHMARKSADPCM TRANSCODER - CCITT G.721REAL_UPDATECOMPLEX_UPDATESDOT_PRODUCTMATRIX_1X3CONVOLUTIONFIRFIR2DIMHR_ONE_BIQUADLMSFFT_INPUT_SCALED21Kurt KeutzerEvolution of GP and DSPGeneral Purpose Microprocessor traces roots back to Eckert, Mauchly, Von Neumann (ENIAC)DSP evolved from Analog Signal Processors, using analog hardware to transform phyical signals (classical electrical engineering)ASP to DSP becauseDSP insensitive to environment (e.g., same response in snow or desert if it works at all)DSP performance identical even with variations in components; 2 analog systems
View Full Document