Viterbi decoder on STI CELL processor Michal Blaek - PowerPoint PPT Presentation

Feb 22, 2023 •203 likes •274 views

IBM - CVUT Student Research Projects Viterbi decoder on STI CELL processor Michal Blaek (blazem2@fel.cvut.cz) Viterbi algorithm introduction Finds the most likely state trajectory given the Works in several steps HMM model and

IBM - CVUT Student Research Projects Viterbi decoder on STI CELL processor Michal Blažek (blazem2@fel.cvut.cz)
Viterbi algorithm introduction • Finds the most likely state trajectory given the • Works in several steps HMM model and observation’s sequence. 1. initializa tion : ≤ ≤ for 1 i N ( ) δ = π ⋅ ( i ) b O 1 i i 1 2. recursion : ≤ ≤ ≤ ≤ for 2 t T, 1 j N [ ] ( ) ( ) ( ) δ = δ ⋅ ⋅ j max i a b O − t t 1 ij j t ≤ ≤ 1 i N [ ] ( ) ( ) ψ = δ ⋅ j arg max i a − t t 1 ij ≤ ≤ 1 i N 3 . terminati on : [ ( ) ] = δ * P max i T ≤ ≤ 1 i N [ ] ( ) = δ * q arg max i T T ≤ ≤ 1 i N 4 . backtracki ng = for t T - 1, T - 2,...,1 ( ) = ψ ⋅ * * q q + + t t 1 t 1 IBM - CVUT Student Research Projects 2
ALF introduction (Accelerated Library Framework) • Programming environment for data and task parallel applications. • Supports the MPMD (multiple-program-multiple- data) programming model. • Optimal tasks scheduling scheme based on tasks dependencies. • Two components: host runtime X accelerator runtime • Input + output + parameters = work block IBM - CVUT Student Research Projects 3
Program realization • Algorithm is divide in four separated TASKs creating a simple pipeline. • Between particular neighbouring TASKs is defined a time dependency for running in right order. • TASKs which using ALF are computed in parallel on 1-6 SPEs. • Program uses “Host data partitioning” method for partition data across the accelerators (SPEs). IBM - CVUT Student Research Projects 4
Conclusion and results • Working correctly. • The most difficult part (recursion) is not completed yet with ALF, for checking results is for now working on PPU(using SIMD) only. • No optimalizations and tune. • Orientation computation time comparison on the same data set (HMM-500x500x500, seq-500): – This: 10,5 sec – Old (1SPE, libspe2, SIMD): 18,3 sec IBM - CVUT Student Research Projects 5

Recommend

F F Fast Transforms using the Cell/B.E. Processor Fast Transforms using the Cell/B.E. Processor

F F Fast Transforms using the Cell/B.E. Processor Fast Transforms using the Cell/B.E. Processor t T t T f f i g th C ll/B E P i g th C ll/B E P David A. Bader joint work with Seunghwa Kang and Virat Agarwal Sony-Toshiba-IBM Center of

612 views • 47 slides

Hardware Architecture of the Cell Broadband Engine Processor LOGO Presented by Wei Wei,

Hardware Architecture of the Cell Broadband Engine Processor LOGO Presented by Wei Wei, 04/20/2009 The CELL/B.E. processor The Cell Broadband Enginee (Cell/B.E.) processor is the first implementation of a new multiprocessor family conforming to

470 views • 22 slides

Optimization algorithms on Cell processor Vladim r T rebick y Optimization algorithms

Optimization algorithms on Cell processor Vladim r T rebick y Optimization algorithms on Cell processor p. 1/6 Cell Power Processor Element (PPE) I/O Controller Memory Controller + Explicitly parallel Element Interconnect Bus

472 views • 6 slides

Scalable String Matching on the Scalable String Matching on the Scalable String Matching on the

Scalable String Matching on the Scalable String Matching on the Scalable String Matching on the Cell BE Processor BE Processor Cell Cell BE Processor Daniele Scarpazza, Oreste Villa, Fabrizio Petrini Applied Computer Science Group Pacific

408 views • 21 slides

Communication Analysis of the Communication Analysis of the Communication Analysis of the Cell

Communication Analysis of the Communication Analysis of the Communication Analysis of the Cell Broadband Engine Processor Cell Broadband Engine Processor Cell Broadband Engine Processor Fabrizio Petrini Pacific Northwest National Laboratory

709 views • 27 slides

Intro This talk will focus on Cell processor Cell Broadband Engine Architecture (CBEA)

Intro This talk will focus on Cell processor Cell Broadband Engine Architecture (CBEA) Power Processing Element (PPE) Synergistic Processing Element (SPE) Current implementations Sony Playstation 3 (1 chip with 6 SPEs)

709 views • 23 slides

Digital Design Disc: RTL Combinatorial Components 2-to-4 Decoder 4-to-16 Decoder 8-bit Shifter

Principles Of Digital Design Disc: RTL Combinatorial Components 2-to-4 Decoder 4-to-16 Decoder 8-bit Shifter Absolute Differences Delay in Adder/Subtractor Comparator 6-to-1 Selector 2-to-4 Decoders Build a 2-to-4 decoder using 1-to-2

605 views • 27 slides

Introduction Playstation 3 (PS3) Game Console Cell Processor Molecular Dynamic (MD)

PS3 GRID .net Building a distributed supercomputer using the Playstation 3 M. J. Harvey G. Giuppone J. Vill-Freixa G. De Fabritiis Presented by: Abadi Kurniawan Introduction Playstation 3 (PS3) Game Console Cell Processor

559 views • 17 slides

6.02 Fall 2012 Lecture #7 Viterbi decoding of convolutional codes Path and branch metrics

6.02 Fall 2012 Lecture #7 Viterbi decoding of convolutional codes Path and branch metrics Hard-decision & soft-decision decoding Performance issues: decoder complexity, post- decoding BER, free distance concept 6.02 Fall 2012

488 views • 20 slides

4 1 3 2 Instruction ALU Registers Memory Fetch and Decode Building Blocks Processor

Processor: Data Path Components 4 1 3 2 Instruction ALU Registers Memory Fetch and Decode Building Blocks Processor datapath Microarchitecture Instruction Decoder Memory Arithmetic Logic Unit Adders Registers Multiplexers

412 views • 14 slides

Systems Architecture The ARM Processor The ARM Processor p. 1/14 The ARM Processor ARM:

Systems Architecture The ARM Processor The ARM Processor p. 1/14 The ARM Processor ARM: Advanced RISC Machine First developed in 1983 by Acorn Computers ARM Ltd was formed in 1988 to continue development Advantages of the ARM

391 views • 14 slides

Search and Decoding Lecture 16 CS 753 Instructor: Preethi Jyothi Recall Viterbi search Viterbi

Search and Decoding Lecture 16 CS 753 Instructor: Preethi Jyothi Recall Viterbi search Viterbi search finds the most probable path through a trellis of time on the X-axis and states on the Y-axis q F end end end end v 1 (2) =.32 v 2

934 views • 23 slides

Scheduling complex streaming applications on the Cell processor Mathias Jacquelin, joint work

Scheduling complex streaming applications on the Cell processor Mathias Jacquelin, joint work with Matthieu Gallet and Loris Marchal INRIA ROMA project-team LIP (ENS-Lyon, CNRS, INRIA) Ecole Normale Sup erieure de Lyon, France Workshop

804 views • 79 slides

UN13750 Programmable Encoder/Decoder Single chip contains both Encoder and Decoder. Schmitt

UN13750 Programmable Encoder/Decoder Single chip contains both Encoder and Decoder. Schmitt Trigger input provides excellent noise immunity. 3V to 11 V operation. Applications. alarm control system, security system On chip oscillator uses

620 views • 6 slides

Benchmarking the Memory Interface Controller bus of the Cell processor Nathalie Casati EPFL

CBEA The benchmarks Getting the full bandwidth Bandwidth graphs Results analysis Whats missing? Benchmarking the Memory Interface Controller bus of the Cell processor Nathalie Casati EPFL December 18, 2007 CBEA The benchmarks Getting

431 views • 20 slides

Pentium 4 Deeply pipelined processor supporting multiple issue with speculation and

Pentium 4 Deeply pipelined processor supporting multiple issue with speculation and multi-threading 2004 version: 31 clock cycles from fetch to retire, 3.2 GHZ clock rate (deep pipeline allows higher clock rate) Front end decoder

692 views • 22 slides

Language, Dialect, and Speaker Recognition Using Gaussian Mixture Models on the Cell Processor

Language, Dialect, and Speaker Recognition Using Gaussian Mixture Models on the Cell Processor Nicolas Malyska, Sanjeev Mohindra, Karen Lauro, Douglas Reynolds, and Jeremy Kepner {nmalyska, smohindra, karen.lauro, reynolds, kepner}@ll.mit.edu

720 views • 60 slides

FPGA co-processor Patrick Dunne for the co-processor group Introduction Co-processor will

FPGA co-processor Patrick Dunne for the co-processor group Introduction Co-processor will take care of data compression and trigger primitive generation Co-processor will also perform data buffering for supernova trigger Sits on FELIX

522 views • 13 slides

An Efficient GPU-based An Efficient GPU-based LDPC Decoder for Long LDPC Decoder for Long

An Efficient GPU-based An Efficient GPU-based LDPC Decoder for Long LDPC Decoder for Long Codewords Codewords Stefan Grnroos Kristian Nybom Jerker Bjrkqvist 18.10.11 bo Akademi University - Turku, Finland 1 Background Background

301 views • 11 slides

Contents PRO-Decoder Function Methods Results Abstract Experiment Computer RBS-Decoder

Contents PRO-Decoder Function Methods Results Abstract Experiment Computer RBS-Decoder TER-Decoder Transcriptional level Translational level Transcriptional level PRO-Decoder Synoproteinerc Transcriptional level Translational level

1.25k views • 87 slides

Exercise 2: Encoder / Decoder Framework Goals : Implement basic framework for encoder and decoder

Exercise 2: Encoder / Decoder Framework Goals : Implement basic framework for encoder and decoder (without any real data processing) Implement block-wise processing of samples Encoder Operation Read image in PGM format Write data to bitstream

153 views • 3 slides

Processor Design Pipelined Processor Hung-Wei Tseng Drawbacks of a single-cycle processor

Processor Design Pipelined Processor Hung-Wei Tseng Drawbacks of a single-cycle processor The cycle time is determined by the longest instruction Could be very long, thinking about fetch data from DRAM Hardware is mostly idle 3

866 views • 47 slides

Cell Communication and Cell Signaling Why is cell signaling important? Why is cell signaling

Cell Communication and Cell Signaling Why is cell signaling important? Why is cell signaling important? Allows cells to communicate and coordinate functions/activities of the organism Usually involves the cell membrane Cell

408 views • 36 slides

CSE 140 Discussion Session 1 Decoder A digital module that converts a binary address to the

CSE 140 Discussion Session 1 Decoder A digital module that converts a binary address to the assertion of the addressed device EN (enable) y 0 0 I 0 0 y 1 1 2 . 1 I 1 3 . 4 5 I 2 2 6 y 7 7 n to 2 n decoder 2 n outputs n inputs

308 views • 11 slides