1 Combining the strengths of UMIST and The Victoria University of Manchester Asynchronous Signal Processing Systems Linda Brackenbury APT GROUP, Computer Science University of Manchester

2 Combining the strengths of UMIST and The Victoria University of ManchesterAgenda Why asynchronous? Applications suited to asynchronous Design examples – DSP design – Viterbi decoder Future work What have we learnt?

3 Combining the strengths of UMIST and The Victoria University of Manchester System Timing Synchronous – uses global clock – any state changes occur on clock edge – system states predictable so good tools Asynchronous – uses events to control timing – timing is more unpredictable – tool support not as good

4 Combining the strengths of UMIST and The Victoria University of Manchester Why Asynchronous? No clock generation or distribution – timing uses local handshake signals Power only consumed when doing useful work No overhead between idle and active Low EMI – switching is spread

5 Combining the strengths of UMIST and The Victoria University of ManchesterApplications Async - no help to some applications – power/performance at full activity similar for synchronous and asynchronous! Async good for portable systems – battery size and lifetime is important – workload is highly variable – lots of idle time – low EMI requirement

6 Combining the strengths of UMIST and The Victoria University of Manchester Low Power DSP GSM chipsets are typically based on microprocessor + DSP DSP performs intensive calculations Challenge is to meet required throughput without excessive power – throughput met with parallelism – area traded for increased speed

7 Combining the strengths of UMIST and The Victoria University of Manchester Asynchronous Contribution Design Philosophy – optimize design for typical operation – support design for rarer conditions usually at expense of increased operation time Simpler logic within processing units – energy and area reduction

8 Combining the strengths of UMIST and The Victoria University of Manchester DSP Design Examples 1 Data dependent adder on critical path – detect completion of carry path – average carry path only half word length Address wrap around in circular buffer – synchronous calculates new and possible corrected value in parallel - two adders – asynchronous new value only – one adder if correction required (rare) this done after

9 Combining the strengths of UMIST and The Victoria University of Manchester DSP Design Examples 2 Register File has eight single-read single-write ported 32-word banks – efficient parallel access to sequential registers from 4 Functional Units (typical) Request conflicts to same bank rare – broadcast mechanism available – genuine conflicts take 1 read cycle per request rather than 1 clock cycle each

10 Combining the strengths of UMIST and The Victoria University of Manchester Viterbi Decoder Two data streams transmitted depends on current and previous data State transitions of encoder with time can be drawn as a trellis Decoder reconstructs trellis

11 Combining the strengths of UMIST and The Victoria University of Manchester Asynchronous Decoder – Clock only used to input and output data all internal operation is asynchronous FIFOs buffer data to meet clock demand

12 Combining the strengths of UMIST and The Victoria University of Manchester Branch Metric Unit Calculates gap between input symbol and four ideal symbols

13 Combining the strengths of UMIST and The Victoria University of Manchester Path Metric Unit j+32 j 2j 2j+1BMa BMb add-compare- select operation previous node metric next node metric node

14 Combining the strengths of UMIST and The Victoria University of Manchester Node Arithmetic Serial arithmetic Counts events Unary numbers – Change of state equals count – one=1111 two=0001 three=1101 etc. When smaller count empties merge stops

15 Combining the strengths of UMIST and The Victoria University of Manchester History Unit Records PMU node winners and global winner over many timeslots No error -global winner is child of last winner Error – need to reconstruct good path – compute parent of global winner and repeat ONLY until it agrees with good path – can have many backtraces in parallel – backtraces decoupled from placing data into HU

16 Combining the strengths of UMIST and The Victoria University of Manchester History Unit

17 Combining the strengths of UMIST and The Victoria University of Manchester Low Power Contribution History Unit – much smaller – highly concurrent independent operation – computation performed minimised Path Metric Unit – most of power – smaller, simple, fast +/- units replace add-compare-select – idea simple but a lot of control complexity so dissipated a lot of power!

18 Combining the strengths of UMIST and The Victoria University of Manchester What Have We Learnt? Asynchronous is advantageous to some applications Asynchronous design looks very different from synchronous design Get very poor results if just translate from a synchronous design Design of asynchronous is harder – timing and control more complex to design