Efficient implementation of iterative receiver components

Channel encoder architecture for energy-efficient WSNs

In this work, we demonstrated that the introduction of a rate-1 encoder into the IEEE 802.15.4 PHYsical layer (PHY) can yield a 3.99 dB reduction in the transmission energy required for reliable transmission, without increasing the bandwidth or reducing the bit rate. In the transmitter, the rate-1 encoder is provided by a simple accumulator, which has a negligible energy consumption compared to the transmission energy saving that it affords. Meanwhile, an iterative decoding process is employed in the receiver. This approach redistributes energy consumption from the transmitter to the receiver, which is beneficial in star-topology Wireless Sensor Networks (WSNs) where the transmitting sensor nodes typically have scarce energy resources and the receiving central node typically has access to a plentiful energy resource, such as mains power.

Augmented IEEE 802.15.4 PHY

Welcome to ePrints Soton - ePrints Soton
The University of Southampton

Welcome to ePrints Soton

Welcome to the University of Southampton Institutional Research Repository, ePrints Soton. This repository contains details and, if available, downloads of our research output.

Information on this website should be updated via PURE, our research management system. For issues and queries on outputs and open access, please contact the ePrints team at eprints@soton.ac.uk or view the University's Pure support pages.

Search Repository

Search the repository using a full range of fields. Use the search field at the top of the page for a quick search.

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of https://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×
Welcome to ePrints Soton - ePrints Soton
The University of Southampton

Welcome to ePrints Soton

Welcome to the University of Southampton Institutional Research Repository, ePrints Soton. This repository contains details and, if available, downloads of our research output.

Information on this website should be updated via PURE, our research management system. For issues and queries on outputs and open access, please contact the ePrints team at eprints@soton.ac.uk or view the University's Pure support pages.

Search Repository

Search the repository using a full range of fields. Use the search field at the top of the page for a quick search.

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of https://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×

Parameter selection for energy-efficient channel decoder architectures

Practical implementations of channel decoders are typically sub-optimal, since well-designed approximations of the optimal decoder can closely approach the same performance, at a fraction of the complexity and therefore energy consumption. However, the design and parameterisation of attractive sub-optimal decoders has previously relied upon time-consuming Bit Error Ratio (BER) simulations. In this work, we demonstrated that the use of EXtrinsic Information Transfer (EXIT) charts can not only significantly expedite the design process, but can also offer unique insights into the causes of performance degradation as the sub-optimality of the decoder is adjusted.

EXIT chart analysis of optimal turbo code parameter values

Welcome to ePrints Soton - ePrints Soton
The University of Southampton

Welcome to ePrints Soton

Welcome to the University of Southampton Institutional Research Repository, ePrints Soton. This repository contains details and, if available, downloads of our research output.

Information on this website should be updated via PURE, our research management system. For issues and queries on outputs and open access, please contact the ePrints team at eprints@soton.ac.uk or view the University's Pure support pages.

Search Repository

Search the repository using a full range of fields. Use the search field at the top of the page for a quick search.

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of https://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×
Welcome to ePrints Soton - ePrints Soton
The University of Southampton

Welcome to ePrints Soton

Welcome to the University of Southampton Institutional Research Repository, ePrints Soton. This repository contains details and, if available, downloads of our research output.

Information on this website should be updated via PURE, our research management system. For issues and queries on outputs and open access, please contact the ePrints team at eprints@soton.ac.uk or view the University's Pure support pages.

Search Repository

Search the repository using a full range of fields. Use the search field at the top of the page for a quick search.

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of https://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×

Channel decoder architecture for energy-efficient WSNs

Previous channel decoder architectures have been designed for achieving high throughputs, at the cost of having high energy consumptions. However, low throughputs are typical in WSNs and energy consumption is the main concern. Motivated by this, we proposed a novel channel decoder architecture which is designed for achieving a low energy consumption. This is achieved by minimising the wastage of energy, by decomposing the entire decoding algorithm into just Add, Compare and Select (ACS) operations, which are executed by a parallel concatenation of highly-optimised ACS units. As a result, we achieved a 71% reduction in the energy consumption of previous decoder architectures.

Novel ACS unit

Welcome to ePrints Soton - ePrints Soton
The University of Southampton

Welcome to ePrints Soton

Welcome to the University of Southampton Institutional Research Repository, ePrints Soton. This repository contains details and, if available, downloads of our research output.

Information on this website should be updated via PURE, our research management system. For issues and queries on outputs and open access, please contact the ePrints team at eprints@soton.ac.uk or view the University's Pure support pages.

Search Repository

Search the repository using a full range of fields. Use the search field at the top of the page for a quick search.

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of https://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×

Li, L., Maunder, R.G., Al-Hashimi, B.M., Zwolinski, M. and Hanzo, L. (2013) Energy-conscious turbo decoder design: a joint signal processing and transmit energy reduction approach. IEEE Transactions on Vehicular Technology, 62 (8), 3627-3638.

Channel decoders for correcting both channel-induced transmission errors and voltage-scaling-induced timing errors

Channel codes may be employed in wireless transmission schemes for correcting transmission errors. This tolerance to channel-induced transmission errors allows reduced transmission energies to be used, at the cost of the additional signal processing energy consumption of the channel decoder. Voltage scaling may be employed for reducing the channel decoder’s energy consumption, but at the cost of potentially introducing timing errors. In this work, we propose channel decoder architectures that are capable of exploiting their inherent error correction capability for correcting not only transmission errors, but also timing errors.

Novel stochastic variable node

Perez-Andrade, I., Zuo, X., Maunder, R.G., Al-Hashimi, B.M. and Hanzo, L. (2013) Analysis of voltage- and clock-scaling-induced timing errors in stochastic LDPC decoders. 2013 IEEE Wireless Communications and Networking Conference (WCNC 2013), China. 07 - 10 Apr 2013. pp. 4293-4298. (doi:10.1109/WCNC.2013.6555268).

Fully-parallel turbo decoding having ultra-high processing throughput and ultra-low latency

State-of-the-art LTE-Advanced mobile phones support data rates of up to 500 Mbit/s. However, if ten mobile phones are connected to the same basestation, then it must support data rates of 5 Gbit/s and even higher data rates will be demanded in the future. The basestation uses turbo decoding to correct transmission errors, but until recently the fastest turbo decoders could only achieve data rates of up to 2.15 Gbit/s. In order to address this, we have developed a fully-parallel turbo decoder, which achieves data rates of 20 Gbit/s and beyond and processing latencies of below 1 us. This has been achieved by eliminating the data dependencies in the turbo decoder processing, dispelling the long-standing myth that this processing is inherently serial. Find out more on the Future Worlds webpage.

Fully-parallel turbo decoder

Maunder, R.G. (2015) A fully-parallel turbo decoding algorithm. IEEE Transactions on Communications, 63 (8), 2762-2775.

Brejza, Matthew F., Li, Liang, Maunder, Robert G., Al-Hashimi, Bashir, Berrou, Claude and Hanzo, Lajos (2015) 20 years of turbo coding and energy-aware design guidelines for energy-constrained wireless applications. IEEE Communications Surveys & Tutorials, 1-21.

23 Responses to “Efficient implementation of iterative receiver components”

  1. Srinivas Says:

    Can U provide the MAtlab Code for this sir

  2. Rob Says:

    Hello Srinivas,

    I’m afraid that these studies were not completed in Matlab. You can download some of our C code from…
    http://users.ecs.soton.ac.uk/rm/resources/cfixedbcjr/

    Take care, Rob.

  3. Ideal Says:

    Dear Rob,

    Have you implemented puncturing in LDPC code or any simple matlab code which implement this puncturing concept? Thanks

  4. Rob Says:

    Hi Ideal,

    Puncturing is quite easy in Matlab. For example, in the transmitter, you might puncture some bits using…
    a2 = reshape(a,2,length(a)/2);
    b = a2(1,:);
    In the receiver the inverse operation for LLRs is…
    a2_tilde = [b_tilde; zeros(1,length(b_tilde)];
    a_tilde = reshape(a2_tilde,1,numel(a2_tilde));

    Take care, Rob.

  5. naveen Says:

    respected sir,
    need a matlab code for decoding ldpc codes using stochastic computation.

  6. Rob Says:

    Hi Naveen,

    I’m afraid that we don’t have Matlab code for the stochastic LDPC decoder -
    we did this work using the IT++ libraries for C++.

    Take care, Rob.

  7. iklass souhir Says:

    Pr Rob Maunder,
    Bonjour, je suis très heureuse de trouver des codes sous matlab du turbocode que vous avez mi à notre disposition, j’aimerai SVP que vous m’envoyez un programme en matlab du récepteur râteau(RAKE receiver) de l’UMTS à l’adresse email iklass_s@yahoo.fr .
    Merci beaucoup
    iklass

  8. Rob Says:

    Hello Iklass,

    I’m afraid that I don’t have any Matlab code for a rake receiver - my UMTS turbo code assumes that all of the channel estimation and equalisation has already been done…

    Take care, Rob.

  9. kahsay kiross Says:

    hello thank u for the notes and scripts . but i am writting my thesis in channel estimation techniques for lte downlink . i am trying to implement it in matlab for different types of fading channels. please if u have a sample matlab code can u help me.

  10. Rob Says:

    Hello Kahsay,

    I’m afraid that I don’t have any Matlab code for channel estimation.

    Take care, Rob.

  11. faisal Says:

    Hi ROB”
    could you please help me in implementation of Differentially encoded LDPC (in context of non coherent detection :no channel estimation ) specifically product accumulatede LDPC.
    http://jwcn.eurasipjournals.com/search/results?terms=jing%20tiffany
    as jing tiffanys’ paper on eurasip journal says that regular LDPC codes doesnt work good for non coherent detection .you either have to use DE-LDPC ,or we have to optimize these codes using exit charts which i realy dont understand.I have just completed my simulation using DQPSK with regular LDPC non coherent detection in rayleigh fading.BER curve of which is nor accurate at high SNRs. I want to change LDPC part of this code to DE-LDPC or equivalent so to get accurate and minimum possible BER.Could you Help me in that I am talking in reference to your expertise in exit charts specially.thanks in advance i am waiting for your answer hope you help me soon.I have followed all the comments of yours in exit charts discussions as well as your code ‘main middle’.

  12. Rob Says:

    Hi Faisal,

    Do you have a specific question that you are struggling to answer? You can read about Irregular LDPC design in this paper by Stephan ten Brink…
    http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1291808&tag=1

    Take care, Rob.

  13. faisal Says:

    Thank you sir for considering my question:
    Actually I am Going to change My regular LDPC code that i have integrated with Differential detection That is not giving me the optimal results.I found out that i need to modify that LDPC Encoder decoder Either based on Exit chart critarian i dont know about that. Or I need another type of LDPC algorith named as Differentially encoded LDPC ,Suggested by jing Li Tiffany.Which I dont have either.Please help me to find out this DE-LDPC .Thank you for your time

  14. Rob Says:

    Hi Faisal,

    It seems to me that a DE-LDPC is obtained by concatenating an LDPC code with an accumulator. It sounds like this is what you have already done, since differential detection uses an accumulator. My suggestion would be to try to figure out why your existing code is not giving you the performance you expect. My advice would be to look for programming errors by using the averaging and histogram methods to draw the EXIT functions of your various system component - these two methods should give very similar results. If not, it suggests that you have a programming error in the corresponding system component. You can read all about this technique in the comments I have put at the following page…
    http://users.ecs.soton.ac.uk/rm/resources/matlabexit/

    Take care, Rob.

  15. faisal Says:

    Thank you for your kind advise .I am working on your suggestions ,thank you again for you dedication and quick reply

  16. Carmen Says:

    Hello!
    Do you a code for turbo equalization?
    Thank you!

  17. Rob Says:

    Hello Carmen,

    I’m afraid that I don’t have any Matlab code for turbo equalisation.

    Take care, Rob.

  18. amita Says:

    Hi Rob,

    do you have the matlab code for 16 QAM frequency offset recovery. Thanks :)

  19. shahzad Says:

    Hi Sir,

    I want to take my research on WSNs especially on the energy efficient techniques on PHY, MAC or Network Layer. Can you please suggest some specific area/areas which has great potential for research. I shall be very grateful.

    Thanks
    Shahzad

  20. Rob Says:

    Hi Shahzad,

    I think that cross-layer optimisation of PHY, MAC and Network layers for energy efficient communication is an area with a lot of untapped potential. My advice would be to look for opportunities to blur the lines between these layers, so that the energy can be optimised. For example, depending on the SNR detected by the PHY layer, the MAC or Network layers can modify their operation.

    Take care, Rob.

  21. Rob Says:

    Hi Amita,

    I’m afraid that I don’t have any Matlab code for a 16QAM frequency offset recovery.

    Take care, Rob.

  22. Mary Says:

    Hello
    do you have matlab code for puntured ldpc codes with awgn channel?
    and can you please tell me if puncturing is going to increase the error rate always?

  23. Rob Says:

    Hello Mary,

    I’m afraid that I don’t have any Matlab code for punctured LDPC codes. However, puncturing is easily to implement. You just need to discard some bits before transmission, then replace them with zero-valued LLRs upon reception. Puncturing will always increase the error rate.

    Take care, Rob.

Leave a Reply

Security Code: