A. Ambroze

University of Plymouth, Plymouth, England, United Kingdom

Are you A. Ambroze?

Claim your profile

Publications (10)4.69 Total impact

  • M. Ferrari · S. Bellini · A. Ambroze · M. Tomlinson ·
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper we propose a coding-decoding scheme to be used for II-type hybrid ARQ on a binary erasure channel model, based on rate compatible punctured binary Turbo Codes, DRP interleaver design and ML decoding at the receiver. We carefully design the interleaver and the puncturing mask, and we describe how to achieve low-complexity Maximum Likelihood decoding. We show for the case of an MPEG2 packet size that this scheme overrides random codes upper bounds, providing throughputs within 0.05 bits from capacity at word error rates greater or equal to 10<sup>-6</sup>.
    Information Theory Workshop, 2008. ITW '08. IEEE; 06/2008
  • Source
    A. Ambroze · M. Tomlinson · G. Wade ·
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper investigates the design and application of data magnitude modulation to reduce power amplifier back-off for QPSK and OQPSK modulations for small satellite Earth stations. The coefficients for data magnitude modulation are obtained by using an iterative design loop. The parameters of this loop are tuned to obtain maximum gain. Roll-off factors from 10% to 100% are considered, with emphasis on small roll-off values for increased bandwidth efficiency. It is shown that a gain of almost 6 dB can ideally be obtained for a roll-off factor of 10%. Error correction coding is used to compensate for the increased sensitivity to noise of the magnitude modulated data. The loss due to this increased sensitivity is reduced from almost 5.5 dB to 1.2 dB for a roll-off factor of 10% by using high rate product codes, thus giving an overall gain of 4.8 dB. Trade offs between spectral regrowth and implementation are addressed.
    Communications, 2003. ICC '03. IEEE International Conference on; 06/2003
  • M. Tomlinson · M. Ferrari · A. Ambroze ·
    [Show abstract] [Hide abstract]
    ABSTRACT: A simplified search for good recursive systematic convolution codes with rate b/(b + 1) is proposed, to be used as component codes in very high rate, parallel concatenated turbo codes. The first optimised parameter is the effective free-distance of turbo codes. The results obtained are compared to known bounds and previously published results where available.
    Electronics Letters 01/2003; 38(25-38):1691 - 1693. DOI:10.1049/el:20021113 · 0.93 Impact Factor
  • M. Tomlinson · G. Wade · P. Van Eetvelt · A. Ambroze ·
    [Show abstract] [Hide abstract]
    ABSTRACT: A tight bound for finite block-length codes is required when developing bandwidth efficient systems, and the paper describes an information-theoretic approach for determining the bound for binary codes. It is derived on the assumption that when the entropy loss in a binary symmetric channel equates to that in a soft-decision channel, the corresponding hard and soft-decision decoders must generate the same error probability. The bound agrees well with the Gallager (1968) bound at large block lengths, and is compared to the performance of iteratively decoded product codes. The paper also describes a recursive breakdown of Shannon's sphere packing bound that permits exact numerical evaluation for information block lengths up to 10<sup>3</sup> bits. The difference between the exact solution and Shannon's approximation is given up to this block length
    IEE Proceedings - Communications 05/2002; 149(2-149):65 - 69. DOI:10.1049/ip-com:20020315 · 0.79 Impact Factor
  • Source
    A. Tomlinson · A. Ambroze · G. Wade ·
    [Show abstract] [Hide abstract]
    ABSTRACT: The objective of this project is to maximise the power and bandwidth efficiency of small satellite Earth communication terminals whose main limitation to satellite capacity is a nonlinear transmitting power amplifier. A form of envelope equalisation is presented in which regular QPSK modulation is used together with instantaneous data magnitude modulation so as to produce a near constant envelope signal when filtered with root 35% raised cosine filtering. The algorithm for deriving the magnitude values is presented along with results using a data lookup table approximation. Bit error rate figures are presented for envelope equalisation combined with a (4096,3249) binary product code and MAP decoding. It is shown that at 10<sup>-6</sup> BER the E<sub>b</sub>/N<sub>0</sub> is 3.52 dB, some 0.26 dB from the linear system. When used with a nonlinear power amplifier, the benefit of envelope equalisation is approximately 3 dB. The power spectrum is identical to that of the linear system with no spectral spreading and no increase in adjacent channel interference.
    Communications, 2002. ICC 2002. IEEE International Conference on; 02/2002
  • Source
    A. Ambroze · G. Wade · C. Serdean · M. Tomlinson · J. Stander · M. Borda ·
    [Show abstract] [Hide abstract]
    ABSTRACT: The operational capacity of a watermark data channel is deduced by examining the correlation distribution at the retrieval end of a spread-spectrum-based watermarking system. This enables capacity to be determined given MPEG-2 compression, geometric attack, visual thresholds and channel coding. Watermarking itself is carried out in the DCT domain using video-dependent and visual perception concepts. An objective of the paper is to determine the capacity improvement provided by advanced FEC. It is found that FEC based on multiple parallel concatenated convolutional codes (3PCCCs) can give an order improvement in capacity for compressed video, and typically gives 0.5 kbit/s capacity under a combined compression-geometric attack
    IEE Proceedings - Vision Image and Signal Processing 03/2001; 148(1-148):54 - 58. DOI:10.1049/ip-vis:20010109 · 0.46 Impact Factor
  • A. Ambroze · G. Wade · M. Tomlinson ·
    [Show abstract] [Hide abstract]
    ABSTRACT: The convergence problem of iterative, block-mode turbo decoders is discussed and the performance of a practical convergence criterion is presented. A fixed-point approach is used, whereby the saturation and stability characteristics of the extrinsic-probability vector for each MAP decoder are determined by simulation and used to terminate iteration. If these vectors are saturated and identical, or non-saturated and stable, the decoder has converged to a fixed point. The paper also examines the effect of interleaver design and machine precision effects on convergence. Sometimes, finite precision can lead to a limit-cycle effect, and practical solutions are discussed. Once convergence has been established, it can also be used to determine with high confidence the effective d<sub>free</sub> of the decoder, even for large block lengths
    IEE Proceedings - Communications 05/2000; 147(2-147):69 - 74. DOI:10.1049/ip-com:20000151 · 0.79 Impact Factor
  • A. Ambroze · G. Wade · M. Tomlinson ·
    [Show abstract] [Hide abstract]
    ABSTRACT: The paper provides detailed computational steps for implementing an iterative serial concatenated convolutional code (SCCC) decoder. These are based on maximum a posteriori probability (MAP) decoding of a single, rate 1/2, recursive, systematic convolutional code, which is reduced to easily implemented equations for forward and backward recursion. In particular, the crucial information exchange between MAP decoders is clarified. Simulation of a rate 1/3 SCCC with memory-2 codes and a coding delay of N=1000 shows a bit error rate of 10<sup>-6</sup> for E<sub>b</sub>/N<sub>0</sub>=1.5 dB, and gives a typical interleaver gain of N<sup>-3</sup>
    IEE Proceedings - Communications 05/1998; DOI:10.1049/ip-com:19981876 · 0.79 Impact Factor
  • A. Ambroze · G. Wade · M. Tomlinson ·
    [Show abstract] [Hide abstract]
    ABSTRACT: A code tree for a rate 1/3, memory-2 turbo code is developed and shown to have a nonuniform branch structure due to the effect of the interleaver. The tree is used to compute the weight spectrum, and the number of terms required for an accurate upper bound to the bit error rate are identified
    Electronics Letters 03/1998; 34(4-34):353 - 354. DOI:10.1049/el:19980251 · 0.93 Impact Factor
  • Source
    Cristian V Serdean · Martin Tomlinson · Graham J Wade · Adrian M Ambroze ·
    [Show abstract] [Hide abstract]
    ABSTRACT: The objective of the paper is to analyse the advantages and disadvantages of spatial, DFT, DCT and DWT domains and highlight the advantages offered by watermarking wavelet coefficients rather than the DCT or FFT coefficients. The reasons for the DWT advantage are analysed and the choice of a particular wavelet basis is explained. As an illustration of these advantages, the paper presents a high capacity blind video watermarking system, which embeds the data payload in the wavelet domain. In this paper the video sequence is regarded as a noisy communications channel, and the multi-bit watermark as the hidden message. In order to maximize the information capacity in the presence of attacks, the payload is embedded according to a HVS model, and is protected by state-of-the-art error correction (Turbo codes). It is shown that the DWT is significantly more robust to scaling and cropping, and gives a useful capacity improvement under a compression attack.

Publication Stats

81 Citations
4.69 Total Impact Points


  • 1998-2008
    • University of Plymouth
      • Centre for Research in Information Storage Technology
      Plymouth, England, United Kingdom
  • 2002
    • University of Newcastle
      • Department of Computer Engineering
      Newcastle, New South Wales, Australia