The Experts below are selected from a list of 48 Experts worldwide ranked by ideXlab platform

Jin Jin - One of the best experts on this subject based on the ideXlab platform.

  • scattered random network coding for efficient transmission in multihop wireless networks
    IEEE Transactions on Vehicular Technology, 2011
    Co-Authors: R Y Kim, Jin Jin
    Abstract:

    Packet error rate dramatically increases when transmissions go over multiple hops in wireless networks, leading to substantial throughput performance degradation. However, in real wireless channels, bit error probabilities vary across different bit positions in one Modulation Symbol, and corruption of the packet is largely due to the incurred errors on those “bad” bit positions. To improve the throughput performance in multihop wireless networks, in this paper, we propose a novel scattered random network coding (which is referred to as S-RNC) scheme, which further exploits the usefulness of random network coding and takes advantage of error position diversity. In S-RNC, the random-network-coded blocks are classified into different groups, and certain groups of blocks are selected as protected blocks. The sender and relays always scatter the bits of these protected coded blocks on “good” bit positions (with low error probability) and the rest on “bad” bit positions (with high error probability). Rather than sharing the same error rate across all blocks in the conventional transmission scheme, the error probabilities of protected blocks in S-RNC significantly decrease, even over multiple hops, which is helpful in achieving overall higher throughput, particularly under poor channel conditions. Corroborating our intuition, our extensive simulation results show that S-RNC substantially improves throughput performance in multihop mode of wireless networks.

Renato Baldini Filho - One of the best experts on this subject based on the ideXlab platform.

  • clipping distortion performance of nonsquare m qam ofdm systems on nonlinear time variant channels
    IEEE Transactions on Vehicular Technology, 2011
    Co-Authors: Luciano Leonel Mendes, Renato Baldini Filho
    Abstract:

    Orthogonal frequency-division multiplexing (OFDM) systems usually make use of a set of square M quadrature-amplitude Modulation ( M-QAM) constellations to obtain a good tradeoff between throughput and Symbol-error robustness. However, the switch to the next constellation increases the number of bits per Modulation Symbol by two. The introduction of nonsquare M -QAM constellations in such systems brings extra advantages such as smoother transition among bit rates and a reduction of the peak-to-average ratio of the OFDM signal. Therefore, this paper unfolds analytical expressions to evaluate the Symbol-error performance of nonsquare M-QAM OFDM on nonlinear time-variant additive white Gaussian noise channels, taking clipping distortion into account. Cross and overlaid M-QAM are considered. Analytical performances are evaluated and compared with computational simulations, which show good agreement.

Marat Burnashev - One of the best experts on this subject based on the ideXlab platform.

Yahong Rosa Zheng - One of the best experts on this subject based on the ideXlab platform.

  • Precoded MIMO Systems With Nonbinary LDPC Codes and Many-to-One Mapping
    IEEE Transactions on Vehicular Technology, 2018
    Co-Authors: Nhat Quang Nhan, Emanuel Radoi, Karine Amis, Philippe Rostaing, Yahong Rosa Zheng
    Abstract:

    A precoding design is proposed for multiple-input multiple-output (MIMO) systems utilizing nonbinary low-density parity check (NB-LDPC) codes and many-to-one mapping. When a high-order Modulation scheme is used, many-to-one mapping converts a group of low-order Galois field (GF) coded Symbols into one modulated MIMO Symbol vector. In contrast, one-to-one mapping maps a high-order Modulation Symbol directly from one high-order GF coded Symbol. With the help of an interleaver between the NB-LDPC encoder and the GF to MIMO Symbol mapper, the many-to-one mapping enables turbo receiver and reduces the computational complexity by more than 95%. The proposed MIMO precoders enhance the error-rate performance of the many-to-one NB-LDPC MIMO system, especially in terms of reducing the error floor and improving the waterfall region. The proposed precoder design modifies the approach that suboptimally maximizes the minimal Euclidean distance between the received MIMO Symbols. Simulation results show that the proposed precoders enhance the robustness of the precoder design and reduce the inner and outer iterations of the turbo receiver.

R Y Kim - One of the best experts on this subject based on the ideXlab platform.

  • scattered random network coding for efficient transmission in multihop wireless networks
    IEEE Transactions on Vehicular Technology, 2011
    Co-Authors: R Y Kim, Jin Jin
    Abstract:

    Packet error rate dramatically increases when transmissions go over multiple hops in wireless networks, leading to substantial throughput performance degradation. However, in real wireless channels, bit error probabilities vary across different bit positions in one Modulation Symbol, and corruption of the packet is largely due to the incurred errors on those “bad” bit positions. To improve the throughput performance in multihop wireless networks, in this paper, we propose a novel scattered random network coding (which is referred to as S-RNC) scheme, which further exploits the usefulness of random network coding and takes advantage of error position diversity. In S-RNC, the random-network-coded blocks are classified into different groups, and certain groups of blocks are selected as protected blocks. The sender and relays always scatter the bits of these protected coded blocks on “good” bit positions (with low error probability) and the rest on “bad” bit positions (with high error probability). Rather than sharing the same error rate across all blocks in the conventional transmission scheme, the error probabilities of protected blocks in S-RNC significantly decrease, even over multiple hops, which is helpful in achieving overall higher throughput, particularly under poor channel conditions. Corroborating our intuition, our extensive simulation results show that S-RNC substantially improves throughput performance in multihop mode of wireless networks.