Self-concatenated coding for wireless communication systems

Abstract

In this thesis, we have explored self-concatenated coding schemes that are designed for transmission over Additive White Gaussian Noise (AWGN) and uncorrelated Rayleigh fading channels. We designed both the symbol-based Self-Concatenated
Codes considered using Trellis Coded Modulation (SECTCM) and bit-based Self- Concatenated Convolutional Codes (SECCC) using a Recursive Systematic Convolutional (RSC) encoder as constituent codes, respectively. The design of these codes was carried out with the aid of Extrinsic Information Transfer (EXIT) charts. The EXIT chart based design has been found an efficient tool in finding the decoding convergence threshold of the constituent codes. Additionally, in order to recover the information loss imposed by employing binary rather than non-binary schemes, a soft decision demapper was introduced in order to exchange extrinsic information with
the SECCC decoder. To analyse this information exchange 3D-EXIT chart analysis was invoked for visualizing the extrinsic information exchange between the proposed Iteratively Decoding aided SECCC and soft-decision demapper (SECCC-ID). Some of the proposed SECTCM, SECCC and SECCC-ID schemes perform within about 1 dB from the AWGN and Rayleigh fading channels’ capacity. A union bound analysis of SECCC codes was carried out to find the corresponding Bit Error Ratio (BER) floors. The union bound of SECCCs was derived for communications over both AWGN and uncorrelated Rayleigh fading channels, based on a novel interleaver concept.
Application of SECCCs in both UltraWideBand (UWB) and state-of-the-art video-telephone schemes demonstrated its practical benefits.

In order to further exploit the benefits of the low complexity design offered by SECCCs we explored their application in a distributed coding scheme designed for cooperative communications, where iterative detection is employed by exchanging extrinsic information between the decoders of SECCC and RSC at the destination. In the first transmission period of cooperation, the relay receives the potentially erroneous data and attempts to recover the information. The recovered information is then re-encoded at the relay using an RSC encoder. In the second transmission period this information is then retransmitted to the destination. The resultant symbols transmitted from the source and relay nodes can be viewed as the coded symbols of a three-component parallel-concatenated encoder. At the destination a Distributed Binary Self-Concatenated Coding scheme using Iterative Decoding (DSECCC-ID) was employed, where the two decoders (SECCC and RSC) exchange their extrinsic information. It was shown that the DSECCC-ID is a low-complexity scheme, yet capable of approaching the Discrete-input Continuous-output Memoryless Channels’s (DCMC) capacity.

Finally, we considered coding schemes designed for two nodes communicating with each other with the aid of a relay node, where the relay receives information from the two nodes in the first transmission period. At the relay node we combine a powerful Superposition Coding (SPC) scheme with SECCC. It is assumed that decoding errors may be encountered at the relay node. The relay node then broadcasts this information in the second transmission period after re-encoding it, again, using a SECCC encoder. At the destination, the amalgamated block of Successive Interference Cancellation (SIC) scheme combined with SECCC then detects and decodes the signal either with or without the aid of a priori information. Our simulation results demonstrate that the proposed scheme is capable of reliably operating at a low BER for transmission over both AWGN and uncorrelated Rayleigh fading channels. We compare the proposed scheme’s performance to a direct transmission link between the two sources having the same throughput.

More information

Identifiers

ORCID for Lajos Hanzo: orcid.org/0000-0002-2636-5214

ORCID for Soon Xin Ng: orcid.org/0000-0002-0930-7194

Catalogue record

Export record