Joint Source And Channel Coding Research Articles

Adaptive Information Bottleneck Guided Joint Source and Channel Coding for Image Transmission

Joint source and channel coding (JSCC) for image transmission has attracted increasing attention due to its robustness and high efficiency. However, the existing deep JSCC research mainly focuses on minimizing the distortion between the transmitted and received information under a fixed number of available channels. Therefore, the transmitted rate may be far more than its required minimum value. In this paper, an adaptive information bottleneck (IB) guided joint source and channel coding (AIB-JSCC) method is proposed for image transmission. The goal of AIB-JSCC is to reduce the transmission rate while improving the image reconstruction quality. In particular, a new IB objective for image transmission is proposed so as to minimize the distortion and the transmission rate. A mathematically tractable lower bound on the proposed objective is derived, and then, adopted as the loss function of AIB-JSCC. To trade off compression and reconstruction quality, an adaptive algorithm is proposed to adjust the hyperparameter of the proposed loss function dynamically according to the distortion during the training. Experimental results show that AIB-JSCC can significantly reduce the required amount of transmitted data and improve the reconstruction quality and downstream task accuracy.

Collaborative Semantic Communication for Edge Inference

We study the collaborative image retrieval problem at the wireless edge, where multiple edge devices capture images of the same object from different angles and locations, which are then used jointly to retrieve similar images at the edge server over a shared multiple access channel (MAC). We propose two novel deep learning-based joint source and channel coding (JSCC) schemes for the task over both additive white Gaussian noise (AWGN) and Rayleigh slow fading channels, with the aim of maximizing the retrieval accuracy under a total bandwidth constraint. The proposed schemes are evaluated on a wide range of channel signal-to-noise ratios (SNRs), and shown to outperform the single-device JSCC and the separation-based multiple-access benchmarks. We also propose a channel state information-aware JSCC scheme with attention modules to enable our method to adapt to varying channel conditions.

Double Polar Codes for Joint Source and Channel Coding

In this paper, we design a joint source and channel coding (JSCC) framework combining the source polar coding and the channel polar coding. The source is first compressed using a polar code (PC), and source check decoding is employed to construct an error set containing the index of all source decoding errors. Then, the proposed JSCC system employs another PC or systematic PC (SPC) to protect the compressed source and the error set against noise, which is called double PC (D-PC) or systematic double PC (SD-PC), respectively. For a D-PC JSCC system, we prove a necessary condition for the optimal mapping between the source PC and the channel PC. On the receiver side, by introducing the joint factor graph representation of the D-PC and SD-PC, we propose two joint source and channel decoders: a joint belief propagation (J-BP) decoder, and a systematic joint belief propagation (SJ-BP) decoder. In addition, a biased extrinsic information transfer (B-EXIT) chart is developed for various decoders as a theoretical performance evaluation tool. Both B-EXIT and simulation results show that the performance of the proposed JSCC scheme has no error floor and outperforms the turbo-like BP decoder.

Joint Source and Channel Coding Using Double Polar Codes

In this letter, a double polar code (D-Polar) for joint source and channel coding (JSCC) is proposed, in which the source compress is implemented by a polar code, the channel error correction is performed by a systematic polar code. Furthermore, a turbo-like belief propagation (TL-BP) decoder consisted of a channel BP decoder and a source BP decoder is proposed for joint source and channel decoding in the receiver. In this TL-BP decoder, the soft information is exchanged between the channel BP decoder and the source BP decoder so as to improve the efficiency of channel decoding in utilizing the source information residual in the compressed bits. Simulation results show that the performance of the proposed D-Polar JSCC scheme with TL-BP decoder is significantly improved compared with the existing source-channel joint polarization scheme.

Wireless Image Retrieval at the Edge

We study the image retrieval problem at the wireless edge, where an edge device captures an image, which is then used to retrieve similar images from an edge server. These can be images of the same person or a vehicle taken from other cameras at different times and locations. Our goal is to maximize the accuracy of the retrieval task under power and bandwidth constraints over the wireless link. Due to the stringent delay constraint of the underlying application, sending the whole image at a sufficient quality is not possible. We propose two alternative schemes based on digital and analog communications, respectively. In the digital approach, we first propose a deep neural network (DNN) aided retrieval-oriented image compression scheme, whose output bit sequence is transmitted over the channel using conventional channel codes. In the analog joint source and channel coding (JSCC) approach, the feature vectors are directly mapped into channel symbols. We evaluate both schemes on image based re-identification (re-ID) tasks under different channel conditions, including both static and fading channels. We show that the JSCC scheme significantly increases the end-to-end accuracy, speeds up the encoding process, and provides graceful degradation with channel conditions. The proposed architecture is evaluated through extensive simulations on different datasets and channel conditions, as well as through ablation studies.

Deep Joint Source-Channel Coding for Wireless Image Transmission

We propose a joint source and channel coding (JSCC) technique for wireless image transmission that does not rely on explicit codes for either compression or error correction; instead, it directly maps the image pixel values to the complex-valued channel input symbols. We parameterize the encoder and decoder functions by two convolutional neural networks (CNNs), which are trained jointly, and can be considered as an autoencoder with a non-trainable layer in the middle that represents the noisy communication channel. Our results show that the proposed deep JSCC scheme outperforms digital transmission concatenating JPEG or JPEG2000 compression with a capacity achieving channel code at low signal-to-noise ratio (SNR) and channel bandwidth values in the presence of additive white Gaussian noise (AWGN). More strikingly, deep JSCC does not suffer from the ``cliff effect'', and it provides a graceful performance degradation as the channel SNR varies with respect to the SNR value assumed during training. In the case of a slow Rayleigh fading channel, deep JSCC learns noise resilient coded representations and significantly outperforms separation-based digital communication at all SNR and channel bandwidth values.

An Efficient Construction Strategy for Near-Optimal Variable-Length Error-Correcting Codes

In this letter, we present an efficient cross-entropy (CE)-based algorithm for the design of variable-length error-correcting (VLEC) codes under the joint source and channel coding (JSCC) framework. The algorithm enables us to construct the near-optimal VLEC codes that have the minimum average codeword length (ACL) with low search complexity. The efficiency of the proposed CE-based algorithm makes it possible to construct the VLEC codes that have small ACL values under various free-distance constraints, especially for large-sized signal alphabets.

Modeling the lossy transmission of correlated sources in multiple access fading channels

In this paper, we develop accurate distortion models for the lossy transmission of two correlated sources in a multiple access Rayleigh fading channel. We focus on a class of real-life communication systems, where the source and channel coders have already been designed separately and can only be configured during the system operation. We investigate three different source coding schemes: distributed source coding (DSC), layered source coding, and independent compression through quantization. With the later scheme the sources are jointly decoded with minimum mean square error (MMSE) estimation at the receiver. We also consider two different transmission schemes: Orthogonal transmissions and interfering transmissions decoded with a successive interference cancellation (SIC) decoder. Our final closed-form analytical models are used to determine the optimal combination of source coding and transmission schemes, as well as their optimal configuration. Hence, we exercise joint source and channel coding (JSCC) by optimizing the system configuration. Through simulations, we first validate the analytical model and illustrate the performance of different schemes. Finally, we demonstrate the JSCC gains achieved by our system.

Flexible iterative receiver architecture for wireless sensor networks: a joint source and channel coding design example

Due to their computational complexity, iterative decoder components such as source and channel decoders are usually implemented using specialised dedicated hardware. This leads to the scenario where each different iterative decoder component of the receiver requires its own hardware. Due to their relatively high complexity, many capacity-approaching techniques proposed in the literature have not yet been invoked in wireless sensor network applications, despite their potential benefits of facilitating a reduced transmission power or extended communication range. Against this background, the authors propose an energy-efficient architecture comprised of multiple computation unit (CU), which is sufficiently flexible for accommodating different iterative decoder components using the same hardware. In this study, the flexible architecture is applied to Joint Source and Channel Coding (JSCC), comprising the Unary Error Correction (UEC) code, a turbo code and an iterative demodulator. The authors conceive a flexible technique for controlling the hardware, which supports a high hardware-exploitation ratio for the CU, reaching a utilisation of 88%, compared with 68% achieved in similar solutions reported in the open literature.

Logical Superposition Coded Video Multicast/Broadcast

Joint source and channel coding (JSCC) has been widely reported for its superb capability in mitigating the legacy channel/user diversity problem in the wireless multicast/broadcast scenario, where the successively refined source information is manipulatively mapped with superimposed channel symbols via superposition coding (SPC). However, the use of SPC involves additional hardware in the physical layer, which may not be in compliance with most standard equipments and protocols. This paper introduces a novel framework of logical SPC (L-SPC) modulation for multicast/broadcast of successively refined sources via the JSCC approach. The framework is characterized by completely software-based L-SPC modulation at the sender, where the modulated signal is generated by mapping successively refined information bits onto a single signal constellation. At the receiving end, the proposed framework only needs simple modifications in the medium access control layer software, while the required demodulation schemes are completely compatible with the standard ones with uniform constellation. Generalized formulations on symbol error rate (SER) are derived for performance evaluation and comparisons with the conventional hardware-based approach for both the 2- and 3-layer cases.

A High-Throughput FPGA Architecture for Joint Source and Channel Decoding

In the wireless transmission of multimedia information, the achievable transmission throughput and latency may be limited by the processing throughput and latency associated with source and channel coding. Ultra-high throughput and ultra-low latency processing of source and channel coding are required by the emerging new video transmission applications, such as the first-person remote control of unmanned vehicles. The recently proposed unary error correction (UEC) code facilitates the joint source and channel coding (JSCC) of video information at transmission throughputs that approach the capacity of the wireless channel. In this paper, we propose the first hardware implementation of the UEC code that achieves the high processing throughputs as well as ultra-low processing latencies required. This is achieved by extending the application of the recently proposed fully parallel turbo decoder (FPTD) from pure stand-alone channel coding to JSCC. This paper also proposes several novel improvements to the FPTD, in order to increase its hardware efficiency and supported frame length. We demonstrate the application of these improvements to both the long term evolution turbo code and the UEC code. We synthesize the proposed fully parallel design on a mid-range field programmable gate array, achieving a throughput of 450 Mbps, as well as a factor of 2.4 hardware efficiency improvement over previous implementations of the FPTD.

Exponential Golomb and Rice Error Correction Codes for Generalized Near-Capacity Joint Source and Channel Coding

The recently proposed unary error correction (UEC) and Elias gamma error correction (EGEC) codes facilitate the near-capacity joint source and channel coding (JSCC) of symbol values selected from large alphabets at a low complexity. Despite their large alphabet, these codes were only designed for a limited range of symbol value probability distributions. In this paper, we generalize the family of UEC and EGEC codes to the class of rice and exponential Golomb error correction codes, which have a much wider applicability, including the symbols produced by the H.265 video codec, the letters of the English alphabet, and in fact any arbitrary monotonic unbounded source distributions. Furthermore, the practicality of the proposed codes is enhanced to allow a continuous stream of symbol values to be encoded and decoded using only fixed-length system components. We explore the parameter space to offer beneficial tradeoffs between error correction capability, decoding complexity, as well as transmission-energy, -duration, and -bandwidth over a wide range of operating conditions. In each case, we show that our codes offer significant performance improvements over the best of several state-of-the-art benchmarkers. In particular, our codes achieve the same error correction capability, as well as transmission energy, transmission duration, and transmission bandwidth as a variable length error-correction code benchmarker, while reducing the decoding complexity by an order of magnitude. In comparison with the best of the other JSCC and separate source and channel coding benchmarkers, our codes consistently offer $E_{\mathrm {b}}/N_{0}$ gains of between 0.5 and 1.0 dB which only appear to be the modest, because the system operates close to capacity. These improvements are achieved for free, since they are not achieved at the cost of increasing transmission energy, transmission duration, transmission bandwidth, or decoding complexity.

Reordered Elias Gamma Error Correction Codes for the Near-Capacity Transmission of Multimedia Information

A novel joint source and channel coding (JSCC) scheme is proposed, which we refer to as the reordered Elias gamma error correction (REGEC) code. Like the recently proposed unary error correction (UEC) code and EGEC code, the proposed code facilitates the practical near-capacity transmission of source symbol values that are randomly selected from a large or infinite alphabet. However, in contrast to the UEC code, both the EGEC and our proposed REGEC codes are universal codes, facilitating the transmission of source symbol values that are randomly selected using any monotonic probability distribution. However, the EGEC code has a complicated structure comprising two parts, where unequal error protection is required to balance the two parts with the aid of a specific parameterization that must be tailored to the source distribution, preventing its employment for unknown or non-stationary sources. By contrast, the proposed REGEC code does not need unequal error protection, and hence its parameterization does not have to be tailored to the particular source distribution, and thus the REGEC code is a more attractive scheme. More explicitly, our REGEC code has a simple structure comprising only a single part, which does not suffer from the delay and loss of synchronization that are associated with the two parts of the EGEC code. In a particular practical scenario, where the source symbols obey a specific Zeta probability distribution, our REGEC scheme is shown to offer gains of up to 0.9 dB over the best of JSCC and separate source and channel coding (SSCC) benchmarkers, when QPSK modulation is employed for transmission over an uncorrelated narrowband Rayleigh fading channel. In the scenario where the source symbols obey the distribution produced by the H.265 video codec, our REGEC scheme is shown to offer a gain of 0.7 dB over the SSCC benchmarker. These gains are achieved for free, without increasing the required transmit-duration, transmit-bandwidth, transmit-energy, or decoding complexity.

Irregular Trellis for the Near-Capacity Unary Error Correction Coding of Symbol Values From an Infinite Set

Irregular joint source and channel coding (JSCC) scheme is proposed, which we refer to as the irregular unary error correction (IrUEC) code. This code operates on the basis of a single irregular trellis, instead of employing a set of separate regular trellises, as in previous irregular trellis-based codes. Our irregular trellis is designed with consideration of the UEC free distance, which we characterize for the first time in this paper. We conceive the serial concatenation of the proposed IrUEC code with an irregular unity rate code (IrURC) code and propose a new EXtrinsic Information Transfer (EXIT) chart matching algorithm for parametrizing these codes. This facilitates the creation of a narrow EXIT tunnel at a low $E_\text{b}/N_0$ value and provides near-capacity operation. Owing to this, our scheme is found to offer a low symbol error ratio (SER), which is within 0.4 dB of the discrete-input continuous-output memoryless channel (DCMC) capacity bound in a particular practical scenario, where gray-mapped quaternary phase shift keying (QPSK) modulation is employed for transmission over an uncorrelated narrowband Rayleigh-fading channel with an effective throughput of $0.508 \,\text{bit}\,{\text{s}}^{-1}\,{\text{Hz}}^{-1}$ . Furthermore, the proposed IrUEC–IrURC scheme offers a SER performance gain of 0.8 dB, compared to the best of several regular and irregular separate source and channel coding (SSCC) benchmarkers, which is achieved without any increase in transmission energy, bandwidth, transmit duration, or decoding complexity.

A Survey on Protograph LDPC Codes and Their Applications

Low-density parity-check (LDPC) codes have attracted much attention over the past two decades since they can asymptotically approach the Shannon capacity in a variety of data transmission and storage scenarios. As a type of promising structured LDPC codes, the protograph LDPC codes not only inherit the advantage of conventional LDPC codes, i.e., excellent error performance, but also possess simple representations to realize fast encoding and efficient decoding. This paper provides a comprehensive survey on the state-of-the-art in protograph LDPC code design and analysis for different channel conditions, including the additive white Gaussian noise (AWGN) channels, fading channels, partial response (PR) channels, and Poisson pulse-position modulation (PPM) channels. Moreover, the applications of protograph LDPC codes to joint source-and-channel coding (JSCC) and joint channel-and-physical-layer-network coding (JCPNC) are reviewed and studied. In particular, we focus our attention on the encoding design and assume the decoder is implemented by the belief propagation (BP) algorithm. Hopefully, this survey may facilitate research in this area.

Performance of Joint Source-Channel Coding Based on Protograph LDPC Codes over Rayleigh Fading Channels

This letter presents a joint protograph extrinsic information transfer (PEXIT) analysis for joint source and channel coding (JSCC) with double protograph-based low-density parity-check (DP_LDPC) codes over single-input multiple-output (SIMO) Rayleigh fading channels. The impacts of source statistics and receive antenna diversity on convergence performance for the JSCC are investigated, and we find that source sparsity plays a more vital role than diversity orders in improving both convergence and error performances under fading environments.

Joint Source/Channel Coding Based on Two-Dimensional Optimization for Scalable H.264/AVC Video

The scalable extension of the H.264/AVC video coding standard (SVC) demonstrates superb adaptability in video communications. Joint source and channel coding (JSCC) has been shown to be very effective for such scalable video consisting of parts of different significance. In this paper, a new JSCC scheme for SVC transmission over packet loss channels is proposed which performs two-dimensional optimization on the quality layers of each frame in a rate-distortion (R-D) sense as well as on the temporal hierarchical structure of frames under dependency constraints. To compute the end-to-end R-D points of a frame, a novel reduced trellis algorithm is developed with a significant reduction of complexity from the existing Viterbi-based algorithm. The R-D points of frames are sorted under the hierarchical dependency constraints and optimal JSCC solution is obtained in terms of the best R-D performance. Experimental results show that our scheme outperforms the existing scheme of [13] with average quality gains of 0.26 dB and 0.22 dB for progressive and non-progressive modes respectively.

Joint Source and Channel Coding

The objectives of this article are two-fold: First, to present the problem of joint source and channel (JSC) coding from a graphical model perspective and second, to propose a structure that uses a new graphical model for jointly encoding and decoding a redundant source. In the first part of the article, relevant contributions to JSC coding, ranging from the Slepian-Wolf problem to joint decoding of variable length codes with state-of-the-art source codes, are reviewed and summarized. In the second part, a double low-density parity-check (LDPC) code for JSC coding is proposed. The double LDPC code can be decoded as a single bipartite graph using standard belief propagation (BP) and its limiting performance is analyzed by using extrinsic information transfer (EXIT) chart approximations.

Joint Source Coding and Network-Supported Distributed Error Control for Video Streaming in Wireless Multihop Networks

Real-time video communication over wireless multihop networks has gained significant interest in the last few years. In this paper, we focus our attentions on the problem of source coding and link adaptation for packetized video streaming in wireless multihop networks when network nodes are media-aware. We consider a system where source coding is employed at the video encoder by selecting the encoding mode of each individual macro-block, while error control is exercised through application-layer retransmissions at each media-aware network node. For this system model, the contribution of each communication link on the end-to-end video distortion is considered separately in order to achieve globally optimal source coding and ARQ error control. To reach the globally optimal solution, we formulate the problem of joint source and distributed error control (JSDEC) and devise a low-complexity solution algorithm based on dynamic programming. Extensive experiments have been carried out on the basis of H.264/AVC codec to demonstrate the effectiveness of the proposed algorithm over the existing joint source and channel coding (JSCC) algorithm in terms of PSNR perceived at the decoder under time-varying multihop wireless links.

Joint Source and Channel Coding using Punctured Ring Convolutional Coded CPM

In this paper, a novel trellis source encoding scheme based on punctured ring convolutional codes is presented. Joint source and channel coding (JSCC) using trellis coded continuous phase modulation (CPM) with punctured convolutional codes over rings is investigated. The channels considered are the additive white gaussian noise (AWGN) channel and the Rayleigh fading channel. Optimal soft decoding for the proposed JSCC scheme is studied. The soft decoder is based on the a posteriori probability (APP) algorithm for trellis coded CPM with punctured ring convolutional codes. It is shown that these systems with soft decoding outperform the same systems with hard decoding especially when the systems operate at low to medium signal-to-noise ratio (SNR). Furthermore, adaptive JSCC approaches based on the proposed source coding scheme are investigated. Compared with JSCC schemes with fixed source coding rates, the proposed adaptive approaches can achieve much better performance in the high SNR region. The novelties of this work are the development of a trellis source encoding method based on punctured ring convolutional codes, the use of a soft decoder, the APP algorithm for the combined systems and the adaptive approaches to the JSCC problem.