Ahmad Abou Saleh

Low-Latency Source-Channel Coding for Fading Channels with Correlated Interference

We investigate the problem of sending a Gaussian source over a Rayleigh fading channel with Gaussian correlated interference known to the transmitter using low-latency codes. For the matched bandwidth case between the source and the channel, we show that among all single-letter codes, the uncoded scheme achieves the lowest mean square error distortion under full correlation between source and interference, and hence it is optimal. To benefit from nonlinear strategies for other scenarios, we derive the necessary conditions for optimality and propose an iterative algorithm based on joint optimization between the encoder and the decoder. A reduced-complexity approach for the implementation of the design algorithm is presented based on Monte-Carlo (at the encoder side) and importance sampling (at the decoder side) techniques. Furthermore, the scalability of our low-latency scheme is improved by modifying the search process at the encoder side using a targeted search method.

Hybrid digital-analog coding for interference broadcast channels

We consider the transmission of bivariate Gaussian sources (V1, V2) over the two-user Gaussian broadcast channel in the presence of interference that is correlated to the source and known to the transmitter. Each user i is interested in estimating Vi. We study hybrid digital-analog (HDA) schemes and analyze the achievable (square-error) distortion region under matched and expansion bandwidth regimes. These schemes require proper combinations of power splitting, bandwidth splitting, rate splitting, Wyner-Ziv and HDA Costa coding. An outer bound on the distortion region is also derived by assuming knowledge of V1 at the second user and full/partial knowledge of the interference at both users. Numerical results show that the HDA schemes outperform tandem and linear schemes and perform close to the derived bound for certain system settings.

Power-constrained bandwidth-reduction source-channel mappings for fading channels

We consider the transmission of a memoryless Gaussian source over a power-constrained Rayleigh fading channel with additive white Gaussian noise. We propose the use of low-delay joint source-channel mappings and consider optimizing the non-parametric mappings through an iterative process. A design algorithm for joint source-channel mapping is proposed and numerically evaluated for 2:1, 3:1, and 4:1 bandwidth reductions. Parametric mappings are also studied. We assume three cases of fading knowledge; in the case of presence of channel state information at both encoder and decoder, optimal power allocation is solved for the parametric mappings in terms of fading gains and average power constraint. It is shown that the proposed non-parametric and parametric mappings, which have a non-linear structure, achieve a graceful and robust performance and considerably surmount the saturation effect of linear systems.

Compressed Sensing with Non-Gaussian Noise and Partial Support Information

We study the problem of recovering sparse and compressible signals using a weighted ℓp minimization with 0<;p≤1 from noisy compressed sensing measurements when part of the support is known a priori. To better model different types of nonGaussian (bounded) noise, the minimization program is subject to a data-fidelity constraint expressed as the ℓq(2≤q<;∞) norm of the residual error. We show theoretically that the reconstruction error of this optimization is bounded (stable) if the sensing matrix satisfies an extended restricted isometry property. Numerical results show that the proposed method, which extends the range of and comparing with previous works, outperforms other noise-aware basis pursuit programs. For p<;1, since the optimization is not convex, we use a variant of an iterative reweighted ℓ2 algorithm for computing a local minimum.

Source-Interference Recovery Over Broadcast Channels: Asymptotic Bounds and Analog Codes

We consider the problem of joint recovery of a bivariate Gaussian source and of interference over the two-user Gaussian degraded broadcast channel in the presence of common interference. The interference, that is available non-causally at the encoder, is assumed to be Gaussian and correlated to the sources. The tradeoff between the distortion of the sources and the interference estimation error is studied; information-theoretic outer and inner bounds based on ideas from rate-distortion theory and hybrid coding are derived, respectively. More precisely, the outer bound is found by assuming additional knowledge at each user; the inner bound, however, is obtained by analyzing the distortion of a layered hybrid scheme based on proper power splitting, Costa and Wyner-Ziv coding. Low delay and complexity coding schemes based on analog mapping are next proposed. More specifically, parametric mappings based on linear and sawtooth curves are studied and optimized by minimizing an upper bound on the systems distortion; nonparametric mappings based on joint optimization between the encoder and the decoder using an iterative algorithm are designed. Numerical results show that for the special cases that are previously considered by Abou Saleh et al. (with no fading), the derived outer bound is tighter and the proposed hybrid scheme has a lower complex structure with no loss in performance. In addition, the proposed low delay nonlinear schemes outperform the linear scheme and perform relatively close to the inner bound under certain system settings.

Source-Channel Coding for Fading Channels With Correlated Interference

We consider the problem of sending a Gaussian source over a fading channel with Gaussian interference known to the transmitter. We study joint source-channel coding schemes for the case of unequal bandwidth between the source and the channel and when the source and the interference are correlated. An outer bound on the systems distortion is first derived by assuming additional information at the decoder side. We then propose layered coding schemes based on proper combination of power splitting, bandwidth splitting, Wyner-Ziv and hybrid coding. More precisely, a hybrid layer, that uses the source and the interference, is concatenated (superimposed) with a purely digital layer to achieve bandwidth expansion (reduction). The achievable (square error) distortion region of these schemes under matched and mismatched noise levels is then analyzed. Numerical results show that the proposed schemes perform close to the best derived bound and to be resilient to channel noise mismatch. As an application of the proposed schemes, we derive both inner and outer bounds on the source-channel-state distortion region for the fading channel with correlated interference; the receiver, in this case, aims to jointly estimate both the source signal as well as the channel-state (interference).

Hybrid digital-analog source-channel coding with one-to-three bandwidth expansion

We consider the problem of bandwidth expansion for lossy joint source-channel coding over a memoryless Gaussian channel. A low delay 1∶3 bandwidth expansion hybrid digital-analog coding system, which combines a scalar quantizer and a 1∶2 nonlinear analog coder, is proposed. It is shown that our proposed system outperforms the 1∶3 generalized hybrid scalar quantizer linear coder in terms of signal-to-distortion ratio (SDR). A lower bound on the system SDR is also derived.

Analog coding for Gaussian source and state interference estimation

We consider zero-delay analog coding of a Gaussian source over a Gaussian channel with additive correlated Gaussian interference known to the transmitter. The receiver aims to jointly estimate the source signal and the state interference. We propose a layered parametric analog coding scheme based on linear and sawtooth mappings. We derive an upper bound on the distortion for the parametric scheme by assuming a suboptimal decoder. To optimize the system parameters, we use two suboptimal methods. The first one is partially numerical and part of the parameters are derived assuming no sawtooth mapping; the other one, however, is based on minimizing the derived upper bound. To improve the performance whenever storage and offline design complexity are not an issue, we design a nonparametric mapping through an iterative process based on joint optimization between the encoder and the decoder using the necessary conditions for optimality. Numerical results show that the nonparametric and parametric mappings outperform the linear scheme and overcome the saturation effect.

Low and high-delay source-channel coding with bandwidth expansion and correlated interference

We consider the problem of sending a Gaussian source over an additive white Gaussian noise channel with Gaussian correlated interference known to the transmitter. We study both low-delay and asymptotically high-delay (in the sense of infinite source and coding block lengths) joint source-channel coding schemes based on purely analog and hybrid-digital analog (HDA) schemes with bandwidth expansion, respectively. The achievable (square error) distortion region of these schemes under matched and mismatched noise power is analyzed. The low-delay scheme uses a non-parametric analog mapping that is designed using a joint optimization of the encoder and the decoder. Numerical results show that the non-parametric approach adapts better to the interference than the classical linear scheme. For the high-delay regime, we establish a lower bound on the systems distortion and propose a layered HDA scheme based on Wyner-Ziv and HDA Costa coding. The proposed HDA scheme is shown to perform close to the derived bound and to be resilient under noise mismatch.

Power-constrained low-complexity coding of compressed sensing measurements

We study a low delay and low complexity sensor-communication system based on compressed sensing (CS) and scalar coding for transmission. The proposed scheme uses a 1 : r channel dimension expansion on the CS measurements for protection against channel noise. Simulation results show that optimizing the choice of r and the power allocation between the r transmissions significantly improve the system performance when compared to existing CS-communication schemes. Moreover, we consider the asymptotic behaviour of our CS system as the channel signal-to-noise ratio grows without bound and show that the proposed scheme achieves the optimal scaling exponent.