Javad Haghighat

Decode-Compress-and-Forward with Selective-Cooperation for Relay Networks

We propose a new signal-processing scheme, referred to as Decode-Compress-and-Forward with Selective-Cooperation (DCF-SC). In DCF-SC, the relay dedicates a certain amount of time to listen to the message broadcasted by the source and then performs Soft-Input Soft-Output (SISO) decoding. The relay then quantizes the Log-Likelihood Ratio (LLR) values received from the SISO decoder, encodes them and then transmits to the destination. The Selective-Cooperation condition determines whether the destination will accept or reject relays collaboration. We consider half-duplex relaying with orthogonal channels at the destination and apply turbo coding at both source and relay nodes. We define a trade-off parameter that determines how much of the relays time should be dedicated to listening and to transmission. We show by simulations that this trade-off factor has an optimal value for which the Block-Error Rate (BLER) is minimized. We compare the error rate performance of the proposed DCF-SC scheme with that of the Decode-Amplify-Forward (DAF) scheme presented in the literature.

Design of lossless turbo source encoders

Lossless turbo source coding with decremental redundancy is an effective approach for compressing binary sources. A large block length lossless turbo source encoder offers compression rates close to the source entropy but with large latency. In this letter, we propose a lossless compression technique for binary memoryless sources using short block length turbo codes. To achieve compression rates close to the source entropy, we modify different components of the encoder. We focus on the design of the parity interleaver for different compression rates. Also, we replace the square shape puncturing array with a rectangular shape array that allows finer puncturing and hence improved compression rates. Finally, instead of a single code, we employ many codes operating in parallel. Given these modifications, we evaluate the encoding complexity of the proposed code

A Power-Efficient Scheme for Wireless Sensor Networks Based on Transmission of Good Bits and Threshold Optimization

We propose a power-efficient transmission scheme for wireless sensor networks. In this scheme, the sensor nodes compress their corresponding source-relay channel state information (CSI) and transmit this compressed CSI sequence along with a selected subset of their received bits from the source, which we refer to as the good bits. We assume slowly varying fading channels between source and relays and analytically derive the compression rate of the CSI sequence. We then study the relay-fusion centre link and find an optimal threshold for our proposed scheme, based on the target bit error rate and the packet delivery ratio of the network. Combining the threshold optimization and the reliable bit transmission schemes, we study the total number of transmitted bits for our proposed system. We show that for slowly varying fading channels, our proposed scheme considerably reduces the number of transmitted bits, and consequently the transmission power of the relay nodes compared with a conventional scheme where all bits are transmitted to the fusion center. We also compare the performance of our proposed scheme with a special decode and forward (SDF) scheme previously introduced in the literature. In order to have a fair comparison, we modify the SDF scheme, such that the modified scheme includes the originally proposed SDF scheme as a special case. We provide detailed comparisons and discussions on the achieved bit error rate, energy efficiency, and feasibility of the proposed and the modified SDF schemes.

Energy efficient relay selection scheme for cooperative uniformly distributed wireless sensor networks

We consider a wireless sensor network (WSN) with identically distributed nodes, and a two phase cooperative protocol where the source transmits and is overheard by multiple relays which in turn transmit to the destination or fusion center (FC). We introduce a selection scheme that will pick a subset of the relays that overhear the message and transmit to the FC. This scheme will aim at making the least number of relays active while minimizing the outage probability and sending the least amount of information enough to reconstruct the message at the FC. The reduced amount of information being transmitted through the network along with an even distribution of active relays leads to a more energy efficient system.

Code detection in turbo source coding

We consider the lossless compression of binary memoryless sources using a library of turbo codes. The message is compressed by each code and the best result along with the index of the applied code is sent to the decoder. Instead of transmitting the code index, we find a criterion to detect the code index using the transmitted parities. Our method helps to reduce the compression rate of short block length turbo source coders.

Design of short block length lossless turbo source encoders

Lossless turbo source coding with decremental redundancy is an effective approach for compressing binary sources. In this method, two sets of parities are generated for each block of data. The parities are interleaved and written into two square shape parity puncturing arrays. Then the encoder gradually punctures the parities using an iterative process, checking the decoders ability to perfectly reconstruct the data. Large block length system has a good compression rate, but also a large latency. To reduce the latency, we modify different components of the encoder. We design a systematic parity interleaver that outperforms pseudo-random interleaver used in literature. Also we replace the square shape puncturing array with a rectangular shape array that allows finer puncturing. Finally, instead of a single code we use many codes operating in parallel to achieve a better compression rate. Our modified scheme can achieve the same compression rate as the one achieved by the original scheme, but with much smaller block length.

A Closed-Form Mutual Information Approximation for Multiple-Antenna Systems With Spatial Modulation

In this letter, a low-complexity approximation for instantaneous mutual information (IMI) of spatial modulation multiple-input single-output systems is presented. Based on this low-complexity approximation, a closed-form approximation for the average mutual information (MI) is derived. The approximation closely follows exact average MI values for different numbers of antennas and a wide range of signal-to-noise ratios. The proposed approximation for the IMI has a considerably lower computational complexity compared to previously proposed bounds. Also, the closed-form approximation of the average MI is considerably less complex than previously proposed bounds that require numerical averaging over randomly generated channel realizations.

Analysis of a Subset Selection Scheme for Wireless Sensor Networks in Time-Varying Fading Channels

One of the main challenges facing wireless sensor networks (WSNs) is the limited power resources available at small sensor nodes. It is therefore desired to reduce the power consumption of sensors while keeping the distortion between the source and its estimate at the fusion centre (FC) below a specific threshold. In this paper, we analyze a subset selection strategy to reduce the average transmission power of the WSN. We consider a two-hop network and assume the channels between the source and the relay sensors to be time-varying fading channels, modeled as Gilbert-Elliott channels. We show that when these channels are known at the FC, a subset of sensors can be selected by the FC to minimize transmit power while satisfying the distortion criterion. Through analysis, we derive the probability distribution of the size of this subset. We also consider practical aspects of implementing the proposed scheme, including channel estimation at relays. Through simulations, we compare the performance of the proposed scheme with schemes appearing in the literature. Simulation results confirm that for a certain range of end-to-end bit-error rates (BERs), the proposed scheme succeeds to achieve a superior power reduction compared to other schemes.

Relay Pre-Selection for Reducing CSI Transmission in Wireless Sensor Networks

Relay selection requires collection and transmission of channel state information (CSI) to a central node. When all relay nodes contribute in CSI transmission, the amount of power required for signaling the CSIs linearly increases with the number of relays. A central node should determine the proper number of pre-selected relays by aid of channel statistics and a decision criterion. We derive an upper bound on this number when end-to-end bit error rate is considered as the decision criterion and the source-relay channels are time-varying fading channels.

Lossless source coding using repeat-accumulate codes

Design of good repeat-accumulate (RA) codes for data compression has been possible by using extrinsic information transfer (EXIT) charts. EXIT charts could also predict the average compression rate achieved by the RA code. However, for RA source codes the compression rate is a random variable and to the best of our knowledge simulation is the only known way to find the probability mass function (pmf) of the compression rate. In this paper, we present a tool to evaluate this pmf when the inner code is a single memory recursive convolutional code. Considering regular RA codes we then study the effect of two parameters on the accuracy of the evaluated pmf: the block length and the rate of the outer code. Simulation results show that this evaluation is fairly accurate for short block length systems and high-rate outer codes.