A Novel Hard Decision Based Simultaneous Target Tracking and Classification Approach.

The high computational complexity of existing joint tracking and classification (JTC) algorithms hampers their application. After presenting a new description of the JTC problem--simultaneous tracking and classification (STC) instead of JTC, we derive two STC algorithms in both exact and approximate forms by applying Hayes' rule to the target state probability density function (pdf) and target class probability mass function (pmf) simultaneously under the assumption that the kinematic and attribute measurement processes are conditional independent. The mutual information exchange between tracker and classifier of the proposed STC algorithms is introduced by defining the simultaneous pdf-pmf of target state and class, the dependence of kinematic measurement on target class, the dependence of attribute measurement on target state and target model, class-dependent kinematic model sets, and class-dependent flight envelopes, etc. The proposed STC algorithms have four distinctive features. First, they have a modularized structure, i.e., they explicitly integrate a multiple-model filter and a Bayesian classifier. Second, the approximate versions, which follow easily from the proposed STC algorithms thanks to their modularized structure, have a closed form with a lower computational complexity and are more suitable for real-time applications. Third, the proposed exact STC algorithms are derived without the hidden approximation made in some existing multiple-model based JTC algorithms. Fourth, one of the proposed STC algorithms has the potential to further reduce the computational load since it has no redundant motion models. Simulation results suggest that the proposed STC algorithms provide a hopeful solution to a class of STC problems.

Network Coding (NC) is used to transmit combined packets rather than sending them separately, resulting in both data rate and bandwidth savings. In this paper, the behavior of hard and soft decision encoding/decoding is investigated using Partial Unit Memory Turbo Code (PUMTC) as the forward error correction method, over the physical layer when communication between users (nodes) is performed via an Amplitude and Forward (AF) Base Station (BS) over white Gaussian noise channel. In the received packets in the BS are combined and then forwarded to the destination. The simulation results show that soft decision encoding/decoding outperforms hard decision decoding, by around 0.4 dB which is regarded as a good improvement, taking into consideration that the system becomes less complex due to the removal of the hard decision block before decoding the received data in the BS and at the destination. Index Terms-Physical Layer, Forward Error Correction, Amplify and Forward.

Binary message-passing decoders for low-density parity-check (LDPC) codes are studied by using extrinsic information transfer (EXIT) charts. The channel delivers hard or soft decisions and the variable node decoder performs all computations in the L-value domain. A hard decision channel results in the well-know Gallager B algorithm, and increasing the output alphabet from hard decisions to two bits yields a gain of more than 1.0 dB in the required signal to noise ratio when using optimized codes. The code optimization requires adapting the mixing property of EXIT functions to the case of binary message-passing decoders. Finally, it is shown that errors on cycles consisting only of degree two and three variable nodes cannot be corrected and a necessary and sufficient condition for the existence of a cycle-free subgraph is derived.

The optimal strategy for erasing symbols in a concatenated coding system, in which the outer code is decoded once, is considered. In general, the strategy involves comparison of a posteriori symbol likelihoods to nonuniform thresholds. The performance of a greedy version of such a rule is evaluated for a number of specific short inner codes, over both memoryless and bursty channels, with hard and soft decisions. It is shown that the gains achievable via erasing can be substantial when hard decision inner decoders are used and when short inner codes are used, but are less so otherwise. It is shown that erasing gains can be very large when applied over channels with memory, with interleaving at the symbol level.

This paper addresses the problem of joint tracking and classification (JTC) of a single extended target with a complex shape. To describe this complex shape, the spatial extent state is first modeled by star-convex shape via a random hypersurface model (RHM), and then used as feature information for target classification. The target state is modeled by two vectors to alleviate the influence of the high-dimensional state space and the severely nonlinear observation model on target state estimation, while the Euclidean distance metric of the normalized Fourier descriptors is applied to obtain the analytical solution of the updated class probability. Consequently, the resulting method is called the “JTC-RHM method.” Besides, the proposed JTC-RHM is integrated into a Bernoulli filter framework to solve the JTC of a single extended target in the presence of detection uncertainty and clutter, resulting in a JTC-RHM-Ber filter. Specifically, the recursive expressions of this filter are derived. Simulations indicate that: (1) the proposed JTC-RHM method can classify the targets with complex shapes and similar sizes more correctly, compared with the JTC method based on the random matrix model; (2) the proposed method performs better in target state estimation than the star-convex RHM based extended target tracking method; (3) the proposed JTC-RHM-Ber filter has a promising performance in state detection and estimation, and can achieve target classification correctly.

본 논문에서는 개별 센싱 결과를 결합하기 위해 센싱의 성능은 연판정 방식보다 낮지만, 시스템의 부담이 적고 비교적 단순한 방식인 경판정 방식을 적용하였으며, 센싱의 성능을 높이기 위해 각 CR 기기의 검출 확률을 반영하여 가중치를 설정하고, 계산된 가중치가 개별 센싱 결과에 재반영되어 결합하는 최적화된 경판정 방식을 제안한다. 기존의 경판정 방식과 달리 가중치를 적용하여 센싱 성능을 높이도록 하였고, 모의 실험 및 분석을 통하여 성능을 나타내었다. 1차 사용자의 신호는 OFDM 기반의 시스템을 가정하였고, 1차 사용자와 CR 기기 사이의 무선 채널은 가우시안(Gaussian) 채널로 모델링하였다. In this paper, we use hard decision method for cooperative spectrum sensing. Sensing performance adopting hard decision is lower than soft decision but system load is low and the process is relatively simple when the combining scheme is hard decision compared to soft decision. In order to improve sensing performance, we propose optimal hard decision method applying weight that is based on a probability of individual sensing. Unlike conventional hard decision, we try to improve sensing performance applying weight and show the performance of the proposed method from the simulation results and performance analysis. The signal of licensed user is OFDM signal and the wireless channel between a licensed user and CR systems is modeled as Gaussian channel.

Prediction of exons locations in deoxyribonucleic acid (DNA) sequences is a significant issue for biologists. This paper proposes a new method to solve this problem. Unlike the published studies, in which the prediction of exons locations depends on hard decisions from a single classifier, the proposed prediction approach depends on fusion of soft decisions from two classifiers. In the proposed approach we utilize the sliding window discrete Fourier transform (DFT), which is normally used to detect exons 3-base periodicity feature, in a different manner. The novelty here depends on obtaining soft decisions, rather than hard decisions, from two classifiers using different numerical mapping schemes, and fuses them in a decision fusion center to obtain a final global decision about the prediction of exons locations. Simulation results based on real data performed on the HMR195 dataset showed that the proposed soft decisions fusion method achieves better prediction performance compared to the traditional hard decision single classifier method. Moreover the proposed method can easily be extended to more than two classifiers.

The aim of this research is to investigate the Dorsch algorithm performance in the decoding process of binary Hamming codes. The conventional decoding method is hard decision based whereby the message of the demodulator is quantized only to two levels. An important feature of the Dorsch algorithm is that it uses soft decision decoding. Soft decision occurs if the output of the demodulator is quantized to more than two levels. The algorithm uses soft decision information to rank the reliability of the received symbols. The high reliability symbols are treated as the soft decision of the information values whereas the low reliability symbols are treated as parity checks for the information values. The error performance of both the decoding methods were evaluated and concluded. It was observed that Dorsch Algorithm produces better performance compared to the conventional hard decision decoding.

Although iterative equalizers are well-known for mitigating inter-symbol interference (ISI), recent results have shown that the schemes perform poorly in the presence of severe ISI, particularly when coding rates are high. Other equalization schemes, such as the non-iterative canonical decision-feedback equalizer (CDFE), have been shown to be asymptotically optimal on any channel. However, the scheme exhibits very high latency. To trade off between performance and latency, we propose an equalizer structure that iterates with both hard- and soft-decisions. In prior works, hard- decision iterative equalizers were seen to perform poorly, chiefly due to error-propagation in the feedback loop. However, the scheme proposed in this paper outperforms the linear turbo equalizer in both strong and weak ISI. On channels with weak ISI, the equalizer outperforms both the CDFE as well as the ideal DFE with perfect feedback. Simulation results on well- known ISI channels support the findings.

The simplified Gaussian clustering algorithm (SGCA) adopting a reduced constellation is derived, then the dual-variance Gaussian clustering algorithm (DVGCA) is proposed, which makes the soft or hard decisions using the SGCA. Since the DVGCA is extremely reduced-complexity and operates in the soft or hard decisions mode according to the variance for error level of the equalizer output, it is effective for blind equalization of high-order quadrature amplitude modulation (QAM) signals. In addition, the blind decision feedback equalizer (DFE) using the DVGCA is presented, which leads to superior performance to ones without a feedback path over the channel with in-band spectral nulls.

Summary form only given. We develop improved joint source-channel (JSC) methods for decoding variable length encoded data based on residual source redundancy. Until very recently, all JSC methods based on residual redundancy assumed fixed length codewords. Recently, a practically realizable system which performed best over a significant range of channel conditions consisting of inner binary convolutional (BC) (bit-level) decoding, followed by outer (symbol-level) approximate maximum a posteriori (MAP) JSC decoding was suggested. Here we suggest two ways of improving on this method. First, a straightforward improvement is realized by using soft/probabilistic bit decisions output by the BC decoder, rather than hard decisions. Second, the JSC decoder can itself generate soft/probabilistic output, at the symbol level. The exact VLC minimum mean-squared error (MMSE) decoder has large complexity, similar to the exact MAP method, because the number of states increases with time. Thus we suggest an approximate MMSE method. In this approximate scheme, we first form a reduced directed graph, using the same MAP state reduction procedure as used for approximate MAP JSC decoding. Next, we rearrange the remaining states to form an (equivalent) directed graph. We then apply the forward/backward algorithm and a state merging procedure to this reduced graph to get approximate a posteriori probabilities, used for MMSE estimation.

Receiver operating characteristics (ROC) has been widely used as a performance evaluation tool to measure effectiveness of medical modalities. It is derived from a standard detection theory with false alarm and detection power interpreted as false positive (FP) and true positive (TP) respectively in terms of medical diagnosis. The ROC curve is plotted based on TP versus FP via hard decisions. This paper presents a three dimensional (3D) ROC analysis which extends the traditional two-dimensional (2D) ROC analysis by including a threshold parameter in a third dimension resulting from soft decisions, (SD). As a result, a 3D ROC curve can be plotted based on three parameters, TP, FP and SD. By virtue of such a 3D ROC curve three two-dimensional (2D) ROC curves can be derived, one of which is the traditional 2D ROC curve of TP versus FP with SD reduced to hard decision. In order to illustrate its utility in medical diagnosis, its application to magnetic resonance (MR) image classification is demonstrated.

In prior researches, low-density-parity-check (LDPC) code is applied on high-speed magnetic random-access-memory (MRAM) for a better correctable bit error rate (CBER). However, on another emerging memory, resistive random-access-memory (RRAM), reset hard breakdown errors have high-reliability log-likelihood ratio (LLR) which significantly degrades LDPC code soft decision CBER and decoding iterations. To solve this issue, two schemes, improved-LLR and reset-check-reverse-flag (RCRF) concatenating systematic LDPC code (RCRF + LDPC), are proposed. The first scheme increases soft decision CBER by using asymmetric LLRs for matching the distribution of reset hard breakdown errors. In addition, RCRF corrects at least one reset error per array subsection. So, both hard and soft decision CBERs are improved. The best performance is obtained by combining these proposals. The hard decision CBER is increased by more than 50% on a wide range of LDPC code options. Moreover, soft decision CBER is improved by 69% and 51%, respectively, on a 1KB codeword, 0.91 code rate quasi-cyclic-LDPC (QC-LDPC) code, and a 2KB codeword, 0.89 code rate Mackey's code.

In this paper, a context aware system is investigated for reading. The experiments are performed by two ultrasound sensors to obtain the users three scenarios of study, relax and nap. In this paper, we proposed soft decision (SD) method for three contexts to improve the accuracy rate of contexts recognition to 93% comparing with the 78% of hard decision (HD) method. In addition, to remove the external noise or interference, the moving windows (MW) are proposed to further improve the accuracy rate to 98%.

Application of cooperative multiple sensors for distributed signal sensing and modulation classification offers significant improvement of classification performance, compared to using single sensor, especially in the presence of multipath fading. In this paper, several cooperative AMC schemes with hard and soft decision fusion were analyzed. Classification performances of cooperative AMC schemes are investigated and evaluated in terms of average probability of correct classification, through Monte-Carlo simulations.