The computational error of the multilevel fast multipole algorithm is studied. The error convergence rate, achievable minimum error, and error bound are investigated for various element distributions. We will discuss the boundary between the large and small buffer cases in terms of machine precision. The needed buffer size to reach double precision accuracy will be clarified.

This paper presents an automated design of analog circuits starting with idealized elements. Our system first synthesizes circuits using idealized elements by a genetic algorithm (GA). GA evolves circuit topologies and transconductances of idealized elements to achieve the given specifications. The use of idealized elements effectively reduces search space and make the synthesis efficient. Second, idealized elements in a generated circuit are replaced by MOSFETs. Through the two processes, a circuit satisfying the given specifications can be obtained. The capability of this method was demonstrated through experiments of synthesis of a trans-impedance amplifier and a cubing circuit and benchmark tests. The results of the benchmark tests show the proposed CAD is more than 10 times faster than the CAD which does not use idealized elements.

In computer simulation of a large number of moving objects (MOs), how to enlarge Δt (the interval between the simulation time steps) without introducing causality errors is one of the primary keys to enhancing performance. Causality errors can be avoided by using the same Δt among related MOs when they are in the scene of detection (SoD). But in a large-scale MO simulation, MOs interact with one another in a complicated manner requiring a large calculation cost to predict the beginning time of SoD. In this paper we propose an event-aware dynamic time step synchronization method (DTSS) for distributed MO simulation, which increases Δt without introducing causality errors and speeds up the simulation. DTSS can be implemented with little calculation cost because: (1) DTSS does not calculate the beginning time of SoD exactly, but calculates the time for possible entry into SoD with a simple mechanisim, and (2) MO simulation consists of a "movement"-phase and a "detection"-phase in which the distance-calculation between MOs requires a heavy load, and DTSS utilizes the distance values to calculate Δt. In this paper, we also discuss a suitable HLA based time management mechanism to implement DTSS on a distributed computing environment. In the performance evaluation of DTSS, the calculation cost of DTSS is implemented by using the HLA suitable time management mechanism. The results show that DTSS can be executed within the ideal time plus its 1% over-cost when a basic scenario of war-game simulation is employed. Therefore if the ratio of SoD to the total simulation is small, the execution time is expected to decrease to nearly this ratio. We also introduce the criterion for determining when DTSS is superior to the conventional method by using the performance evaluation results. The results presented in this paper are effectively utilized when DTSS is applied to practical applications.

In this letter, a blind frequency offset estimation algorithm is proposed for OFDM systems. The proposed method exploits the intrinsic phase shift between neighboring samples in a single OFDM symbol, incurred by a frequency offset. The proposed algorithm minimizes a novel cost function, which is the squared error of the candidate frequency offset compensated signals from two different observation windows. Also, the solution of the proposed algorithm is given by an explicit equation, which does not require any iterative calculation. It is shown that the performance of the proposed method is better than those of the conventional methods, especially in the presence of multipath channels. This is due to the fact that the proposed method is insensitive to inter-symbol interference (ISI).

This paper presents a semi-blind algorithm for multiuser interference cancellation and fading amplitude estimation for downlink MIMO DS-CDMA systems with multipath fading channels. Taking advantage of the space-time information of the parametric multipath channel, the proposed algorithm first uses a space-time channel decoupler to suppress multiuser interference and then decomposes the channel into a set of parallel subchannels each containing the signal of the desired user on an individual multipath. Two criteria, the complementary orthogonal projection (COP) and the minimum variance distortionless response (MVDR), are employed by the space-time decoupler to achieve interference suppression and signal separation. The fading amplitudes can then be estimated from the eigen space of the output of the space-time channel decoupler. It follows that the signal of interest can be maximally combined in a pathwise manner and then differentially decoded.

In this paper, a novel cost effective interconnection network for two-way pipelined SIMD-based reconfigurable computing processor is proposed. Our reconfigurable computing engine is composed of the SIMD-based function units, flexible interconnection networks, and two-bank on-chip memories. In order to connect the function units, the reconfigurable network is proposed to connect all neighbors of each function unit. The proposed interconnection network is a kind of full and bidirectional connection with the data duplication to perform the data-parallelism applications efficiently. Moreover, it is a multistage network to accomplish the high flexibility and low hardware cost.

A novel pitch-synchronous auditory-based feature extraction method for robust automatic speech recognition (ASR) is proposed. A pitch-synchronous zero-crossing peak-amplitude (PS-ZCPA)-based feature extraction method was proposed previously and it showed improved performances except when modulation enhancement was integrated with Wiener filter (WF)-based noise reduction and auditory masking. However, since zero-crossing is not an auditory event, we propose a new pitch-synchronous peak-amplitude (PS-PA)-based method to render the feature extractor of ASR more auditory-like. We also examine the effects of WF-based noise reduction, modulation enhancement, and auditory masking in the proposed PS-PA method using the Aurora-2J database. The experimental results show superiority of the proposed method over the PS-ZCPA and other conventional methods. Furthermore, the problem due to the reconstruction of zero-crossings from a modulated envelope is eliminated. The experimental results also show the superiority of PS over PA in terms of the robustness of ASR, though PS and PA lead to significant improvement when applied together.

Recently, a versatile user cooperation method called coded cooperation diversity has been introduced, in which the codewords of each mobile are partitioned and transmitted through independent fading channels instead of simple repetition relay, to achieve remarkable gains over a conventional (non-cooperation) system, while maintaining the same information rate and transmission power. In this paper we present an adaptive space-time (AST) coded cooperation scheme based on the decoding operation on the first partition of the codeword at the base station and enables practical adaptive arrangement of resources to adopt the channel condition. Performance analysis and simulation results have proved that the proposed scheme greatly improves error rate performance and system throughput, compared with the previous framework.

Pulse-shaping OFDM is well-known that it performs well in a mobile environment compared with conventional OFDM. However, in a highly mobile environment intersymbol and intercarrier interferences (ISI/ICIs) increase and can no longer be neglected. These ISI/ICIs deteriorate the performance of the systems. Proper channel equalization is needed for further improvement of the systems. In this paper, a more general case, namely Biorthogonal Frequency Division System based on Offset QAM (BFDM/OQAM) is considered. We propose a multi-input multi-output (MIMO) transversal filter to equalize the time-frequency dispersive channel. Tap weights are calculated using the zero-forcing (ZF) algorithm. We also propose maximum-likelihood channel estimator and its low-complexity version. The proposed system can significantly improve the performance of BFDM/OQAM systems in the highly mobile environment.

For super-parallel video processing, we proposed a power- and area-efficient SRAM core architecture with a segmentation-free access, which means accessibility to arbitrary consecutive pixels, and horizontal/vertical access. To achieve these flexible accesses, a spirally-connected local-wordline select signal and multi-selection scheme in wordlines are proposed, so that extra X-decoders in the conventional multi-division SRAM can be eliminated. Consequently, the proposed SRAM reduces a power and area by 57-60% and 60%, respectively, when it is applied to a 128 parallel architecture. The proposed 160-kbit SRAM with 16-read ports (2-read port SRAM with eight-parallel architecture) is implemented to a search window buffer for an H.264 motion estimation processor core which dissipates 800 µW for QCIF 15-fps in a 130-nm technology.

A highly collision-resistive RFID system multiplexes communications between thousands of transponders and a single reader using TH-CDMA based anti-collision scheme. This paper focuses on the back-end design consideration of such an RFID system with the deployment of high-level modeling techniques, accompanying a technical comparison of physical-level description, hardware-based emulation, and software-based simulation. A new rapid-prototyping simulation system was constructed to evaluate the robustness of a multiplexed RFID link system with more than 1,000 channels in the presence of field disturbances, and the design parameters of the back-end digital signal processing that dominated anti-collision performance were explored. Finally, the derived optimum parameters were applied to the design of a back-end digital integrated circuit to be installed in collision-resistive transponder circuitry.

We designed a CMOS image sensor capable of capturing variably smoothed images. This sensor uses a negative-feedback technique to set photodiode (PD) capacitance in the pixel circuit to any intermediate voltage during charge accumulation and it provides a neighboring-pixel operation by using their average value when resetting the PD capacitance. Smoothing-filter coefficients are changed by adjusting timing of the pixel-readout and neighboring-pixels operations. The performance of the proposed sensor was evaluated by SPICE simulation and numerical analysis.

This paper presents an asynchronous multiple-valued current-mode data-transfer controller chip based on a 1-phase dual-rail encoding technique. The proposed encoding technique enables "one-way delay" asynchronous data transfer because request and acknowledge signals can be transmitted simultaneously and valid states are detected by calculating the sum of dual-rail codewords. Since a key component, a current-to-voltage conversion circuit in a valid-state detector, is tuned so as to obtain a sufficient voltage range to improve switching speed of a comparator, signal detection can be performed quickly in spite of using 6-level signals. It is evaluated using HSPICE simulation with a 0.18-µm CMOS that the throughput of the proposed circuit based on the 1-phase dual-rail scheme attains 435 Mbps/wire which is 2.9 times faster than that of a CMOS circuit based on a conventional 4-phase dual-rail scheme. The test chip is fabricated, and the asynchronous data-transfer behavior of the proposed scheme is confirmed.

Combined input-crosspoint buffered (CICB) packet switches have been of research interest in the last few years because of their high performance. These switches provide higher performance than input-buffered (IB) packet switches while requiring the crosspoint buffers run at the same speed as that of the input buffers in IB switches. Recently, it has been shown that CICB switches with one-cell crosspoint buffers, virtual output queues, and simple input and output arbitrations, provide 100% throughput under uniform traffic. However, it is of general interest to know the maximum throughput that a CICB switch, with no speedup, can provide under admissible traffic. This paper analyzes the throughput performance of a CICB switch beyond uniform traffic patterns and shows that a CICB switch with one-cell crosspoint buffers can provide 100% throughput under admissible traffic while using no speedup.

In this paper, we compare two signal designs for uplink quasi-synchronous code division multiple access (CDMA) channels in order to optimize the trade-off between bandwidth efficiency and power efficiency. The design we call band-limited DS/CDMA design, is based on the time-domain assignment of Gold sequences, just as in the ordinary DS/CDMA, but with band-constrained cyclic chip interpolation functions, which is unlike the ordinary DS/CDMA. The other design, MC/CDMA design, is based on frequency-domain assignment of the sequences, as in the ordinary MC/CDMA. In both designs, we assume insertion of guard intervals at the transmitter and frequency-domain processing in reception. Assuming quasi-synchronous arrival of CDMA signals at the CDMA base station and FFT in the effective symbol interval, the intersymbol interference is evaded in both designs. First we identified the signal parameters that optimize bandwidth efficiency in each of the band-limited DS design and MC design. Second, we clarified the signal parameters that optimize the power efficiency as functions of frequency efficiency in each of the two designs. Finally, we derived and compared the trade-off between the bandwidth efficiency and power efficiency of band-limited DS and MC designs. We found a superiority of band-limited DS design over MC design with respect to the optimized trade-off.

This paper discusses the on-line frequency identification problem for a measured sinusoidal signal by using the adaptive method and filter theory. The proposed method is based on an identity between the sinusoidal signal and its second order derivative. For a set of chosen parameters, the proposed method is robust to the initial phase, the amplitude, and the frequency in a wide range. The convergence rate can be adjusted by the chosen parameters. The estimation error mainly depends on the frequency of the sinusoid, the measurement noise and a key design parameter.

This paper considers systems consisting of communicating processes which can move between specified locations. Mobile telephone systems and intelligent transport systems are examples of them. The processes can exchange data as well as channels (e.g. frequencies) to be used in further communication. It may happen that two processes (e.g. telephones or cars) in different locations could communicate directly because they share a communication channel, but they cannot as there is a physical obstacle between the locations. A possible solution is to allow one process to send a message to another one through other processes and locations. To see if this is possible, a notion of connectivity of processes has been devised in the literature. A process was defined to be connectable to another one if a message from the first one could reach the other one by using any existing communication actions, allowed locations, and process moves in the system. A process-algebraic approach for checking connectivity was proposed. In this paper, it is shown how the temporal logic of actions (TLA) can be employed for the same purpose. In both approaches, connectivity of a process with another one is basically checked as follows. The first process includes a marking message in all its sending actions. Every process that receives this message gets marked and includes it in its own sending actions. The first process is connectable to the second one if there exists such a system execution sequence that the latter gets marked in it. Since TLA is a linear-time temporal logic, it can generally not express such a property. This paper, however, shows that it can be expressed and verified for a given TLA system specification. It also shows how to specify the marking operations and provides an example of connectivity checking.

Irregular Repeat-Accumulate (IRA) codes, introduced by Jin et al., have a linear-time encoding algorithm and their decoding performance is comparable to that of irregular low-density parity-check (LDPC) codes. Meanwhile the authors have introduced detailedly represented irregular LDPC code ensembles specified with joint degree distributions between variable nodes and check nodes. In this paper, by using density evolution method [7],[8], we optimize IRA codes specified with joint degree distributions. Resulting codes have higher thresholds than Jin's IRA codes.

In this letter, we define the generalized doubly stochastic processing via Jacket matrices of order-2n and 2n with the integer, n≥2. Different from the Hadamard factorization scheme, we propose a more general case to obtain a set of doubly stochastic matrices according to decomposition of the fundaments of Jacket matrices. From order-2n and order-2n Jacket matrices, we always have the orthostochastoc case, which is the same as that of the Hadamard matrices, if the eigenvalue λ1 = 1, the other ones are zeros. In the case of doubly stochastic, the eigenvalues should lead to nonnegative elements in the probability matrix. The results can be applied to stochastic signal processing, pattern analysis and orthogonal designs.

Multiple-input multiple-output (MIMO) systems using eigenbeam space division multiplexing (E-SDM) perform well and have increased capacities compared with those using conventional space division multiplexing (SDM). However, channel state information (CSI) is required at a transmitter, and the performance of E-SDM systems depends much on the accuracy of the CSI at a transmitter and a receiver. In time-varying fading environments, the channel change between the transmit weight determination time and the actual data transmission time causes the system performance to degrade. To compensate for the channel error, a linear extrapolation method has been proposed for a time division duplexing system. Unfortunately, the system performance still deteriorates as the maximum Doppler frequency increases. Here, two new techniques of channel extrapolation are proposed. One is second order extrapolation, and the other is exponential extrapolation. Also, we propose maximum Doppler frequency estimation methods for exponential extrapolation. Simulation results for 4tx 4rx MIMO systems showed that using the proposed techniques, E-SDM system performs better in a higher Doppler frequency region.