Shiunn-Jang Chern

A generalized decomposition approach to multi-h continuous phase modulation

A decomposition approach to continuous phase modulation (CPM), mostly due to the work of Rimoldi (1988), is significant in the design of combined coding and communication signaling scheme. In this paper the generalized decomposition model is developed and applies to the multi-h CPM. It is shown that any multi-h CPM can also be realized by a continuous phase encoder (CPE) concatenated with a memoryless modulator (MM). Moreover, depending on the manner that the modulation indexes are selected for each symbol interval, the CPE can be either a linear or nonlinear feedback convolutional encoder, in which both the CPE and MM are time-variant. Finally, some applications are employed to verify the advantage of the proposed decomposition model, namely, (1) natural trellis representation of the CPE for M-ary single-h CPM with different modulation indexes that have the same denominator P, (2) a reduction of the number of states of the MLSD receiver for conventional multi-h CPM, (3) asymmetrical multi-h CPM has no gain over conventional multi-h CPM, (4) nonlinear CPM has a CPE that is nonlinear.

A low cost wireless video transmission network with padding-based distributed video coding

Distributed Video Coding (DVC) based on Distributed Source Coding (DSC) theory has been developed since 2002. The main concerns of DVC are to move high computational complexity (CC) in the encoder of video coding to the decoder side and to lower the cost of portable devices. Motivated by this idea, in this paper we present a new wireless video transmission network architecture, which combines the padding-based DVC, Digital Video Transcoding (DVT) and Decoding-Friendly Encoder Design (DFED). With this new configuration, we could shift the CC from the portable device to the network side, such that to lower the cost of portable device and whole video network system. The advantages of the proposed wireless video network framework are shown, which could achieve lower cost and retain the desired quality, at the same time.

Differential Detection of Serially Concatenated Precoded GMSK with Iterative Decoding

Differential detection for a serially concatenated coding system consists of an outer convolutional encoder, an interleaver and an inner rate-1 precoded Gaussian minimum-shift keying (GMSK) modulation is addressed. Using differential phase detector (DPD) combined with an iterative MAP (maximum a posteriori) decoding scheme is devised to improve the system performance. It is shown that the combination of GMSK modulation and DPD results in a feedforward tapped delay line. Therefore, a rate-1 recursive precoder is employed in front of the GMSK to produce a combined inner MAP decoder with a trellis of four states only. Numerical results show that the proposed scheme outperforms previous work about 2 dB in the AWGN channel and about 9 dB in the flat Rayleigh fading channel at BER is equal to 10-4 respectively.

Adaptive semi-blind channel estimation for ST-BC MIMO-CDMA systems with hybrid user signature

We extend our previous work to present a new semi-blind transceiver for the direct-sequence code division multiple access (DS-CDMA) system that uses multiple transmit and receive antennas (MIMO) system, equipped with space-time block code (ST-BC). In the transmitter we design a new hybrid augmented user signature (AUS) that composes of the desired user signatures and the prefix/postfix zero-padding sequences with respect to individual transmit-antennas. The hybrid AUS is devised to resolve the phase ambiguity problem which occurs in all blind receivers. At the receiver we propose a Capon-like semi-blind two-branch filter bank receiver, based on the linearly constrained constant modulus (LCCM) criterion, followed by the AUS-assisted semi-blind channel estimation and power method for block symbol recovery. This enables us to partially alleviate the effects of inter-block interference (IBI) and the multiple access interference (MAI). In the ST-BC MIMO-CDMA receiver with two-branch filterbank, we build on the generalized sidelobe canceller (GSC) structure with the RLS for implementing the adaptive semi-blind LCCM receiver. Via intense simulations it reveals that our proposed new transceiver has robust performance against the users acquisition inaccuracies comparing with current available algorithms and to resolve the phase ambiguity problem.

Performance of the MIMO CS-PRP-OFDM systems with complementary codes

This paper presents a new transceiver framework of the MIMO-OFDM systems, with the space-time block code (ST-BC). Unlike the conventional Pseudo-Random Postfix (PRP) - OFDM in our proposed framework the null samples of zero-padding (ZP)-OFDM is replaced by known cyclic postfix sequence (CPS), weighted by a pseudo-random (PR) scalar. Since the CPS is implemented by the cyclic-shift (CS) complementary code (CC) sequences, the proposed transceiver scheme is referred to as the MIMO CS-PRP-OFDM systems. By exploring the useful property of CC sequences, convolved with channel information, the receiver design associated with the semi-blind channel estimation of the proposed MIMO CS-PRP-OFDM systems is affected only by the background noise. It avoids the interference of the transmitted signals, and yields achieving better system performance, in terms of symbol error rate, compared with the conventional PRP-OFDM based systems, with less complexity. This is especially true when the signal-to-noise ratio is increased.

A simple spectra computing approach for continuous-phase phase modulation (CPPM)

Recently, we have developed a special class of continuous-phase modulation (CPM) scheme, called the continuous-phase phase modulation (CPPM); it transmits the PAM information sequences using time-limited phase shaping pulses. In this paper, the power spectra for this novel CPM are investigated. We employ autocorrelation function approach to modify the CPM spectra computing formula for time-limited phase shaping pulse. Since CPPM is a CPM modulator with time-limited phase shaping function, previous research believed that such modulator will have data-independent spectral lines and suggested to be avoided. However, we show that the spectral lines can be diminished by choosing appropriate modulation index, modulation level and phase shaping pulse. Spectra for two classes of CPPM schemes are given. It is shown that the bandwidth of CPPM scheme can be effectively reduced by using smaller modulation index and/or increasing the correlative length and modulation level. The resultant CPPM schemes have attractive spectral occupancy as compared to conventional CPM.

A novel class of continuous-phase modulation (CPM) with separable phase property

Continuous phase modulation (CPM) is a power and bandwidth efficient modulation yet the receiver complexity is usually too high. Therefore, receiver complexity reduction techniques become very important. Among the complexity reduction techniques, the PAM decomposition approach for CPM signals have recently gained much attention, due to its effectiveness and excellent BER performance. The PAM decomposition approach decomposes CPM signals into a set of PAM waveforms and employs only significant part of the PAM waveforms to design receiver. More recently, a PAM waveforms reduction method for a special class of CPM signals, which have the so-called separable phase property, enables the CPM to be reformulated as an correlatively encoded equivalent CPM scheme, such that the equivalent CPM can be represented as less PAM waveforms than the original CPM. Nevertheless, most of the known CPM family does not satisfy the separable phase property. This paper proposes a new class of CPM with phase separable property. It is found that the new class of CPM can reduce the number of filters to 25 percent of the MLSD receiver required. Moreover, the bandwidth and power efficiency of this new set of CPM are very close to the conventional CPM and sometimes are even better.

New ST-BC MIMO-CDMA transceiver with augmented user signatures

This paper presents new transceiver framework for the DS-CDMA systems that use the multiple-input multiple-output (MIMO) antennas, with space-time block code (ST-BC). In the transmitter, new hybrid (prefix/postfix) zero-padding sequences combining with the desired user signature is exploited to form the hybrid augmented user signature sequences, with respect to different transmit-antennas. While in the receiver under the GSC framework a new Capon-like semi-blind MIMO-CDMA filterbank detector based on the linearly constrained constant modulus (LCCM) criterion, is derived. Also, the adaptive GSC-RLS algorithm is developed for adaptively implementing the blind LCCM MIMO-CDMA filterbank of the receiver. With the specific design of the hybrid augmented user signature sequences, we are able to resolve phase ambiguity problem of blind channel estimation. Computer simulations verify the merits of the proposed new transceiver framework; it can be employed to achieve better performance, and is more robust against the discrepancy due to Capon channel estimation.

Padding block based DVC coding scheme with mutual bi-directional frame coding at decoder

In this paper, we apply the mutual bi-directional frame coding (MB-frame coding) to our previous proposed padding block based Distributed Video Coding (DVC) scheme for performance improvement. Basically, in the MB-frame coding approach, we adopt the mutual forward and backward video sequence frames to process within reference frames in each individual Group of Pictures (GOP) until no skip blocks could be padded at decoder. It is worth to noting that with this approach the computation complexity does not increase at encoder side, which is consistent with the basic concept to meet the low complexity DVC requirement. Indeed, it achieves the overall system performance improvement with the cost of increasing for average 30 % of computation complexity at decoder side. Via computer simulation we show that the experimental results with the MB frame coding approach has about 0.2 dB gain over the one which does not employ the MB-frames coding scheme.

A low complexity video coding for combining RFID and video surveillance with padding based DVC

Radio frequency identification (RFID) tag has been extensively used in diverse applications, recently. For instance, we may use it to track freights on a warehouse, to catch a thief for lost luggage and to rapidly locate lost children or passengers who fail to arrive on time at the departure gates of major airport. It can also be employed for transmitting multimedia content (especially for video). Presently, many research works focus on designing the RFID tag with high transmission data rate. In this paper we present a new practical scheme, it integrates the video surveillance into RFID tag at range of 250 kbps data rate for video data transmission. We believe that has not been addressed before in the literature. Here the Distributed Video Coding (DVC) scheme is integrated into RFID tag, where the so-called padding-based DVC scheme, proposed by us recently, is deployed. It is known that the padding-based DVC scheme has the advantages of requiring very low complexity and achieving low bit rate. Via computer simulations we show that the proposed scheme can easily work at data rate less than 250 kbps, and achieve better performance (PSNR) than the conventional approaches.