The problem of jointly optimizing the transmitter and receiver beamforming weights in the downlink of a cyclic prefix-free MC-CDMA system over multipath fading channels is addressed in this paper. Both the base station and user terminals are equipped with antenna arrays to leverage on the spatial information provided by the antenna arrays. The preceding and equalization matrices are designed with the aim of minimizing the overall mean-squared-error (MSE) of the system subject to a power constraint for each transmitter antenna, which has the additional benefit of reducing the probability of clipping of the transmitted signals from the transmitter. This is highly favorable for a MC-CDMA system due to the multi- carrier modulation involved. The performance of the optimization process is supported by simulation results.

In this paper, we develop and present a DFT-SOFDM system (operating in both distributed and localized FDMA modes) for transmission and reception through a frequency-selective, slow fading channel. This system allows us to use well-known linear receiver techniques such as the zero-forcing (ZF) and minimum mean-squared error (MMSE) equalization. More importantly, we proposed using a sphere decoding algorithm (SDA) on DFT-SOFDM to obtain a maximum-likelihood (ML) estimate on the received signal, hence achieving superior bit error rate (BER) performance without the complexity of an exhaustive search. Simulation results show that SDA gives superior BER performance for both distributed and localized DFT-SOFDM compared to suboptimal ZF or MMSE receivers.

A block-spreading code division multiple access (BS-CDMA) system is proposed for broadband uplink transmission, giving rise to a significantly improved multiuser performance without using complex multiuser detection techniques. This is because the code orthogonality is easily maintained when channel variation across the consecutive blocks, in a block-by-block high-speed transmission, is negligible. The proposed system uses frequency domain equalisation at the receiver to combat multipath interference efficiently over frequency selective fading channels. We propose despreading before equalisation, which reduces the frequency domain process to symbol-wise operation. A cell-specific scrambling code is employed to suppress other-cell interference for uplink transmission in a multicell system. Our analytical and simulation studies show that the proposed BS-CDMA system has superior multiuser performance over the conventional direct sequence CDMA and cyclic prefix CDMA systems for uplink transmission.

The transmission of a Discrete Fourier Transform Spread Orthogonal Frequency Division Multiplexing (DFT-SOFDM) signal through a time-selective fading channel suffers from severe performance degradation due to the Doppler shifting of the transmitted spectrum. In this paper, we propose a multiple-input multiple-output (MIMO) DFT-SOFDM scheme and a new technique to overcome the adverse effects caused by Doppler shift. By appropriately interleaving the transmitted signals in space and time, our scheme allows us to use linear equalization techniques such as the zero-forcing (ZF) and minimum mean-squared error (MMSE) equalization. In addition, the use of sphere decoding algorithm (SDA) provides superior BER performance compared to the ZF or MMSE solution. Simulation results show that at a BER of 10 3, SDA provides a 13.5 dB gain over ZF solution, and a 13 dB gain over MMSE solution. The system with SDA shows negligible performance degradation when the system load in the uplink is increased: at Eb/No= 21 dB, a BER of 1.88 times 10-5is achieved when there are 2 users in the system, but degrades only to 2.19 times 10-5for a 8 user system. In addition, SDA used with our proposed signal model offers significantly lower worst case complexity as compared to the conventional DFT-SOFDM with SDA.

We propose an open-loop two-dimensional linear pre-transformed (2DLPT) multiple-input multiple- output (MIMO) orthogonal frequency division multiplexing (OFDM) system. ThenTNtimesnTN2DPLT, withnTthe number of transmit antennas and N the FFT size in the OFDM modulation, is applied in both spatial and frequency domains and is unitary in individual as well as both domains. While the spatial domain transform converts the transmit diversity from the other transmit antennas to frequency diversity, the frequency domain transform fully exploits the frequency domain diversity made available. More over, the unitary transform guarantees the same ergodic capacity as the non-transformed spatial division multiplexing MIMO-OFDM channel. Therefore, the proposed 2DLPT MIMO-OFDM system achieves full diversity ofnTnRnLwithnRandnLrepresenting the number of receive antennas and the number of multipaths in each transmit-receive antenna pair channel, and full capacity.

We present a new probability density function (PDF) based on the generalized Laguerre expansion that, for the first time, completely characterize the widest range of small scale fading distributions in wireless communications. This unified PDF formula characterizes the most general case of small scale fading consisting of an arbitrary number of specular waves plus diffused power. The same formula covers all existing distributions for small scale fading, which include the Nakagami, Rician, and Rayleigh fadings, as special cases.

The rate-one semi-orthogonal algebraic space-time (SAST) codes is recently introduced. In this paper, the authors propose a new signal design method for SAST codes so that the transmitted symbols can be divided into four groups for maximum likelihood (ML) detection; the authors call this new design of SAST codes four-group decodable SAST (4Gp-SAST) codes. The decoding complexity of 4Gp-SAST codes is greatly reduced compared with that of the original SAST codes, where the ML decoding of the transmitted symbols is made into two groups. The exact pair-wise error probability of 4Gp-SAST codes is derived and it is used to optimize the signal designs. Simulation results show that 4Gp-SAST codes perform better than several low complexity space-time block codes such as orthogonal and quasi-orthogonal codes.

The generic algebraic structure of a quasi-orthogonal space-time block code (QO-STBC) has been derived, and group-constrained linear transformation (GCLT) as a mean to optimise the diversity and coding gains of a QO-STBC with square or rectangular QAM constellations has been proposed. Compared with QO-STBC with constellation rotation (CR), QO-STBC with GCLT requires only half the number of symbols for joint detection, hence lower maximum-likelihood decoding complexity. The optimum GCLT parameters for QO-STBC with square QAM constellation have been derived analytically. The optimised QO-STBCs with GCLT are able to achieve full-transmit diversity and have negligible performance loss compared with QO-STBCs with CR at the same code rate.

In this paper, we proposed a new multicarrier, direct sequence code division multiple access, the so-called premulticoded MC-DS-CDMA system. In this scheme, the basic MC-DS-CDMA system is augmented with a precoder and a multicode encoder at the transmitter. The multicode encoder is an inverse Walsh-Hadamard transform (IWHT) unit, while the precoder is a recently suggested ldquoconstant amplituderdquo encoder that guarantees that the IWHT output stream has binary level. While IWHT, multicarrier (MC) transmission, and direct sequence spreading provide us with frequency diversity, the constant amplitude precoder ensures that the transmitted waveform is spread BPSK signal, minimizing peak to average-power ratio (PAPR) problem. The multicode and (MC) combination also addresses the need for flexible high data rate services in present day wireless communications. In the receiver, after matched filtering or correlation demodulation for each subcarrier and WHT operation, a detection strategy based on minimum-mean-square-error (MMSE) criterion is used. We have derived an upper bound for the bit error rate (BER) performance in Nakagami fading channel. Numerical results are compared with those obtained from a single carrier DS-CDMA system (SC-DS-CDMA) and a MC-DS-CDMA scheme without the precoding and IWHT units. We show that our proposed MC-DS-CDMA scheme has a much better BER performance compared with both of these schemes.

Proper partitioning of a unitary space-time (UST) signal set is essential for the optimal design of trellis coded unitary space-time modulation (TC-USTM). The unique properties of these UST signals have necessitated a different partitioning methodology from that of the conventional two dimensional constellations. In this letter, we propose a systematic set partitioning through a novel subset-pairing strategy, for an arbitrary UST signal set. This approach leads to a geometrically congruent partitioning, i.e., subsets of the same size (order) have identical intra-distance profiles. Based on this partitioning, the resulting TC-USTM can achieve a minimum bit error probability.

We study space-time precoding techniques for asymmetric multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) channels with more transmit than receive antennas which is a typical downlink setup. Under the condition of perfect channel state information (CSI) at the receiver but no CSI at the transmitter, we consider linear dispersion coding (LDC), groupwise space-time block coding (GSTBC), quasi-orthogonal space-time block coding (QSTBC), and spatial spreading (SS) and study their capacity and diversity performances. Our results show that for both ergodic and outage capacities, SS is worse than LDC, GSTBC and QSTBC, while the latter three have very close/same performance. We also show that by adjusting the phase shift values in the SS matrices, its outage capacities can be improved. In terms of diversity, LDC, GSTBC and QSTBC have both transmit and receive spatial diversity, and LDC's diversity order is slightly higher than the latter two. SS, on the other hand, has not only receive spatial diversity, but also introduces additional frequency diversity. This contributes to their very close bit error rate (BER) performance when turbo processing is deployed at the receiver. Considering the implementation flexibility and complexity, we therefore recommend SS and GSTBC as the promising precoding schemes for asymmetric MIMO-OFDM channels.

Frequency-domain equalization (FDE) is an effective technique to mitigate the distortion induced by long-spanning inter-symbol interference, but the receiver requires to estimate the channel information. In this paper, we present a training-based frequency-domain channel frequency response (CFR) estimation scheme, which is hardware efficient when integrating with FDE for space-time (ST) coded multiple-input multiple-output (MIMO) transmissions over frequency-selective fading channels. In this proposed scheme, training sequences are ST block encoded at the transmitter. Similar to when performing FDE, the training sequences are processed after the FFT operation to obtain the CFR. A thorough analysis of the CFR estimation mean square error (MSE) is provided based on both the least squares (LS) and minimum MSE (MMSE) criteria, with the transmit power constraint and the optimal choices of training signals being taken into account. Finally, we demonstrate that when the number of transmit antennas is equal to 2, this CFR estimation could be implemented recursively by means of certain adaptive algorithms. Simulation results are presented to demonstrate the effectiveness of our proposed scheme in joint with FDE in multipath fading channels.

We show that there exists an interesting relationship between the algebraic constraints of Amicable Orthogonal Design (AOD) matrices and the dispersion matrices of Quasi-Orthogonal Space-Time Block Code with minimum decoding complexity (MDC-QOSTBC). This insight leads to a novel concept of "Preferred AOD Pair", and we show that MDC-QOSTBC can be constructed from two AODs that form a Preferred AOD Pair. With the Preferred AOD Pair concept, we are able to derive the lower bound on code rate of square MDC-QOSTBC, and show them to be 1 and 3/4 for four and eight transmit antennas respectively. We also present a systematic method to construct Preferred AOD Pair and use them to obtain new MDC-QOSTBCs.

In this paper, we present a hybrid receiver scheme to overcome the shortcoming of the conventional interference cancellation (IC) receiver, for multiuser multicode CDMA (M-CDMA) systems. In conventional IC receivers, RAKE receiver is used in the first stage to obtain initial data estimation but substantial interference at the output of the RAKE receiver renders the initial data estimation highly inaccurate. Our proposed scheme consists of an equalizer for obtaining initial data estimation, followed by a parallel interference cancellation (PIC) section for enhancing the accuracy of data estimation. Simulation results are presented, which show that our proposed scheme achieves much better BER performance than the RAKE-PIC scheme

We show a new design of unitary differential space-time modulation (DSTM) with low decoding complexity, i.e. its decoding can be performed by two parallel decoders where each of the decoder has a search space of only radicN for a codebook with N codewords. The design is based on orthogonal space-time block code (O-STBC) and spherical code. We separate the symbols of an O-STBC into two groups and jointly modulate the symbols within a group using a joint constellation set constructed from spherical code. The proposed unitary DSTM scheme has a lower decoding complexity than many other DSTM schemes including those based on group codes and Sp(2) with better or comparable decoding performance. It also achieves a better performance than existing DSTM schemes with similar decoding complexity

The maximum likelihood (ML) estimator and the Cramer Rao Bound (CRB) for frequency offset estimation are derived and compared against each other in a multiband orthogonal frequency division multiplexing (MB-OFDM) ultra-wideband (UWB) system. A total of 100 UWB channel realizations, as proposed by IEEE 802.15 Channel Modeling sub-committee, for each channel model (CM) 1: line-of-sight for 0-4m and CM2: non-line-of-sight for 0-4m are simulated to demonstrate the convergence of the ML estimator variance with the CRB for high signal-to-noise ratio (SNR).

In this paper, we study various frequency-domain equalizers (FDE) for single-carrier space-time block coded transmissions over frequency-selective fading channels. The linear and decision feedback equalizers based on zero forcing (ZF) and minimum mean-square error (MMSE) criteria are considered. These methods offer different tradeoff in terms of performance and the a priori knowledge of the data and receiver noise statistics, which is discussed. Besides, we also consider the adaptive FDE based on some low-complexity block-wise recursive algorithms, which obviates the need for distinct channel estimation and realizes joint channel estimation and equalization. Performance of these FDE schemes are investigated by applying them to the EDGE system in the multipath fading environment. Simulation results demonstrate their effectiveness and also provide implications for selection of these FDE schemes in practice

We study space-time precoding techniques for asymmetric multiple-input multiple-output (MIMO) channels with more transmit than receive antennas which is a typical downlink setup. Under the condition that channel state information (CSI) is perfectly known at the receiver, but not known at the transmitter, we consider groupwise space-time block coding (GSTBC), quasi-orthogonal space-time block coding (QSTBC), linear dispersion coding (LDC), and spatial spreading (SS). We show that for narrow-band i.i.d. circularly symmetric complex Gaussian (CSCG) MIMO channels, SS has the lowest ergodic capacity, while GSTBC, QSTBC and LDC have almost the same ergodic capacity. We then compare their input-output mutual information when fixed-order modulation is deployed and show that the same performance order maintains. Finally we look into their bit error rate (BER) performance for uncoded systems, and show that LDC slightly outperforms GSTBC and QSTBC in high SNR regions and below BER of 10-2, and SS has the worst performance. Due to the good flexibility for various asymmetric channels and its simple implementation, we therefore recommend GSTBC as the asymmetric MIMO precoding scheme

In this paper, a joint symbol timing and frequency-offset estimator based on the maximum likelihood (ML) criterion, is derived for OFDM systems over frequency selective fading channels. .We use the linear minimum mean square estimation technique (LMMSE) to estimate the nuisance parameters such as transmitted signals and fading gains, and arrive at an algorithm to jointly estimate symbol timing and frequency offset. Simulation results show that our proposed symbol timing estimator performs better than the estimator reported by van de Beek et al (1997), while our proposed frequency offset estimator achieves a lower MSE at higher SNR (/spl ges/ 15 dB). Simulation results also show that our proposed symbol timing and frequency offset estimator has better MSE performance than the cyclostationarity-based estimator studied in T. Pollet and M. Moeneclaey (1995) at higher SNR (for example 15 dB). However, in terms of probability of correct timing synchronization and MSE, our symbol timing estimator achieves roughly the same performance as the estimator by Lee et al (2002), although a different approach is used in our estimator. Our proposed estimator is next modified to two practical ones independent of SNR, furthermore, one of which needs only the knowledge of the maximum delay spread to perform estimation. We also extend our proposed estimator to applications where multiple OFDM symbols can be observed to perform synchronization.

We present a novel differential space-time modulation (DSTM) scheme that is single-symbol decodable and can provide full transmit diversity. It is the first known single-symbol-decodable DSTM scheme not based on orthogonal STBC (O-STBC), and it is constructed based on the recently proposed minimum-decoding-complexity quasi-orthogonal space-time block code (MDC-QOSTBC). We derive the code design criteria and present systematic methodology to find the solution sets. The proposed DSTM scheme can provide higher code rate than DSTM schemes based on O-STBC. Its decoding complexity is also considerably lower than DSTM schemes based on Sp(2) and double-symbol-decodable QO-STBC, with negligible or slight trade-off in decoding error probability performance

We show that minimum-decoding-complexity quasi-orthogonal space-time block code (MDC-QOSTBC) can be used to form a new differential space-time modulation (DSTM) scheme to provide full transmit diversity with single-symbol decodable complexity. This is the first known single-symbol decodable DSTM not based on orthogonal STBC. We derive the DSTM code design criteria and use them to construct customized constellation sets that provide full diversity and maximum coding gain. Our new DSTM scheme can provide a higher code rate and better decoding performance than DSTM schemes based on rate-1/2 orthogonal STBC. It also has a lower decoding complexity than the other DSTM schemes, as its maximum likelihood decoding only requires the joint detection of two real symbols.

We address the throughput and delay performance of a direct sequence spread spectrum (DS-SS) slotted ALOHA packet radio system, where all users use a common spreading code to communicate with a central station. Unlike in previous works where packets are constrained to arrive on a finite set of discrete times, we consider in this paper that packet arrivals are continuously and uniformly distributed over an arbitrary uncertainty interval. We derive a closed form multi-packet capture probability and obtain the throughput and delay expressions for the DS-SS slotted ALOHA packet radio system. We use simulations to validate our analysis.

In this paper, we present low-complexity adaptive frequency-domain equalization (FDK) for space-time block coding (STBC) direct-sequence code-division multiple access (DS-CDMA) downlink system. The proposed scheme is powerful and computationally efficient, as the sophisticated matrix decomposition operations are not required in the processing and some block recursive processing, such as block least mean square (BLMS) algorithm and modified block recursive least squares (BRLS) algorithm can be easily incorporated for equalizer adaption. Simulation results demonstrate the effectiveness of such adaptive FDE with regard to different Doppler shifts.

An algorithm based on distinguishing the first significant multipath (ETA) is proposed to address the critical issue of symbol timing synchronization and carrier frequency offset estimation in Ultra wideband (UWB) Multi-band OFDM (MB-OFDM) systems. FTA pinpoints the frame synchronization sequence and the start of its Fast Fourier Transform (FFT) window by accumulating multipath energies and then discerning for first significant multipath component through threshold comparison between consecutive accumulated energy samples. Thereafter, the frequency offset (FO) is estimated from phase differences between training symbols. Two channel models (CM), as proposed by IEEE 802.15 Channel Modeling sub-committee, CM1: line-of-sight for 0-4m and CM2: non-line-of-sight for 0-4m are simulated with 100,000 noisy realizations per threshold. The timing offset mean-squared errors for CM1 is 0.371 OFDM samples and CM2 is 4.548 OFDM samples under a typical signal-to-noise ratio (SNR) of 17 dB for high data rate applications. These are orders of magnitude lower compared to conventional OFDM-extended correlation methods. A maximum likelihood FO estimation algorithm is derived and its performance is compared with FTA simulation results. Variance of the FTA FO estimate approaches that of maximum likelihood FO estimate with a SNR loss of less than 0.5 dB.

We propose and construct a new class of full-rate full-diversity Quasi-Orthogonal Space-Time Block Code (QO-STBC), namely Four-Group QO-STBC (4Gp-QOSTBC), which can support any number of transmit antennas. 4Gp-QOSTBC can linearly separate the symbols into four independent groups, such that symbols within a group is orthogonal to all other symbols in another groups, and the maximum-likelihood (ML) decoder of the code only needs to jointly decode the symbols within the same group. The number of symbols required for joint detection of the newly proposed 4Gp-QOSTBC is half of that required by the existing QO-STBCs with the same code rate, and their full-diversity decoding performances are comparable. We also extend the proposed code construction to obtain 6Gp-QOSTBC and 8Gp-QOSTBC, which have even lower decoding complexity, albeit at a slight loss in code rate.

In this paper, several new high-rate STBCs with transmit diversity gain of two are proposed. The proposed STBCs can flexibly support either three or four transmit antennas, and a code rate of between two to four. We show that they have good symbol dispersion property that leads to better BER performance under spatially-correlated MIMO channels or when the number of transmit antennas is reduced from four to three. They also have better coding gain and lower decoding complexity than some existing high-rate STBCs at the same spectral efficiency.

A new class of quasi-orthogonal space-time block code (QO-STBC) namely minimum-decoding-complexity QO-STBC (MDC-QOSTBC) has recently been proposed in the literature. In this paper, we analyze some of its essential code parameters and code properties. Specifically we derive its maximum achievable code rate expression for any number of transmit antennas. We also derive the closed form expression of its diversity product and use it to obtain the optimum constellation rotation in order to achieve optimum decoding performance. Other performance benefits of MDC-QOSTBC, such as low decoding complexity, good coding gain and power distribution property, flexibility in supporting any constellation and number of transmit antennas, are also presented and discussed.

We extend the throughput optimization technique of Qiu and Chawla (1999) for adaptive modulation, to combine power and rate adaptation in wireless cellular systems. We develop new combined power and rate control algorithms for wireless multimedia systems, in which the transmitted powers and rates of different media users are adapted based on the signal-to-interference power ratio. Using simulations, we show that with appropriately chosen power and rate limits, our proposed combined power and rate control algorithms can achieve a higher throughput when compared to previously proposed algorithms with power control only.

In this paper, we consider a quasi-orthogonal (QO) space-time block code (STBC) with minimum decoding complexity (MDC-QO-STBC). We formulate its algebraic structure and propose a systematic method for its construction. We show that a maximum-likelihood (ML) decoder for this MDC-QO-STBC, for any number of transmit antennas, only requires the joint detection of two real symbols. Assuming the use of a square or rectangular quadratic-amplitude modulation (QAM) or multiple phase-shift keying (MPSK) modulation for this MDC-QO-STBC, we also obtain the optimum constellation rotation angle, in order to achieve full diversity and optimum coding gain. We show that the maximum achievable code rate of these MDC-QO-STBC is 1 for three and four antennas and 3/4 for five to eight antennas. We also show that the proposed MDC-QO-STBC has several desirable properties, such as a more even power distribution among antennas and better scalability in adjusting the number of transmit antennas, compared with the coordinate interleaved orthogonal design (CIOD) and asymmetric CIOD (ACIOD) codes. For the case of an odd number of transmit antennas, MDC-QO-STBC also has better decoding performance than CIOD.

In this paper, we propose a multiuser detector (MUD) that is based on iterative interference cancellation (IIC), for a multiple-access multi-carrier M-ary frequency-shift-keying (MC-MFSK) system. We evaluate the error probability performance of this MC-MFSK system over Rayleigh fading channels, through both theoretical analyses and computer simulations. We show that the IIC-based MUD algorithm can be easily implemented and can improve system performance considerably.

By using joint modulation and a customized constellation set, we show that quasi-orthogonal space-time block coding (QO-STBC) can be used to form a new differential space-time modulation (DSTM) scheme to provide full transmit diversity with non-coherent detection. Our new scheme can provide a higher code rate than existing DSTM schemes based on orthogonal STBC. It also has a lower decoding complexity than other DSTM schemes, such as those based on group codes, because it only requires a joint detection of two complex symbols. We derive the design criteria for the customized constellation set and use them to construct a constellation set that provides a wide range of spectral efficiency with full diversity and maximum coding gain.

We first derive the generic algebraic structure of a quasi-orthogonal STBC (QO-STBC). Next we propose group-constrained linear transformation (GCLT) as a means to optimize the diversity and coding gains of a QO-STBC. Compared with the constellation rotation (CR) technique reported in the literature, we show that QO-STBC with GCLT requires only half the number of symbols for joint detection in the maximum-likelihood (ML) decoder when square or rectangular QAM constellations are used, hence it has lower decoding complexity than QO-STBC with CR. We also derive analytically the optimum GCLT parameters for QO-STBC with a generic square or rectangular QAM constellation for four and eight transmit antennas. The resulting QO-STBCs with GCLT can achieve full transmit diversity, and they have negligible performance loss compared with QO-STBCs with CR at the same code rate.

Amicable complex orthogonal design (ACOD), which is the complex version of amicable orthogonal design first reported in Wolfe (1976) is proposed and its existence is proven. Using ACOD, new orthogonal space-time block codes (O-STBC) for four and eight transmit antennas are constructed. Their maximum achievable code rates are proven to be as high as the existing O-STBC designs. In addition, the new O-STBC are shown to have better practical implementation features than the existing O-STBC, as they do not require any transmit antenna to be turned off intermittently (which is beneficial for the power amplifier design), and they do not have irrational-number coefficients inside the codeword (which simplifies the hardware implementation).

We consider the problem of channel estimation and symbol and frequency synchronization for orthogonal frequency division multiplexing (OFDM) systems in multipath fading environment with pilot data. We developed an algorithm to estimate the channel, symbol timing and frequency offset simultaneously by using a technique based on maximum-likelihood theory. Unlike existing joint algorithms which operate on frequency domain (FFT demodulated data), our algorithm works on raw received data, avoiding the need of a coarse synchronization algorithm. Thus, it is much simpler yet robust. Simulation results also show that our algorithm is more accurate and reliable than existing algorithms.

We propose a novel iterative receiver for coded multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) systems. Starting from the derivation of the optimal minimum mean squared error (MMSE) estimate of the transmitted data, we show that for BPSK and QPSK modulated signals, the soft estimate is a hyperbolic tangent function of the extrinsic information from the soft-input soft-output (SISO) decoder and the decision statistic from the detector. Compared with the conventional iterative receiver which computes the data estimate using only the decoder output, the proposed structure has significant performance gain, for example, it is 2.0 dB better for groupwise space time block coded (GSTBC) MIMO OFDM system at BER of 10/sup -6/, and 1.5 dB better for vertical Bell Lab layered space-time (VBLAST) OFDM system at BER of 10/sup -5/.

In this paper, we formulate the algebraic structure of quasiorthogonal STBC with minimum decoding complexity (MDC-QOSTBC), whose maximum likelihood (ML) decoder only requires the joint detection of two real symbols, for any numbers of transmit antennas. We also propose a systematic method to construct an MDC-QOSTBC from an Orthogonal-STBC (O-STBC). The maximum code rate of the resultant MDC-QOSTBC is shown to be the same as that of the lower-order O-STBC used to construct it. We also find the optimum constellation rotation angle for the proposed MDC-QOSTBC construction to achieve full diversity and optimum coding gain. We can show that the proposed MDC-QOSTBC has more even power distribution over the transmit antennas, better scalability in adjusting the number of transmit antennas, and more superior decoding performance than the coordinate interleaved orthogonal design (CIOD) and asymmetric CIOD

A power control scheme with outage probability specifications is derived and studied for interference-limited wireless cellular systems in Rician/Rician fading channels. We first define a performance index, so-called signal-to-interference power ratio margin (SIRM) for each desired communication link in Rician fading channels. We also explain and discuss the relationship between the outage probability of a desired link and its SIRM. A power control scheme is then proposed to balance the outage probabilities of all the desired links and we show that applying the well-known Perron-Frobenius theory can solve such a problem. Simulation results verify the effectiveness of the proposed power control scheme. As expected, our scheme is shown to cover a previous scheme for Rayleigh fading channels as a special case.

We propose a multi-user detection (MUD) based on iterative interference cancellation (UC) for a multiple-access multi-carrier M-ary frequency-shift-keying (MC-MFSK) system. The MC-MFSK is a new multiple-access technique with the desirable properties of frequency diversity and near-far resistance. The error performance of MC-MFSK system with the IIC-based MUD algorithm is studied over noise-free and Rayleigh fading channels. Both analytical and simulation results are presented, and it will be shown that the IIC-based MUD algorithm can be easily implemented and improve system performance considerably.

Quasiorthogonal space-time block code (QO-STBC) may be preferred to orthogonal STBC as it can achieve a higher code rate and to nonorthogonal STBC as it can achieve a lower decoding complexity. In this paper, the algebraic structure of QO-STBC is derived. The search of QO-STBC with rate greater than one for four transmit and at least two receive antennas is carried out. With the aids of graph theory, it is found that if the transmit diversity level is four, the maximum code rate is limited to 5/4; while if the transmit diversity level is lowered to two, the maximum code rate can go up to 4. The maximum likelihood decoding of these QO-STBCs can be computed in two separated groups, hence lead to a lower decoding complexity as compared to other nonorthogonal STBCs.

This paper presents a complete analysis on the envelope probability density functions (PDF) for a new analytical model of small-scale, local area fading experienced by narrowband receivers in wireless communications. A closed-form formula of the envelope PDF is derived in terms of a series of Laguerre polynomials by using the orthogonal expansion. As shown in numerical results, this closed-form formula can be easily computed for either analytical study or computer simulation.

A new multipath interference cancellation (MIC) scheme is proposed and studied for multicode CDMA systems. The proposed MIC scheme can be divided into two sections: firstly, successive multipath interference cancellation for obtaining initial data estimation, and secondly, parallel multipath interference cancellation to further enhance the accuracy of data estimation. BER performance of the proposed MIC scheme is evaluated and compared with the parallel interference cancellation (PIC) scheme in J Chen et al. (2002) using simulation. New simulation results show that the MIC scheme achieves improvement in BER performance than the PIC scheme in J Chen et al. (2002).

A closed-form formula for symbol-error rate (SER) of an orthogonal frequency-division multiplexing (OFDM) system with M-ary differential phase-shift keying (MDPSK) in frequency domain over Rayleigh fading channels is obtained. It is found that, by MDPSK in frequency domain, identical SERs can be achieved on all subcarriers. However, both time and frequency dispersion in the channel will introduce error floors. A comparison between OFDM-MDPSK in frequency domain and that in time domain reveals that the former system offers superior SER performance in a fast fading environment, while the latter performs better if the channel is mainly frequency selective. Moreover, the former system has lower implementation complexity.

We propose and investigate a multiple-trellis-coded unitary space-time modulation (MTC-USTM) for a Rayleigh-flat-fading channel, where k/spl ges/2 unitary space-time signal blocks are assigned to a single branch in the trellis. We have also evolved design criteria which are aimed at minimizing the pairwise error-event probability (PEP) and a partitioning scheme for the k-fold Cartesian product of the UST signal set for MTC-USTM. We demonstrate through simulations as well as theoretical analysis that MTC-USTM can produce superior bit-error probability performance over the trellis-coded USTM (single signal per branch) of the same information rate and number of states, especially in the region of high signal-to-noise ratio.

We propose an efficient two-stage DS-CDMA multiuser detector (MUD). The first stage is a reduced linear detector and the second stage is a matrix graduated nonconvexity annealed neural network (MGN-ANN). By using a first stage linear detector, the computational complexity in the second stage can be significantly reduced. We carried out extensive simulations in order to compare the error probability performance of our proposed detector with other competing multiuser detectors, which are based on conventional matched filter detector (CD), Hopfield neural network (HNN), multistage detector (MS-10) and annealed neural network (ANNMD). We also use the optimum multiuser detector (OMD) as comparison benchmark for all the MUDs. We show that the error probability performance of our proposed detector is significantly better than the other suboptimum MUD's and approaches the performance of the OMD.

In this paper, we propose multiple trellis-coded differential unitary space-time modulation (MTC-DUSTM), by assigning k/spl ges/2 signal blocks to a single trellis branch. We present the derivation of the formulae for the pairwise error-event probability (PEP) and the bit error probability (BEP) for MTC-DUSTM. Design criteria and a systematic code design have also been proposed. When compared with the trellis coded differential unitary space-time modulation where k=1 signal block is assigned to a branch, we demonstrate that MTC-DUSTM has a better BEP performance when the two schemes have the same number of states.

In this paper, we propose the trellis coded differential unitary space-time modulation (TC-DUSTM) and present the derivation of the formulae of the pairwise error-event probability (PEP) and the bit error probability (BEP) for this scheme. We found that in a multi-input multi-output (MIMO) system with M transmit and N receive antennas, apart from the diversity gain of MN, a coding gain of l/sub min/ is also obtained for the TC-DUSTM, where l/sub min/ is the length of the shortest error event. Additional coding gain comes from the product of the dissimilarities along the path of the shortest error event. These performance analyses suggest design criteria for the TC-DUSTM to obtain a best BEP performance. Through simulations we show that these analyses are accurate, especially at high signal-to-noise ratio (SNR).

Space-time block coding (STBC) has been proposed to achieve transmit diversity in wireless communication systems. Most works on STBC system assume the availability of perfect channel state information (CSI) at the receiver. Also assume that the antennas are well separated so that the received signals are uncorrelated. However, in practice, the CSI is not always known at the receiver and the received signals are sometimes correlated. In this paper, a computationally efficient channel estimation scheme is presented for STBC system under spatially correlated Rayleigh fading channel. A decorrelation algorithm is introduced to decorrelate the correlated received signals before this computationally efficient channel estimator is used. Some simulation results are presented to demonstrate the performance of proposed channel estimator in STBC system under spatially correlated Rayleigh fading channel.

Quasiorthogonal space-time block code (QO-STBC) can provide full-rate transmission and low decoding complexity. Full-level of transmit diversity can be achieved if constellation rotation technique has been used. In this paper, the design and application of noise whitening filter for the receiver of QO-STBC are discussed in detail. Such filter facilitates the use of standard equalizer or decoder that have been designed for white noise on the received QO-STBC signals, which tend to suffer from colored noise after matched filtering. This eliminates the need for dedicated decoding metric for signals with colored noise. It is found that only maximum likelihood decoding and sphere decoding can take advantage of the constellation rotation code design technique to achieve full diversity. Analytical results show that the SNR of the estimated symbol is independent to the angle of constellation rotation if symbol-by-symbol decoding schemes are used. This confirms with the simulation results where the performance of linear equalizers and decision feedback equalizers are the same for both constellation rotated and nonconstellation rotated QO-STBC.

A closed form average irreducible bit error rate (AIBER) expression for space-time block coding (STBC) in frequency selective Rayleigh fading channel is derived using the probability distribution of the general quadratic form. A 2 tx/1 rx antenna pair and BPSK modulation are assumed. We extended Alamouti's decoding algorithm, which is optimum in the flat fading channel, to the frequency selective fading scenario. Our result provides an efficient way to evaluate the effects of selective fading with various forms of power delay profiles (PDP) and pulse shaping on error rate performance of STBC. This is a significant improvement over previous simulation based approach.

In this paper, the formulae for the pairwise error-event probability (PEP) and the bit error probability (BEP) of the trellis coded unitary space-time modulation (TC-USTM) in a piecewise constant Rayleigh fading channel are derived. These analyses demonstrate that a multi-input multi-output (MIMO) system with M transmit and N receive antennas, apart from a diversity gain of MN for the space-time signal design, a coding gain of l/sub min/ can also be obtained for the TC-USTM, where l/sub min/ is the length of the shortest error event. Additional coding gain comes from the product of the dissimilarities along the path of the shortest error event. These analyses suggest the design criteria of the TC-USTM encoder to obtain an optimal BEP performance. Propositions are also presented to satisfy these criteria. We demonstrate through simulations that these analyses are accurate, especially at sufficiently high signal to noise ratio (SNR).

In this paper we present several results of our study on training sequence assisted channel estimation for multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) systems. After developing a linear matrix algebraic model for the cyclic prefix based MIMO OFDM systems, we define a generalised preamble structure which is a simple extension of the preambles used in single-input single-output (SISO) OFDM so that its good properties, such as low peak to average power ratio, can be maintained. We then derive the least squares (LS) and linear minimum mean squared error (LMMSE) channel estimation algorithms based on the proposed preamble design. In order to reduce the preamble length, we further propose a switched subcarrier preambles scheme in which the transmit antennas are divided into groups, and preambles are transmitted in alternative subset of subcarriers in each group. A LMMSE filter-based interpolation scheme and a DFT-based LS interpolation scheme are then used to obtain the channel estimates for all the subcarriers of interest. In all the proposed schemes in this paper, the filter parameters can be fixed and robust performance is obtained even when mismatched SNR and channel statistics are used in the filter parameter calculations.

We present analysis on the system performance of multiple-access multicarrier M-ary frequency-shift-keying (MC-MFSK) systems over Rician fading channels. A novel approach is used to exploit the fact that in multiple-access MC-MFSK system, the line-of-sight (LOS) carriers of the multiple signals can be combined into a single LOS carrier, and the multiple components of the multiple signals can combined with other Gaussian process. We derive a new probability density function (pdf) for the output of the envelope detector in MC-MFSK systems, which can be applied to compute the system performance and capacity. We also propose a chi-square approximation to the new pf for a large number of signals, which can be efficiently evaluated, and show that this approximation is very close to the exact results. Furthermore, we provide physical explanations on the effect of the number of multiple sub-carriers on the system performance.

In this letter, we extend the baseband impulse noise modeling over the subscriber loop conducted over the German Telephone Network to the passband scenario. We apply an in-phase (I) and quadrature (Q) resolution of the passband impulse noise and model these components as a generalized error distribution. We use simulation to verify our postulation. Closed-form symbol error rate (SER) expressions for M-ary quadrature amplitude modulation over this passband impulse noise channel are then derived. From the computed SER results, we conclude that the system performance is not sensitive to the dependency factor between the I and Q noise components.

A new multi-carrier, direct sequence code division multiple access (MC-DS-CDMA) system with multi-code is proposed. Transmitted data are first serial-to-parallel converted to D parallel data streams and a pre-coding with a rate of D/M is followed. Next, these pre-coded parallel streams are multiplexed by an inverse Walsh-Hadamard transform (IWHT) (M/spl times/M) and M super-bit streams are formed. Due to using the pre-coding, the super-bit streams will be two-level symbol streams. Then, the super-bits are once more serial-to-parallel converted to form M parallel super-bit sub-streams. Finally, all the M parallel super-bit streams are direct sequence spread spectrum (DS-SS) modulated and transmitted using M orthogonal carriers. In the receiver, after demodulation and WHT operation, the strategy of minimum mean square error (MMSE) is used for detection. Based on a correlator receiver for each carrier, its upper bound of bit error rate (BER) performance over Nakagami fading channel is analyzed. Numerical results are compared with those of a single carrier DS-CDMA system (SC-DS-CDMA). It is shown that the proposed MC-DS-CDMA scheme has much better BER performance compared to the SC-DS-CDMA system.

This paper presents a new multi-carrier DS-CDMA system with multi-code in which input data bits are first coded at a low rate (I/R) and resulting R coded binary bits are then multiplexed with a set of multi-codes to produce code division multiplexing (CDM) super-bits. After serial-to-parallel conversion, they are spread-spectrum transmitted using M orthogonal carriers. Based on a conventional matched-filter and RAKE receiver for each carrier, the performance is analyzed partially and simulated. The numerical results show that the proposed scheme has close bit error rate (BER) performance, but better anti-jamming ability compared to the normal multi-carrier DS-CDMA system.

We derive exact expressions of the symbol error rate (SER) performance of orthogonal, frequency-hopped, noncoherent M-ary frequency-shift-keying (FH/MFSK) receivers in the presence of independent multitone jamming (MTJ), as well as thermal and other wideband Gaussian noises. Both the signal tones and the multiple jamming tones are affected independently by Rician fading. Both the slow-hopping and fast-hopping systems are considered. In our derivation, we exploit the fact that the multipath components of the signal and jamming tones can be combined with the other Gaussian noises to form a single Gaussian process. Using this technique, we are able to obtain compact exact SER expressions which can be computed efficiently. From the computation results we provide physical explanations on the effects of several system parameters, such as jamming fraction, fading conditions and number of hops per symbol, on the SER performance of slow and fast FH/MFSK systems under independent MTJ.

Orthogonal frequency division multiplexing (OFDM) has established itself as a very appealing technique for high-rate data transmission over dispersive channels. The main disadvantage of OFDM is that the signal exhibits Gaussian-like time-domain waveform with high peak-to-average power ratio, which limits the efficiency of the transmitter amplifier and degrades the received signal to noise ratio. The companding technique can be employed to mitigate quantization noise and the peak-to-average-power ratio of OFDM system. The symbol error rate of the companded OFDM systems is investigated. The performances of the systems with and without companding are compared.

In this paper, the bit-error-rate (BER) performance of two signalling schemes for high rate data transmission in the 3rd generation direct-sequence, wide-band code-division-multiple-access (DS/WCDMA) systems are compared. The transmission channel is frequency selective with Rician and Rayleigh fading, and a RAKE receiver is employed to achieve diversity gain with maximal ratio combining (MRC). In the first scheme, referred to as the single code scheme, high rate data transmission is achieved by using a single spreading pseudo-noise (PN) code with varying processing gain (PG) that might be as low as 4. In the other scheme, called the multi-code scheme, the high rate data stream is split into several low rate substreams. Each substream is spread by a concatenated Walsh/PN sequence and all the substreams are then transmitted in parallel. It is shown that notable performance improvement can be achieved by using multi-code scheme in the multipath fading channel when the channel delay spread is small.

A new expression for the average bit-error rate (BER) of differential quadrature phase-shift keying in slow frequency-nonselective Nakagami fading is derived for a space-diversity receiver having a cascade arrangement of L groups of M-branch selection combiners and an L-branch maximal-ratio combiner (MRC). This allows the use of a large number of antennas, for performance improvement, and a small number of M inputs to the combiners, for low complexity. The average BER performance of the cascade receiver is investigated for different fading severity conditions characterized by the Nakagami m factor, and compared with the conventional MRC receiver. As the fading gets less severe (m increases), performance improvement over the conventional MRC scheme is only noticeable for the larger range of average signal-to-noise ratio.

In this paper, we propose and investigate a new signature sequence scheme for asynchronous direct-sequence code-division multiple-access (DS-CDMA) communication systems. The new sequence set offers good mean-square cross-correlation (MSCC) properties in both even and odd correlation functions, which are important for system capacity, acquisition and demodulation performance in DS-CDMA systems. A new asynchronous phase-coded DS-CDMA system model is proposed. An analysis of the new system model is presented, and a new aperiodic cross correlation function is derived. The correlation properties of the new sequence set are presented and compared with those of the conventional PN sequence sets. Emphasis is placed on the average rather than the worst case performance. Numerical results show that MSCC value of the new sequence set is only about half of those of the conventional PN sequence sets. This may result in an improvement of 3 dB in the multiple access interference (MAI) performance.

A new bit error rate (BER) expression for a noncoherent frequency-hopped binary-frequency-shift-keying (FH/BFSK) receiver in Rician channels, subjected to independent multitone and wideband noise interference, is derived. Unlike previously published analyzes, we combine the signal and jammer multipath components, the additive white Gaussian noise (AWGN), and other wideband interference as a single Gaussian process. This method leads to a BER expression which consists of only a single rather than a triple integral requiring numerical computation. We also present a new derivation of the probability density function (PDF) of the ratio of two Rician random variables, used in a previously published approximate BER analysis.

New expressions for the average level crossing rate (LCR) and average time fade duration (AFD) of the signal envelope in microcellular mobile radio channels, characterized by Nakagami (1960) fading and lognormal shadowing, so called Nakagami-lognormal (NLN) channels, are derived. New LCR and AFD are computed, and the effects of the Nakagami m factor and shadowing spread /spl sigma//sub s/ on these fade statistics are discussed. The generality of the NLN model is demonstrated by matching some LCR curves with previously reported measurement for the land mobile satellite channel, commonly described as the Rice-lognormal channels.

New expressions for the average level crossing rate (LCR) and time fade duration (TFD) of the signal envelope in microcellular mobile radio channels, characterized by Rician fading and lognormal shadowing, the so-called Rice-lognormal (RLN) channel, and by Nakagami fading and lognormal shadowing, so-called Nakagami-lognormal (RLN) channel, are presented. New LCR and TFD curves as functions of the Rice K factor, the Nakagami m factor and shadowing spread /spl sigma//sub s/ are presented and discussed. The LCR and TFD expressions for the RLN channel reduce to the published formulae for the Rayleigh-lognormal (Suzuki, 1982) channel as a special case, and for very large Rice K factor, approach those formulae obtained by Loo (1985).

New exact expressions involving hypergeometric functions are derived for the symbol-error rate (SER) of M-ary quadrature amplitude modulation (MQAM) for L branch diversity reception in Rayleigh fading and additive white Gaussian noise (AWGN). The diversity combining techniques considered are maximum ratio combining (MRC) and selection combining (SC). MRC with identical channels and dissimilar channels are analyzed.

An expression for the bit error rate (BER) of 16 STAR-QAM modulation with differential encoding and detection in a Rician fading channel with diversity reception is obtained. BER curves for various ratios of the line-of-sight power to the multipath power, Doppler frequencies and orders of diversity are presented. It is shown that 16 STAR-OAM outperforms 16 DPSK under the same power limited condition.

A simple tight upper bound on the BEP of 2DPSK over the AWGN channel with phase noise in the received signal is obtained. The phase is modeled as a Gaussian random process which is slowly varying compared to the bit rate so that a piecewise-constant approximation can be made. The bound is verified by computer simulations, and it provides good estimates of the error probability. It shows that for high SNR the error probability decreases as the reciprocal of the square-root of the SNR. The results are applicable in particular to heterodyne optical communications.>

We present a new closed-form bit error probability formula for narrow-band FSK with limiter-discriminator-integrator detection in Rayleigh fading channels. This formula eliminates the tedious numerical integrations required in previous publications. The computed results are verified by experiment. We provide physical explanations, on how the system parameters such as fading, IF filter bandwidths and FM carrier frequency deviations affect the error performance, not previously available in the literature.>

A single edge detector for coded mark inversion, data transmission is proposed. Its detection performance depends mainly on the false transition probability if the bandwidth of the receiver low-pass filter is reasonably large, say equal to twice the data rate. An approximate expression for the bit error rate (BER) performance of this detector operating in white Gaussian noise is derived by considering the level crossing probability of the noise process. Measurements show that this approximation is closer to the actual BER performance than results obtained from a previously established theoretical approach in some cases.>

Approximate results for the bit error probability (BEP) of binary phase shift keying (BPSK) in the presence of a noisy carrier phase reference are presented. The results show correctly the behavior or the BEP as a function of SNR. The accuracy of the approximations is verified by simulations and numerical integration of the BEP formulas. The results are compared with existing bounds.>

Error rates for a narrowband digital FM system with discriminator detection in a land mobile radio channel characterized by fading and Doppler frequency shift are obtained using a new analytical method. The fading rate is assumed to be much slower than the bit rate so that signal distortion through a narrow bandpass filter can be analyzed via a quasistatic approach. An experiment was conducted using a Rayleigh fading simulator, and the error rate measurements are shown to agree well with the calculated results, except at high SNR (signal-to-noise ratio) where an irreducible error rate was observed.>

An analytical method for calculating error rates for duobinary FM systems with discriminator detection and Viterbi sequence estimation decoding is presented. Gaussian noise and intersymbol interference are included in the channel model, and high signal-to-noise ratio and precoding of the binary input message are assumed in the analysis. The computed error rates are compared to previously published measurements. The agreement is very good, confirming that the assumptions are valid and useful.>

Band-limited digital FM systems employing discriminator detection are analyzed. The error-rate performance of binary FM with premodulation shaping and duobinary FM with the same occupied bandwidth are compared. At bandwidths above 1.1 times the bit rate, it is found that binary FM gives a lower error rate than duobinary FM. For binary FM to meet lower bandwidth requirements, frequency deviation ratios below 0.4 times the bit rate must be used. At these low deviations, binary FM does not perform as well as duobinary FM with the same bandwidth. In addition, if a more complex receiver is used which makes use of Viterbi decoding after the discriminator, the performance can be made better than binary FM even at the larger occupied bandwidths.

Of the more band-efficient constant envelope digital modulations, duobinary FM is attractive in terms of its potential error performance, as indicated by the minimum distance of the signal structure. In this paper, the error performance of narrow-band duobinary FM, with discriminator detection in an additive Gaussian noise background, is evaluated for a range of system bandwidthsBTand frequency deviation ratiosh. The effects of intersymbol interference, due to the restricted bandwidth, are taken into account in the analysis. In addition, at the narrower bandwidths,0.5 \leq BT < 1.0, the filtered background noise is correlated, and these effects are included. It is shown that with the usual duobinary detector, "clicks," which are a major contributor to errors in conventional digital FM, do not contribute significantly to errors for the low frequency deviation ratios which are commonly used. The error rate theory agrees well with results obtained experimentally. The deviation ratios that give the best error performance for a given system bandwidth are presented. In a narrower bandwidth such asBT = 1/2, where duobinary FM is proposed, duobinary FM with a frequency deviation ratio ofh = 0.7performs better than binary FM.

Two new series representations for the Rice functionIe (k, x)are presented. One of the series involves the modified Struve functions and the other involves the modified Bessel functions. These two series complement each other in their convergence speeds as functions of the values ofkandx. The truncation error bounds are derived for both series. Therefore, they can be used alternatively with high efficiency and known precision.

This paper presents a new series for evaluating the conditional bit error rate (BER) of a given transmitted data sequence in a band-limited binary differential-phase-shift-keyed system. This series involves modified Struve functions and it was found to have converged quickly. The conditional BER can then be averaged over all possible bit sequences to yield the system BER. The results obtained are close to those reported by G. J. Marshall. In our analysis the noise samples a bit period apart were assumed to be uncorrelated.

This note presents a new Fourier series expansion of the probability density function of noise at the output of a phase-shift keyed receiver. A recurrence relation for the coefficients is derived and a bound on the series truncation error is obtained. The convergence rate of the series is then estimated and discussed.

Due to the large discrepancies m the published results, the bit error rate (BER) performance of narrow-band Manchestercoded frequency-shift-keyed systems (MCFSK) with discriminator detection is reviewed and new results are presented which agree closely with measurements. It is found that a minimum BER is obtained for a peak-to-peak frequency deviation of about 1.35 times the bit rate and a receiver bandwidth of about 1.8 times the bit rate. The published results are shown to be rather optimistic. A coherent MCFSK demodulator is then shown to perform 3 dB better than discriminator detection. It is further shown that in the range of receiver bandwidths larger than the bit rate, it is sufficient to consider the intersymbol interference effects to have come only from the two bits adjacent to the bit being detected. Finally, if the phase noise components are assumed to be Gaussian distributed, the error probability formulas obtained do not predict the correct error performance.

In agreement with a recent paper by Pierce, it is shown that ternary, phase-shift keyed (PSK) signals attain the lowest bit error rate at high signal-to-noise ratio (SNR) if more than4/3binary digits can be mapped to a symbol. This point is further demonstrated by using a 3B-2T Code which maps 1.5 bits to a symbol. The spectrum and bit error rate are evaluated for the 3B-2T coded Ternary. PSK signal and compared with those of quaternary phase-shift keyed (QPSK) signal.

In this paper, the Manchester-coded frequency-shift-keyed (MCFSK) signal is shown to be representable in terms of an in-phase(I)and a quadrature(Q)carrier term. TheI-term carries the bit timing and theQ-term carries the message information. Based on this representation, a simple derivation of the power spectral density for MCFSK is presented. The effect of premodulation pulse shaping on the spectral occupancy of MCFSK is then investigated using trapezoidal, triangular and cosine-squared pulses. Formulas for the power spectral density are derived and curves of fractional in-band power are obtained which will be useful for band occupancy determination. Finally, the bit error rate (BER) performance for MCFSK in an ideal channel is derived and shown to be the same as that for coherent FSK.

A bit-synchronization method in a frequency-shift-keying (FSK) data channel has been investigated experimentally and results are presented. The bit-timing information is conveyed by a square wave, with a frequency of half the data bit rate, modulated onto the signal amplitude. It has been shown that for a system using a peak-peak-to-peak frequency deviation of 0.7 times the bit rate and a receiver bandwidth restricted to 1 times the bit rate, the minimum error rates are obtained for an amplitude modulation index of 0.1. Further, in order for the system to have a performance comparable with an ordinary FSK system with the same values of frequency deviation and receiver predetection bandwidth and with perfect timing available at the receiver, an additional signal power of 0.9 dB is needed.