BER expressions for differentially detected π/4 DQPSK modulation

Closed form bit-error rate (BER) expressions for differentially detected π/4-shifted differentially encoded quadrature phase-shift keying (QPSK) modulation (π/4 DQPSK) are derived for both additive white Gaussian noise (AWGN) and Rayleigh-fading channels. The derivations are carried out in an exact and most general manner in that in-phase (I) and quadrature (Q) channel bit-error probabilities P_cl and P_cQ are separately obtained in terms of the same-quadrature and cross-quadrature noise Correlation functions, including a measure of noise nonstationarity. We then specialize the general expressions for uncorrelated noise and equal noise powers in successive symbol periods to obtain a useful bit-error probability expression for the AWGN channel in the form P_e≈Q(√(1.1716·E_b/N₀)) where Q(·) is the Gaussian distribution Q-function and E_b /N₀ is the bit energy-to-noise density ratio. Exact BER expressions for the Rayleigh-fading channel that involve the noise parameters are also given and are extended to the case of L-fold diversity combining     

Bit error rate performance of π/4 DQPSK for Nakagami-lognormalchannels

A new expression is derived for the bit error rate (BER) of the π/4 DQPSK in a channel modelled by the Nakagami-lognormal (NLN) fading process. The NLN model includes the well-known Suzuki model as a special case. New BER curves with varying degrees of fading severity and shadowing spread are presented and discussed     

Bit error rate performance of π/4 DQPSK for Nakagami-lognormalchannels

A convergent infinite series has been presented for the bit error performance of π/4 DQPSK for Nakagami-lognormal channels if the fading figure m is an integer. The authors show that this infinite series can be replaced by a finite series. A convergent series for the BER is also derived when the fading figure, m, is real     

π/4-QPSK modems for satellite sound/data broadcast systems

The use of π/4 quadrature phase-shift keying(QPSK) modems for satellite sound broadcast systems (SSBS) broadcasting to mobile or portable receivers is proposed. Three different differential detectors (including the FM-discriminator followed by integrate-sample-and-dump filter) and a novel coherent detector are discussed. The degradation caused by the frequency offset between the local oscillator and the unmodulated carrier in the baseband differential detector is studied. The performance of both coherently and differentially detected π/4-QPSK in a Gaussian channel is also studied. It is shown that with a frequency offset of more than 3% of the symbol rate, the performance degradation is more than 1 dB at 10^-4. The out-of-band power of the nonlinearly amplified bandlimited π/4-QPSK signals is reduced from -13 dB to -37 dB if a 2-dB output back-off amplifier is used instead of a hardlimiter. The performance of the π/4-QPSK is equivalent to that of QPSK, although the π/4-QPSK has the advantage of less spectrum restoration after nonlinear amplification. The coherent demodulator and differential decoder avoid the three-level detection and achieve the same bit-error-rate (BER) performance as DEQPSK with a simple circuit     

Diversity π/4-DQPSK on microcellular interference channels

The average bit-error rate (BER) performance of π/4-shifted differential quadrature phase shift keying (π/4-DQPSK) in the presence of multiple independent cochannel interferers in frequency-nonselective fading environments is analyzed. Nyquist shaped pulses are used and post detection diversity combining is employed at the receiver. The use of both L-branch post detection equal gain combining (EGC) and L-branch post detection selection combining (SC) are considered. Two approaches for representing the cochannel interference are investigated. These are a Gaussian interference model and a synchronous interference model. The accuracies of both of these models are assessed by comparing their BER performances with precise BER results. The precise BER results are obtained by using a combination of analysis and simulation. Attention is given, in particular, to the outdoor microcellular environment. The results of this paper indicate that the synchronous interference model may overestimate the effects of interference while the Gaussian interference model may underestimate the effects of interference     

π/4-FQPSK: an efficiency improved, standardized π/4-DQPSKcompatible modulation/nonlinearly amplified RF wireless solution

A new modulation solution suitable for digital wireless communications and broadcasting applications, named Feher's π/4-FQPSK, is presented. This solution is an implementation of quadrature crosscorrelated Feher et. al. (1986, 1987) inventions. It has the combined advantages in spectral efficiency and power efficiency, as well as compatibility with IS-54 and IS-136 standardized π/4-DQPSK. Many researchers demonstrated that conventional π/4-QPSK has a significant spectral spreading in nonlinearly amplified systems. To reduce spreading, we introduce an in-phase (I) and quadrature (Q) or I/Q cross correlated baseband processor into the conventional π/4-DQPSK architecture, which reduces the envelope fluctuation significantly. While conventional π/4-DQPSK requires linearized, relatively expensive and low power efficient RF stage, Feher's π/4-FQPSK can be operated in nonlinearly amplified, e.g. “C-class” power efficient systems. That is desirable for applications requiring high transmit efficiency, such as cellular/wireless/personal communications and broadcasting systems. The architecture of the π/4-FQPSK modem is described and its performance for different applications is evaluated through computer simulations and compared with some other modulation schemes including quadrature crosscorrelated GMSK. The compatibility of the π/4-FQPSK with the US digital cellular standard (IS-54 and IS-136) is also studied. The results demonstrate that π/4-FQPSK is both power and spectrally efficient for many applications     

Baseband pulse shaping for π/4 FQPSK in nonlinearly amplifiedmobile channels

We apply baseband pulse shaping techniques for π/4 QPSK in order to reduce the spectral regeneration of the bandlimited carrier after nonlinear amplification. These Feher's patented techniques, namely, π/4 FQPSK (superposed QPSK) and π/4 CTPSK (controlled transition PSK), may also be noncoherently demodulated. Application of these techniques is in fast fading, power efficient channels, typical of the mobile radio environment. Computer simulation and experimental studies demonstrate that with these baseband waveshaping techniques, carrier envelope fluctuations are significantly reduced, and the out-of-band power after nonlinear amplification is suppressed by up to 20 dB compared to π/4 QPSK. In frequency noninterleaved land or satellite mobile radio systems operating in a nonlinear, fading and ACI (adjacent channel interference) environment, these techniques may achieve 20%-50% higher spectral efficiency compared to π/4 QPSK. In mobile cellular systems using π/4 QPSK, such as the new North American and the Japanese digital cellular systems, the application of these baseband pulse shapes may allow more convenient and less costly amplifier linearization     

用于数字闭环加速度计的差动电容检测电路

数字闭环加速度计是加速度计领域最新的研究课题,它具有直接输出数字信号、累积误差小,采样时间短等优点,因此对差动电容检测电路的要求也很高。本文为满足数字闭环加速度计的需要设计的单载波调制解调型差动电容检测电路,分析了方案优缺点,给出了系统和检测电路的原理框图,以及各个功能模块的电路图和传递函数,最后通过实验验证,得到该电路满足灵敏度高,分辨性能好的要求,并且该电路在其他微弱信号检测领域也具有一定的实用意义。     

Low-complexity equalization for π/4 DQPSK signals based on themethod of projection onto convex sets

An equalizer adaptation technique for compensation of degradations caused by multipath Rayleigh fading channels to π/4 differential quadrature phase shift keying (DQPSK)-modulated signals is presented. The technique is applied to linear and nonlinear transversal-filter-type equalizers. It is based on the method of projection onto convex sets (POCS), realized in a particular form of the iterative least mean squares (LMS) procedure. The convergence speed of the proposed equalizer coefficient adaptation technique and its computational complexity depend on the newly introduced look-back parameter. Both can be tailored to the characteristics of the channel. For achieving convergence speeds comparable to speeds of recursive least squares (RLS) techniques, the computational load of the presented equalization is of the order of the load required of RLS techniques. However, its algorithmic implementation is notably simpler and its code and storage size requirements are smaller. The technique is numerically stable, and it is suitable for low-power implementations in digital signal processors or custom very large-scale integration (VLSI) circuits. Performed simulations verify good performance of the technique in various channel conditions for 900-MHz multipath fading radio channels     

Performance of π/4 SQAM in a hard-limited channel in thepresence of AWGN

A strategy that reduces the spectral spreading when an ideal hard-limiter is used as a first approximation to a fully saturated power amplifier is presented. This strategy combines superposed quadrature amplitude modulation (SQAM) filtering with the π/4-shift quadrature phase-shift keying (QPSK) digital transmission format adopted for the first generation of US digital cellular systems. Simulation results showed that this π/4 SQAM filtering strategy increased capacity by 35% in comparison to hard-limited π/4 QPSK. Using computer-aided design, a receive filter that would limit the degradation of E _b/N₀ to less than 1.4 dB at a bit error rate of P_e=10^-4 was selected     

An improved π/4-QPSK with nonredundant error correction forsatellite mobile broadcasting

An improved π/4-quadrature phase-shift keying (QPSK) receiver that incorporates a simple nonredundant error correction (NEC) structure is proposed for satellite and land-mobile digital broadcasting. The bit-error rate (BER) performance of the π/4-QPSK with NEC is analyzed and evaluated in a fast Rician fading and additive white Gaussian noise. (AWGN) environment using computer simulation. It is demonstrated that with simple electronics the performance of a noncoherently detected π/4-QPSK signal in both AWGN and fast Rician fading can be improved. When the K-factor (a ratio of average power of multipath signal to direct path power) of the Rician channel decreases, the improvement increases. An improvement of 1.2 dB could be obtained at a BER of 10^-4 in the AWGN channel. This performance gain is achieved without requiring any signal redundancy and additional bandwidth. Three types of noncoherent detection schemes of π/4QPSK with NEC structure, such as intermediate frequency band differential detection, baseband differential detection, and FM discriminator, are discussed. It is concluded that the π/4-QPSK with NEC is an attractive scheme for power-limited satellite land-mobile broadcasting systems     

Bit error simulation for π/4 DQPSK mobile radio communicationsusing two-ray and measurement-based impulse response models

An accurate software/hardware bit-by-bit error simulator for mobile radio communications is described. Simulation results in indoor and outdoor channels are compared with theoretical results. Bit error rate (BER) results in simulated frequency-selective fading channels generated by several channel models such as two-ray, constant amplitude, and simulated indoor radio channel impulse models (SIRCIMs) are presented. It is shown that BER is not only dependent on the RMS delay spread, but also on the distribution of temporal and spatial multipath components in local areas. An important result is that a two-ray Rayleigh fading model is a poor fit for indoor wireless channels and, if used, can underestimate the BER by orders of magnitude. A real-time bit error simulation of video transmission using the bit-by-bit error simulator is described. The simulator, called BERSIM, is shown to be a useful tool for evaluating emerging data transmission products for digital mobile communications     

Bit error rate performance of π/4-DQPSK in a frequency-selectivefast Rayleigh fading channel

The bit error rate (BER) performance of π/4-differential quadrature phase shift keying (DQPSK) modems in cellular mobile communication systems is derived and analyzed. The system is modeled as a frequency-selective fast Rayleigh fading channel corrupted by additive white Gaussian noise (AWGN) and co-channel interference (CCI). The probability density function of the phase difference between two consecutive symbols of M-ary differential phase shift keying (DPSK) signals is first derived. In M-ary DPSK systems, the information is completely contained in this phase difference. For π/4-DQPSK, the BER is derived in a closed form and calculated directly. Numerical results show that for the 24 kBd (48 kb/s) π/4-DQPSK operated at a carrier frequency of 850 MHz and C/I<20 dB, the BER will be dominated by CCI if the vehicular speed is below 100 mi/h. In this derivation, frequency-selective fading is modeled by two independent Rayleigh signal paths. Only one co-channel is assumed in this derivation. The results obtained are also shown to be valid for discriminator detection of M-ary DPSK signals     

Postdetection selection diversity reception with correlated,unequal average power Rayleigh fading signals for π/4-shift QDPSKmobile radio

The diversity gain degradations due to fading correlation and unequal average power are investigated for practical, two-branch postdetection selection diversity reception. The average bit error rate (BER) of π/4-shift QDPSK is theoretically analyzed taking into account additive white Gaussian noise (AWGN), cochannel interference, and multipath channel delay spread; exact diversity gain degradations are calculated. Simple and useful approximate expressions for the gain degradations are also presented     

Bidirectional decoding of convolutional codes for indoor cellularradio

Bidirectional suboptimal breadth-first decoding of convolutional codes is an attractive technique for slowly varying and quasi-static fading channels as it restricts the extent of decoding errors due to correct path loss to very heavy noise or interference regions. This paper compares the performance of such a decoding scheme to the Viterbi algorithm over wideband TDMA indoor radio links where equalization and space diversity are also used to combat dispersive fading and cochannel interference. On the basis of equal computational complexity and equal decoding delay, suboptimal, breadth-first, bidirectional decoding of a long constraint length convolutional code is shown to be superior to Viterbi decoding of a shorter constraint length code. Furthermore, this advantage increases as the outage criterion (in terms of bit error rate) becomes more stringent which makes bidirectional decoding particularly attractive for data applications and makes channel coding a more attractive alternative to increasing the space diversity order at the receiver     

π-heterostructure field effect transistors for VLSI applications

A quantum-well heterostructure field effect transistor (HFET) with a p⁺ gate and fairly heavily doped p-type buffer is described. This device is called a π-HFET, meaning p-type insulated gate HFET. The effective barrier height in a π-HFET is considerably larger than in other compound semiconductor FETs, and the gate current at maximum gate voltage swing can be made negligible, even at room temperature. Based on the tradeoff between the noise margin and speed, 0.7 and 1.5 V are proposed as the minimum power supply voltages for the direct-coupled FET logic at 77 and 300 K, respectively. Calculations demonstrate that this π-HFET technology can meet all requirements for VLSI applications, and that high electron velocity and mobility in π-HFETs lead to an increase in speed of output drivers by a factor of four at 77 K and a factor of 10 at 300 K (compared to Si NMOS which is faster than CMOS). The full advantages of π-HFET technology can be realized only on a submicrometer scale, where source and drain series resistances play a dominant role in determining the noise margin     

Eigenvalues for ridged and other waveguides containing corners ofangle 3π/2 or 2π by the finite element method

Superelements developed to enable the finite-element method to be used for computing eigenvalues of the Laplacian over domains containing reentrant corners of angle 3π/2 or 2π are discussed. The superelements embody mesh refinement and include basis functions which emulate the singular behavior of the solution at the corner. Being compatible with linear or bilinear elements, the superelements are easily incorporated into standard finite element programs. The method which has been used to compute transverse electric (TE) and transverse magnetic (TM) mode eigenvalues for ridges and other waveguides is described. The results agree well with those obtained using various other methods     

Nonredundant error correction analysis and evaluation ofdifferentially detected π/4-shift DQPSK systems in a combined CCI andAWGN environment

The application of the nonredundant error correction (NEC) technique to the North American and Japanese digital cellular modulation standard, π/4-shift differential quadrature phase shift keying (DQPSK), in a combined additive white Gaussian noise (AWGN) and cochannel interference (CCI) environment is proposed, analyzed, and theoretically evaluated. The performance for NEC receivers with single, double, and triple error correction capability is theoretically analyzed and evaluated. For the CCI, the general model, which includes M statistical independent interferers also employing the π/4-shift DQPSK modulation format, is adopted. The theoretical symbol error probability versus carrier-to-noise ratio have been obtained with M and the carrier-to-interference ratio (C/I) as parameters. The results indicate significant performance improvements over conventional differentially detected systems. Some of the results have been verified by computer simulation. The gains offered by the NEC receivers increase as C/I decreases and/or M increases. Significant error floor reductions have been observed     

Performance of the π/4-DQPSK, GMSK, and QAM modulation schemesin mobile radio with multipath fading and nonlinearities

As mobile communications have become so indispensable, every possible effort should be spent to achieve the optimum operating conditions. In addition to the normal impairments associated with wireless communications, in general, the mobile channel suffers from particular limitations that confine the performance of a mobile radio system. Among those impairments are the bandwidth limitation, interference, and multipath fading. With the strong motivation toward portable radio and personal communication systems, power limitation has manifest itself in the picture, and, consequently, nonlinear operation of the amplifiers involved (hence, the channel) will have to be dealt with. Constant envelope modulation schemes have been used in digital mobile radio systems recently installed. The Gaussian minimum shift keying (GMSK) is employed in the Global System for Mobile (GSM) communications installed in Europe, while in the North American IS-54 system, the modulation scheme used is the π/4-DQPSK. As the quest for higher data rates has kept on increasing, multilevel modulation schemes have been proposed with their performance over nonlinear channels having been overlooked. The paper provides a comparative study, based on simulation, and tests the performance of various modulation schemes operating over a wide variety of mobile radio channel conditions. The effective throughput of all systems is also considered     

Generation and transmission of FM and π/4 DQPSK signals atmicrowave frequencies using harmonic generation and optoelectronicmixing in Mach-Zehnder modulators

A novel method of using the harmonic generation and optoelectronic mixing properties of Mach-Zehnder modulators to generate modulated subcarrier signals at high-order harmonics of the input signals is presented. The method permits the simultaneous transmission over optical fiber of a modulated and an unmodulated signal, both at high-order harmonic frequencies of the input signals, for the purpose of transmitting both a local oscillator tone and the modulated signal required at a base station for microcellular applications. We present the theory of operation and demonstrate the validity of the concept with a narrow band single-tone FM experiment as well as a 20-Mb/s π/4 DQPSK experiment