The Wiener filter (WF) estimate of a desired signal from a vector observation is optimal in the x₀[n] is optimal in the Minimum Mean Square Error (MMSE) sense, and is employed in many applications because it is easily implemented and only relies on second order statistics. If the observation vector is of high dimensionality, though, a reduced-rank approach is needed in order to reduce computational complexity and lessen sample support requirements. In the Principal Components (PC) method, the observation signal is transformed to lower dimensionality by a matrix composed of the principal eigenvectors of the autocorrelation matrix of the data. However, the PC method is suboptimum as it only relies on the autocorrelation matrix and does not factor in the cross-correlation vector between the desired signal and the data in choosing the basis vectors for the reduced dimension subspace. Goldstein, Reed, and Scharf recently developed the Multi-Stage Nested Wiener Filter (MSNWF) in which the reduced dimesion subspace is inherently the Krylov subspace generated by the autocorrelation matrix and the cross-correlation vector. The MSNWF provides better peformance than the PC method at a substantially reduced computational cost. We here provide an overview of the MSNWF and a number of recent results related to both our conceptual understanding of the MSNWF and efficient implementations of the MSNWF. An application of the MSNWF to space-time equalization for the CDMA Forward Link for Third Generation cellular communications is presented demonstrating its efficacy.

The Multi-Stage Nested Wiener Filter (MSNWF) and the Conjugate Gradient (CG) method yield the solution of the Wiener-Hopf equation in the Krylov subspace of the covariance matrix of the observation and the crosscorrelation vector between the observation and the desired signal. Using the Lanczos algorithm instead of the Arnoldi algorithm for the MSNWF simplifies the computation of the Krylov subspace basis. In this paper, we show the relationship between the CG method and the Lanczos based MSNWF and finally derive that the MSNWF may be mathematically transformed into the CG algorithm. Consequently, we present a new implementation of the MSNWF where the weight vector and the Mean Square Error (MSE) is directly updated as each new stage is added. The new algorithm is applied to an Enhanced Data rates for GSM Evolution (EDGE) system where it linearily equalizes the received signal. Simulation results demonstrate the ability of the MSNWF to reduce the receiver complexity while maintaining the same level of system performance.

A modified constant modulus algorithm (MCMA) algorithm for adaptive channel equalization for QAM signals is proposed. The proposed algorithm minimizes a cost function that consists of amplitude- and phase-dependent terms. The phase term complements the amplitude-dependent term that is provided by the conventional constant modulus algorithm (CMA). This term should satisfy several properties that guarantee extremum values at the QAM signal levels and at maximum deviations from these levels. Further, the properties should allow symmetry and uniformity across all alphabets. It is shown that the even power co-sinusoidal functions are appropriate for phase term representations. The MCMA is compared with the CMA and the Stop-and-Go Algorithm (SGA) for blind equalization. The performance is evaluated for indoor wireless channels using both transient and steady-state behaviors of the mean square error (MSE). A channel model is developed based on spectral observations of the unlicensed U-NII frequency band, and shown to be time-varying and to exhibit frequency selective fading. While the CMA is successful in achieving good performance, it is shown that both MCMA and SGA are superior and more robust in low SNR environments. The paper also presents a performance comparison of symbol-spaced and sample-spaced equalizations for a single and two-antenna receiver. Simulation results demonstrate that using multi-antenna receiver can effectively improve adaptive channel equalizations by increasing convergence rate and decreasing steady-state mean square error.

All-digital single-user Multi-Carrier Spread Spectrum (MC-SS) transceivers are studied in this paper and compared with Direct-Sequence Spread-Spectrum (DS-SS) systems under different scenarios. Closed-form bit error rate (BER) expressions are derived for sub-optimum matched filter and optimum minimum mean-square error reception in the presence of additive white Gaussian noise (AWGN) and Narrow-Band Interference (NBI). Although digital MC-SS turns out to outperform DS-SS, the difference is not as pronounced as in their analog counterparts. Closed-form BER expressions are also obtained for MC-SS and DS-SS transmissions propagating through frequency-selective multipath fading channels, and optimality is established for MC-SS when the channel taps are independent identically distributed (i.i.d.). But when the channel taps are non-i.i.d., MC-SS is shown to exhibit near-optimal behavior at high SNR and to outperform DS-SS even though their difference diminishes as the spreading factor increases and proper spreading codes are chosen for DS-SS. Finally, BER performance is carried out for MC-SS and DS-SS in the joint presence of multipath fading and NBI. Simulated BER evaluation illustrates the advantages of MC-SS over DS-SS in the presence of multipath and corroborates its robustness against AWGN and NBI.

Circular Trellis Coded Modulation (CTCM) defines a family of (block) trellis codes which use a unique algebraic constraint, imposed on the start state, to produce a strong tail-biting property without the inefficiency of driving the encoder state to zero, using a sequence of input zeroes. Previous papers have investigated specific practical results in the case of binary and 4-ary signaling, using the elements of the Galois field GF(2^m) to label the trellis. The `4-fly' has proven to be the central feature of the trellis structure, allowing exceptional performance. The present paper generalizes the 4-fly structure, naturally related to GF(2^m), to the case of p^j-flies, which support efficient p^j-ary signaling using trellises labeled by the field GF(p^m), valid both for p equals 2 and for odd characteristic. This opens numerous practical system design options, allows greater flexibility for the transmitter design, and lays the mathematical groundwork required to support a much more systematic and general analysis of the CTCM trellis structure. This paper gives the general definitions and records the first elements of the structural analysis. Detailed knowledge of the trellis structure is key to the minimization of decoder complexity.

In this paper, we present an optimal, low-complexity method of transmitting digitally compressed imagery through interference-dominated and frequency-selective fading channels. The proposed method combines a wavelet-based image coder that employs phase scrambling and variable quality-of-service (VQoS) channel-optimized trellis-coded quantization (COTCQ), and VQoS multicarrier (MC) power allocation across the channel. Optimal image quality is achieved through a joint iterative process between the image coder and MC allocation algorithm. Using our VQoS allocation method, we show that the image quality can be improved significantly, as compared to an approach that uses a fixed QoS allocation across the channel. Further, this improvement extends to a variety of channels, including those that exhibit jamming interference and frequency-selective fading. The method is also shown to be computationally efficient.

The prohibitive -exponential in the number of users- computational complexity of the maximum likelihood (ML) multiuser detector for direct-sequence code-division-multiple-access (DS/CDMA) communications has fueled an extensive research effort for the development of low complexity multiuser detection alternatives. Notable examples are the zero-forcing (``decorrelating'') and minimum-mean-square-error (MMSE) linear filter receivers. In this paper, we show that we can efficiently and effectively approach the error rate performance of the optimum multiuser detector as follows. We utilize a multiuser zero-forcing or MMSE filter as a pre-processor whose output magnitude provides a reliability measure for each user bit decision. An ordered reliability-based error search sequence of length linear to the number of users returns the most likely user bit vector among all visited options. Numerical and simulation studies for moderately loaded systems that permit the exact implementation of the optimum detector indicate that the error rate performance of the optimum and the proposed detector are nearly indistinguishable. Similar studies for higher user loads (that prohibit comparisons with the optimum detector) demonstrate error rate performance gains of orders of magnitude in comparison with straight decorrelating or MMSE multiuser detection.

This paper discusses an experimental neural network based smart antenna capable of performing direction finding and the necessary beamforming. The Radial Basis Function Neural Network (RBFNN) algorithm is used for both tasks and for multiple signals. The algorithm operates in two stages. The field of view of the antenna array is divided into spatial sectors, then each network is trained in the first stage to detect signals emanating from sources in that sector. According to the outputs of the first stage, one or more networks of the second stage can be activated so as to estimate the exact location of the sources. No a priori knowledge is required about the number of sources, and the networks can be designed to arbitrary angular resolution. Some experimental results are shown and compared with other algorithms, such as, the Fourier Transform and the MUSIC algorithm. The comparisons show the superior performance of the RBFNN and its ability to overcome many limitations of the conventional and other superresolution techniques, specifically by reducing the computational complexity and the ability to deal with a large number of sources.

In this paper, we study the possibility of applying the smart antenna technology to repeater design. The difficulties of using an array antenna in a repeater have to do with the coherence of the desired signal and the interfering signal as well as the sensitivity to the array calibration. To cope with the difficulties, we propose a robust beamforming algorithm which uses a interference projection matrix in Generalized Sidelobe Canceller (GSC). The matrix is designed to prevent the desired signal cancellation caused by array error. To verify the algorithm, we use the data generated by computer simulation and the field data measured through a custom-built W-CDMA test-bed.

This paper will discuss the on-going research at Army Research Laboratory (ARL) supporting long distance communication for networked sensor arrays. As the Army continues to move forward in the development of remote, distributed, unattended, intelligence sensors, the need for the resultant information to reach the command and control center increases. In the case of sensor array networks the command and control center provides configuration information for deployment and battlefield visualization tools to interpret and integrate the data received. These sensor array networks may be separated by long distances from the command and control center. To allow the communication to take place, an experimental long distance communication link was developed and demonstrated. The link consists of a variable node ad-hoc wireless network and controlling software. This network provides the pre-processed data for situational awareness to the remote command control center over a low bandwidth link. This link must be robust in that it must tolerate node failures and adapt. Although mobility is not a large issue, the network must tolerate a variable network topology and must therefore be ad-hoc. In this paper we will discuss the design requirements for this communication link along with the protocols developed to overcome limitations imposed by the noisy, failure prone battlefield wireless environment. We will also discuss the software and hardware architecture design and implementation for the generalized long distance sensor array communication link. Currently, operational components of this link function over an 80 mile distance with very high noise levels, node failures, intermittent connectivity, and severe bandwidth constraints.

We present a pilot symbol-assisted beamforming algorithm and a simulation tool of smart antennas for Wideband Code Division Multiple Access (WCDMA) in reverse link. In the 3GPP WCDMA system smart antenna technology has more room to play with than in the second generation wireless mobile systems such as IS-95 because the pilot symbol in Dedicated Physical Control Channel (DPCCH) can be utilized. First we show a smart antenna structure and adaptation algorithms, and then we explain a low-level smart antenna implementation using Simulink and MATLAB. In the design of our smart antenna system we pay special attention for the easiness of the interface to the baseband modem; Our ultimate goal is to implement a baseband smart antenna chip sets that can easily be added to to-be-existed baseband WCDMA modem units.

Need for exchanging a wide range of information among many users in a timely manner, has resulted in a rapid expansion of frequency utilization, in addition to bandwidth expansion. High resolution video, infrared sensors, and various types of radars now are expected to be readily exchanged among users, and these users can be located on the surface, in the air and these users can be expected to be not only mobile, but highly mobile. To exchange this high volume information reliably among these many users requires a high data rate and the use of directional antennas. The use of directional antennas would also tend to reduce the possibility of interference among the users and also reduce the RF terminal power consumption requirements. This paper describes a multi-functional phased array antenna design, which can operate at either Ku-Band or Ka-Band without the need for switching or reconfiguration; and can be readily switched to provide information in an timely manner to the various users.

The maximin algorithm adapts the weights of an antenna array to provide simultaneous interference suppression and beamforming in a frequency-hopping communication system. The algorithm is derived and a digital implementation is presented. Simulation experiments illustrate the ability of the maximin algorithm to rapidly form deep nulls in the directions of interference sources.

In direct-sequence code-division-multiple-access (DS-CDMA) systems, the pre-detection signal-to-interference-plus-noise ratio (SINR) at the output of the single-user minimum-mean-square-error (MMSE) filter is a function of the specific user spreading code (signature). In this paper, we consider the adaptive optimization of the user signature assignment such that the output SINR of the MMSE filter is maximized under a transmitter power constraint. In the context of binary signatures, the complexity of the signature optimization procedure is exponential in the processing gain. A low-cost suboptimum adaptive binary signature assignment algorithm is derived based on conditional optimization principles. We use this algorithm to design an efficient system-wide multiuser adaptive signature set assignment scheme. The performance of the proposed scheme is evaluated under asynchronous multipath fading DS-CDMA channel models and is compared to the performance of systems with arbitrarily chosen signature sets.

Subspace projection techniques are effective in excising FM interferers in GPS receivers. The FM jammers are instantaneously narrowband and have clear time-frequency (t- f) signatures that are distinct from the GPA C/A spreading codes. The instantaneous frequency (IF) estimate, provided by the time-frequency distribution, or any other estimator, is used to form the jammer subspace. The interference rejection is implemented by projecting the received data onto the orthogonal subspace of the jammer subspace. Errors in IF estimations, however, perturb the projection matrix and allow some of the jammer power to escape the projection operation. This in turn leads to degraded receiver performance and lower SINR. This paper derives the signal- to-interference-and-noise ratio (SINR) of the GPS receivers in the presence of Gaussian IF estimation errors. It is shown that IF estimation errors can substantially impede interference rejection in GPS using subspace projection techniques.

Automatic classification of the digital modulation type of a signal is a branch of non-cooperative communication techniques, which has found many applications in electronics warfare, surveillance and threat analysis. We propose in this paper a classifier based on the likelihood method to distinguish BPSK signal and unbalanced QPSK signal embedded in additive white Gaussian noise (AWGN). Unbalanced QPSK signal occurs when the in-phase and quadrature components of a QPSK signal have different power. The proposed classifier first calculates the log-likelihood functions of BPSK and UQPSK signals using AWGN model and IID data sequence. The ratio of the two log-likelihood functions is dependent on the unbalanced factor. When the unbalanced factor is not known to a receiver, it will be replaced by the maximum likelihood estimation, yielding the generalized likelihood ratio test to classify the two modulation types. The proposed classifier performs better than the classifier without taking into account the unbalanced factor when the SNR and the unbalanced factor are high. At low SNR, the performance of the proposed classifier is comparable to the one without considering the unbalanced factor.

When the characteristics of a digitally modulated signal are not known, the estimation of its modulation parameters is crucial for many emitter interception applications. This paper presents the Modified Cramer-Rao Lower Bound (M-CRLB) on the modulation parameters of a QAM signal embedded in AWGN. The modulation parameters considered are the carrier frequency, carrier phase, signal amplitude and symbol time. The M-CRLB results indicate that the carrier frequency has the highest estimation accuracy, followed by the symbol time, the signal amplitude and the carrier phase. The M-CRLBs of the four parameters are affected by the modulation level and the pulse shaping function. When the modulation level increases, the estimation accuracy decreases. When a raised cosine pulse shaping filter is used, the M-CRLBs on the four parameters all increase as the raised cosine roll-off factor increases. For lowpass pulse shaping filter, the M-CRLB on the symbol time decreases as the filter bandwidth increases. However, the M-CRLBs on the other three parameters increase.

Next-generation wireless networks have been designed to transport integrated multimedia services based on a cellular extension of a packet-switched architecture using variants of the Internet protocol. Each call, arriving to or active within the network, carries demand for one or more services in parallel, where each service type has a guaranteed quality of service. Admission of new calls to the wireless IP network (WIN) from the gateway of a wired network or from a mobile subscriber (MS) is allowed by call admission control procedures. MS roaming among the nodes of the WIN is controlled by handoff procedures between base stations (BSs), or BS controllers, and the MSs.

Discrete-time linear systems that possess scale-invariance properties even in the presence of continuous dilation were proposed by Zhao and Rao. The paper presents results of subsequent investigation characterizing self-similarity properties of discrete-time signals synthesized by these systems. It is shown that white noise inputs to these linear scale invariant systems produce self-similar outputs regardless of the marginal distribution of the noise. We investigate this with different types of inputs and in most instances the outputs are fractional Gaussian and self-similar. This is confirmed by generating the fractional Gaussian noise from the fractional Brownian motion and comparing its characteristics with the system output. For heavy tailed input distributions, the output is also heavy-tailed and self-similar. It is also shown that it is possible to synthesize statistically self-similar signals whose self-similarity parameters are consistent with those observed in network traffic.

Nanosatellites operating singly or in clusters are anticipated for future space science missions. To implement this new communications paradigm, we are approaching cluster communications by first developing an infrared (IR) intra- craft wireless bus capability, following initially the MIL- STD-1553B protocol. Benefits of an IR wireless bus are low mass, size, power, and cost, simplicity of implementation, ease of use, minimum EMI, and efficient and reliable data transfer. Our goals are to maximize the reliable link margin in order to afford greater flexibility in receiver placement, which will ease technology insertion. We have developed a concept demonstration using a high-speed visible-band silicon PIN photodiode and a high-efficiency visible LED operating at a data rate up to 4 Mb/sec. In designing an internal IR wireless bus, we have characterized various candidate materials, emitters, and geometries, assuming a single reflection. Thus, we have measured the bidirectional reflectance distribution function (BRDF) for five different materials characteristic of typical spacecraft structures, which range from nearly Lambertian to highly specular. We have fit our data to empirical BRDF functions and modeled the detected irradiance anywhere in the plane of incidence for a divergent emitter. We have also determined the angular limits on the link geometry to remain within the required bit error rate by determining the received signal-to-noise ratio for minimum values of irradiance received at the detector.

For a number of years high speed space to ground data links have been a topic of interest. Improvements in mass storage devices, laser transmitters, EO modulators, and laser receivers have made implementation of a high speed optical down link physically and financially practical. This paper highlights a concept for a many gigabit per second optical data down link using a laser transmitter as the core of a down link device. The laser communication system is scalable using wavelength division multiplexing (WDM). Impact of probability of cloud free field of view (PCFFOV) is addressed on data storage needs. The trade space of the baseline concept between number of channels, data transmit size, number of ground stations, laser power, and climatology is presented. The resulting concept uses COTS equipment and consists of 16 channels multiplexed together producing an effective data transmit rate of 32 Gbps. This optical down link will transmit in 1 minute what an X-band transmitter needs more than 11/2 hours to down link.

Recall the adage `a chain is as strong as its weakest link'- -a phrase that could serve as the official mantra of computer security. Operating Systems are difficult system to administer because it is not only complex and cantankerous but also hard to secure. They are enormous configurability, the fact that vendors don't ship secure systems, and that it requires significant amounts of time, resources, and expertise to safeguard a host are only some of the reasons that so many systems are insecure any type of network commercial or tactical. To compound the problem, like all modern operating systems it not only becomes less secure as time goes on (simply due to usage), but with the rapidly changing security field, it also requires considerably effort to stay abreast of the latest information. Army Research Labs is trying to address the security of the operating system in a tactical wireless environment. Through the use of public domain and/or commercial mans. ARL is evaluating monitoring, deployment, and auditing techniques to the wire commercial domain. By evaluating the wire domain ARL will determine what works and how they work in the tactical area. There are numerous ways to protect the wire/wireless network via public domain or commercial software.

The International Maritime Satellite (Inmarsat) communication service can be used to provide a wide area network (WAN) connection for a local area network (LAN) to the home LAN from almost anywhere on the earth. The communication link of the WAN connection is achieved through the satellite network of Inmarsat and a land-earth digital telephone network, such as Integrated Services Digital Network (ISDN). We determined the throughput data rate of data transmission in the TCP/IP environment over the Inmarsat link and compared it with the throughput data rate over a null-modem link (assumed to be error free) for the WAN connection. We also used the protocol boosters that provide forward error correction for the TCP/IP environment to test the improvement of data transmission over the satellite link. Communication over satellite tends to be noisy and delay prone. However, our analysis shows that the Inmarsat satellite-ISDN link is so reliable that the data transmission rate is nearly an error-free link in the TCP/IP network, and the application of forward error correction on the network causes a reduction in throughput data rate.

Wireless networks have several special characteristics. First, the bandwidth is extremely limited and hence a careful sharing is needed in order to fully utilize this scarce resource. Second, due to the unreliable nature of wireless channels, quality of service becomes very important to ensure good communications among mobile users. Third, the network structure changes frequently as users may move around from one location to another. This dynamic change in network structure causes some problems in network resource management. The key idea of this paper is to manage bandwidth based on predicted future traffics in the system. Conventional approaches to network management are reactionary in nature. That is, the decision on bandwidth allocation is based on past measurement of traffic and hence is slow and inefficient in response. In contrast, our traffic prediction based bandwidth management tool is a look-ahead approach and will be more efficient and quick in bandwidth management. Moreover, much less collisions will occur among users and less bandwidth will be wasted. Here we summarize some preliminary results on traffic prediction by using neural networks. Future work will include the integration of traffic prediction tool with network bandwidth management.