Our interest is service to typical home computers not to portable devices so first we first review the state of high speed access to the Internet by wired means-the systems that wireless techniques must compete with. We then discuss the use of 3G wireless systems to provide high speed wireless access and also of extensions of 2G wireless systems as well as MMDS and LMDS. Finally we consider the use of overhead repeater platforms for this purpose. This class includes both satellites and high altitude systems operating in the stratosphere. For this class of wireless system we spend some time examining its advantages and problems.

A two-tier differential pulse-code modulation (DPCM) codec for videoconference applications is described that utilizes the so-called masking effect of motion. In this approach, an image is separated into high-frequency and low-frequency components. The low-frequency signal is processed by an adaptive spatiotemporal DPCM codec. A motion indicator derived from the low-frequency signal is used to control the corresponding block of high-frequency component pixels. The system is aimed at adaptively removing the high-frequency component of an image in those areas which contain motion. Since less information is needed to be sent, the data rate is lower. Since the missing high-frequency information occurs only in moving areas, the human eyes are less likely to notice. In this way, the data rate is lowered without unduly giving up on the subjective image quality. Simulation results have shown that the amount of visible degradation in image quality is small while the reduction in data rate is substantial.>

Simple techniques for unauthorized descrambling of a random line inversion scrambled television signal are proposed. An analysis of the complexity of the proposed descrambling process is presented and it is shown that high-quality estimates of moderately long encoding sequences can be obtained with reasonable memory and delay. An improvement to the basic scheme which substantially reduces memory requirements with only a slight increase in delay is also evaluated. Finally, it is shown that varying code lengths among a finite number of alternatives does not significantly increase descrambler complexity.

In this paper a relationship between difference triangles and third order intermodulation interference is demonstrated, and two applications of this relationship are developed. In the first application, the relationship allows us to find intermodulation-free channel assignments where such arrangements are possible. The second application concerns cases where this is not possible. In the second application, arguments connected with difference triangles are used to bound, from below, the total intermodulation interference and the maximum interference falling on any assigned channel for the optimum channel assignment. To determine the tightness of this bound, a random coding type argument is used to bound the total intermoduiation for the optimum channel assignment from above, and an upper bound on the maximum intermod is established using previously performed computer experiments.

Locking a phase-locked loop (PLL) with a sinusoid that is 100-percent AM modulated by a low-duty cycle square wave (i.e., comes in bursts) is an old and well-known technique. Despite this, certain aspects of the technique have not been systematically considered. One of the most important aspects is how to avoid sidelockan undesired mode where the loop locks to a frequency other than the sinusoid's frequency. In this paper, we explain how sidelock arises, how it can be avoided and how to provide a good lock while still avoiding sidelock. Throughout the paper our emphasis is on the situation, that is usually not considered, where the sinusoid's frequency can be considerably different than the center frequency of the voltagecontrolled oscillator (VCO).

An often-discussed scheme for providing maximal ratio combining in mobile radio is to set the complex branch gain according to the amplitude and phase of a pilot sent with the carrier. The in-phase component of the branch signal in such a system has a simple doubleexponential distribution. These and other relevant statistics are derived.

The probability of error for the transmission of one ofM-orthogonal, equally likely, equal-energy signals over the white Gaussian noise channel is known. For large values ofTit is shown that this result is asymptotically equivalent to the error probability for the same signal set transmitted over a different channel, termed here the generalized incoherent channel of orderv. Forv = 2this channel is identical to the phase incoherent channel. This problem, for larger values ofv, also arises in a number of other applications, some of which are given.

The transmission of the linear sum ofmbiphase modulated signals in a time intervalTis considered for an additive Gaussian white noise channel of bandwidthW. Previous analyses consider the case wherem \leq n \equiv 2WT. In this paper, the probability of error is derived form > n, a situation which arises when the channel bandwidth is insufficient to support the data rate. Two distinct problems are considered. In the first, termed uncoded transmission, allmsignals are independently biphase modulated. It is shown, for this case, that if the channel signal-to-noise ratio increases linearly withn, the error probability can be made to go to zero approximately exponentially innfor any value ofm/n. In the second problem, termed ceded transmission, onlyk < mof the signals are independently modulated. (The remaining(m - k)signals carry redundant information.) By using a suboptimum receiver, it is shown that for a fixed channel signal-to-noise ratio the error probability goes to0exponentially innifk/nis less than some numberC^{\ast}. For high signal-to-noise ratio,C^{\ast}is greater than1, a situation which could not occur ifm \leq n.