Upper and lower bounds on the capacity of a continuous-time additive white Gaussian noise (AWGN) channel with bilevel (/spl plusmn//spl radic/P) input signals subjected to a minimum inter-transition time (T/sub min/) constraint are derived. The channel model and input constraints reflect basic features of certain magnetic recording systems. The upper bounds are based on Duncan's relation between the average mutual information in an AWGN regime and the mean-square error (MSE) of an optimal causal estimator. Evaluation or upper-bounding the MSE of suboptimal causal estimators yields the desired upper bounds. The lower bound is found by invoking the extended "Mrs. Gerber's" lemma and asymptotic properties of the entropy of max-entropic bipolar (d, k) codes. Asymptotic results indicate that at low SNR=PT/sub min//N/sub 0/, with N/sub 0/ designating the noise one-sided power spectral density, the capacity tends to P/N/sub 0/ nats per second (nats/s), i.e., it equals the capacity in the simplest average power limited case. At high SNR, the capacity in the simplest average power limited case. At high SNR, the capacity behaves asymptotically as T/sub min//sup -1/ln(SNR/ln(SNR)) (nats/s), demonstrating the degradation relatively to T/sub avg//sup -1/ lnSNR, which is the asymptotic known behavior of the capacity with a bilevel average intertransition-time (T/sub avg/) restricted channel input. Additional lower bounds are obtained by considering specific signaling formats such as pulsewidth modulation. The effect of mild channel filtering on the lower bounds on capacity is also addressed, and novel techniques to lower-bound the capacity in this case are introduced.

The concept of iterative multiuser detection for a multiple access adder AWGN channel is coupled with turbo code used for each user. The iterative decoding exchanges the likelihood information both among the two subcodes of each user and among the users. The information exchange among users is performed through subtraction of the mean values of contributions of other users from each symbol. In addition, the variance of the equivalent noise is reestimated for each symbol and each user at each step. When the iterative procedure is applied to equal-power equal-rate case, it fails to converge at aggregate rates significantly beyond 0.5 bit/symbol, due to excessive S/(N+other users) ratio in the first iteration. We are interested in a multiple access scheme which is symmetric with respect to signaling method used by each user, but with significantly higher aggregate capacity. We overcome this problem by splitting each user into several subusers of equal rate but unequal power. Assignment of different powers to different users creates an "onion structure". This structure is used implicitly by the iterative decoding process, as the higher power users converge first, and then the lower power users converge as the residual multi-user noise reduces. This breaks the 0.5 bit/symbol barrier and enables to reach 2 bits/symbol and beyond, at the expense of a linear increase in the number of iterations required. The approach of unequal power assignment can be directly applied to multiple users with power disparities, where each user can be interpreted as an independent user. Application to a degraded broadcast channel is addressed.

The channel output entropy property introduced by A.D. Wyner and J. Ziv (ibid., vol.IT-19, p.769-762, Nov.1973) for a binary symmetric channel is extended to arbitrary memoryless symmetric channels with binary inputs and discrete or continuous outputs. This yields lower bounds on the achievable information rates of these channels under constrained binary inputs. Using the interpretation of entropy as a measure of order and randomness, the authors deduce that output sequences of memoryless symmetric channels induced by binary inputs are of a higher degree of randomness if the redundancy of the input binary sequence is spread in memory rather than in one-dimensional asymmetry. It is of interest to characterize the general class of schemes for which this interpretation holds.>