| J. Y. N. Hui and P. A. Humblet, "The capacity region of the totally asynchronous multiple-access channels," IEEE Trans. Inform. Theory, vol. IT-30, pp. 207--216, Mar. 1985. |
....II, Section IV, the effect of a random Poisson distributed number of users in adjacent cells on the interference characteristics is examined. formation rate nats per channel use (unless otherwise stated) The time discrete model in (2. 1) pre assumes a symbol and frame synchronous system [100] [101] though, as will become clear in the following, the frame synchronous assumption can be relaxed in certain cases when time sharing is not employed. Both these assumptions are obviously idealized and are invoked for the sake of simplicity and tractability. The received signal at cell site is ....
J. Y. N. Hui and P. A. Humblet, "The capacity region of the totally asynchronous multiple-access channels," IEEE Trans. Inform. Theory, vol. IT-30, pp. 207--216, Mar. 1985.
....and marginal Doppler spread have been assumed which yield the discrete time representation in [1, eq. 2. 1) The results associated with a WB approach and no ICTS do not change if frame synchronization is relaxed as in this case it readily follows that the Poltyrev Hui Humblet [26] [27] conclusion holds for the sum rate 7 . All the results are definitely affected by symbol asynchronism [11] and hence also in this respect the model examined is idealized. Note, however, that even for the time asynchronous scenario when no ICTS is employed, the full Gaussian achievable rate ....
J. Y. N. Hui and P. A. Humblet, "The capacity region of the totally asynchronous multiple-access channels," IEEE Trans. Inform. Theory, vol. IT-20, pp. 207--216, Mar. 1985.
....delay. Recent general results as in [300, and references therein] are useful in this setting. Models which account more closely for classical constraints such as the inherent lack of synchronization, presence of memory, and the associated information theoretic implications (see, for example, [135], 219] 306] 309] in the fading regime are yet to be studied. 2662 IEEE TRANSACTIONS ON INFORMATION THEORY, VOL. 44, NO. 6, OCTOBER 1998 Arbitrarily varying channels and compound multiuser channels are intimately related to efficient communication over timevarying channels [164] and, as ....
J. Y. N. Hui and P. A. Humblet, "The capacity region of the totally asynchronous multiple-access channels," IEEE Trans. Inform. Theory, vol. IT-20, pp. 207--216, Mar. 1985.
....of this work appear in [4] 8] Cover, McEliece and Posner [9] have shown that the same capacity region is valid for a mildly frameasynchronous multiple access channel in which the relative delays between users are negligible compared to the codeword length. Poltyrev [11] and Hui and Humblet [12] showed that the capacity region under complete frame asynchronism differs from that of the synchronous case only in the absence of a convex hull operation. Verd u [13] has determined the capacity region for channels with finite memory and symbolasynchronism. All of this work considers only the T ....
....(discrete memoryless) multiple access channel is defined by a transition probability W : X 1 Theta X 2 Theta Delta Delta Delta Theta XM Y, where the input alphabets X i , i = 1; M , and the output alphabet Y, are finite sets. The M users are assumed to be frameasynchronous [11] [12]; i.e. user i s transmission is delayed by an integer valued time offset d i . The delays d = d 1 ; dM ) 2 Z M are unknown to all users and can never be learned since there is no feedback. Moreover, we assume they are initially unknown to the receiver. For any set Y, let D(Y) denote ....
J. Y. N. Hui and P. A. Humblet, "The capacity region of the totally asynchronous multipleaccess channels," IEEE Trans. Inform. Theory, vol. IT-31, pp. 207--216, March 1985.
....Perhaps the two best known prior examples of nonconvexity are the capacity region of the frame asynchronous multiple access channel and the capacity region per unit cost of the multiple access channel subject to input constraints. The first example was obtained independently by Hui and Humblet [10] and Poltyrev [12] This capacity region is not generally convex because the lack of a common time reference between asynchronous encoders precludes time sharing. In sharp contrast, for certain MAVCs, time sharing with the auxiliary variable V enlarges the set of achievable rate pairs, but fails ....
J. Y. N. Hui and P. A. Humblet, "The capacity region of the totally asynchronous multiple-access channel," IEEE Trans. Inform. Theory, vol. 31, pp. 207-216, Mar. 1985.
....region C = P X 1 P X 2 Delta Delta DeltaP X M R[W ; P X 1 P X 2 Delta Delta Delta PXM ] 9) where the union is over all product input distributions. The capacity region of the asynchronous multiple access channel with arbitrarily large shifts between time indices is the closure of C [16, 17], whereas if shifts are bounded or the multiple access channel is synchronous then its capacity is the closure of the convex hull of C [18, 19] In this paper we consider only asynchronous channels. It follows that any point in the interior of the capacity region must be in R[W ; P X 1 P X 2 ....
J. Y. N. Hui and P. A. Humblet, "The capacity region of the totally asynchronous multiple-access channel," IEEE Trans. Inform. Theory, vol. IT--31, pp. 207--216, Mar. 1985.
....Perhaps the two best known prior examples of nonconvexity are the capacity region of the frame asynchronous multiple access channel and the capacity region per unit cost of the multiple access channel subject to input constraints. The first example was obtained independently by Hui and Humblet [10] and Poltyrev [12] This capacity region is not generally convex because the lack of a common time reference between asynchronous encoders precludes time sharing. In sharp contrast, for certain MAVCs, timesharing with the auxiliary variable V enlarges the set of achievable rate pairs, but fails to ....
J. Y. N. Hui and P. A. Humblet, "The capacity region of the totally asynchronous multiple-access channel," IEEE Trans. Inform. Theory, vol. 31, pp. 207--216, Mar. 1985.
....O(log n) in [19] In our view, this assumption on the clock differences is reasonable since, within a local enough area to be able to share a common resource, clocks would usually agree to within a few minutes. With the same motivation, Hui Humblet consider a somewhat different problem [11]. Another way of looking at this result is that since the expected waiting time for packets is a crucial parameter, yet another payoff is seen for building accurate clocks. Our above result is shown quite easily from the main construction we have, which is a stable protocol for the infinite case ....
J. Y. N. Hui and P. A. Humblet. The capacity region of the totally asynchronous multiple-access channel. IEEE Trans. on Information Theory, IT-31:207--216, 1985.
....(S) convex hull 8 : p X R Gamma p X ; S Delta 9 = 3) The convex hull operation results from the possibility of time sharing of channel codes. Removing the convex hull operation gives the capacity region of the symbol synchronous channel in which the users have no common clock [4, 5]. The analysis in this paper will concentrate on the sum capacity C sum (S) max R2C(S) K X k=1 R k = max p X I (X ; Y j S) 4) In [1] and [6] it is shown that the sum capacity of this channel is given by C sum (S) 1 L log det (I SW S ) 5) where W = diag(w 1 ; w 2 ; ....
J. Y. N. Hui and P. A. Humblet, "The capacity region of the totally asynchronous multiple access channel," IEEE Trans. Inform. Theory, vol. IT-31, pp. 207--216, March 1985.
....I(X S ; Y jX S c ) 8S [M ] 1) where R(S) 4 = X i2S R i ; 2) and X S 4 = X i ) i2S ; 3) and let C = p X 1 p X 2 Delta Delta Deltap X M R[W ; p X 1 p X 2 Delta Delta Delta p XM ] 4) where the union is over all product input distributions. Poltyrev [1] and Hui and Humblet [2] showed that the asynchronous 1 capacity region is the closure of C. The closure of the convex hull of C is the synchronous capacity region which was first proved by Ahlswede [3] and Liao [4] and subsequently, as a special case of a more general result, by Slepian and Wolf [5] In many instances ....
.... Gamma1 , we first define a generalized order A [M ] according to the following recursion. Let A [1] Delta = 11) and for i integer greater than 1 let A [i] be the list obtained by interleaving A [i Gamma1] with (i1; i2; Delta Delta Delta ; i(2 i Gamma1 ) For example, A [1] 11) A [2] = 21; 11; 22) and A [3] 31; 21; 32; 11; 33; 22; 34) Fig. 8 shows the equivalent layered representations. 11 A [1] 21 11 22 A [2] 31 21 32 11 33 22 34 A [3] Figure 8: The generalized orders A [3] A [2] and A [3] Two entries of A [M ] are consecutive if their first index (the user ....
[Article contains additional citation context not shown here]
J. Y. N. Hui and P. A. Humblet, "The capacity region of the totally asynchronous multipleaccess channel," IEEE Trans. Inform. Theory, vol. IT--31, pp. 207--216, Mar. 1985.
....can synchronize with the corresponding transmitter. This is a much more realistic assumption. The purpose of this section is to show that the weaker assumption is indeed sufficient. For discrete (time and alphabet) memoryless channels, the lack of (frame) synchronism has been studied in [30] and [31]. Under the assumption that the signal waveforms have disjoint support (which is a suboptimal situation for bandlimited channels) the continuous time Gaussian multipleaccess channel has been investigated in [32, 33] The effect of asynchronism on the cut off rate was studied in [34] We consider ....
J. Y. N. Hui and P. A. Humblet, "The capacity region of the totally asynchronous multipleaccess channel," IEEE Trans. Inform. Theory, vol. IT--31, pp. 207--216, Mar. 1985.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC