Carolin Huppert

On Achievable Rates in the Two User AWGN Broadcast Channel with Finite Input Alphabets

Achievable rate regions in broadcast channels are usually determined for infinite Gaussian input alphabets. Since these alphabets are not used for practical systems, we investigate the influence of finite signal constellations in the two user AWGN broadcast channel. We show that results derived with Gaussian inputs are not transferable to the finite case. For scenarios where the rate of one user should be maximized whilst a required rate for the other user is delivered we establish design rules.

Required Transmit Power Applying Tomlinson-Harashima-Precoding in Scalar and MIMO Broadcast Systems

Tomlinson-Harashima-precoding can be used as dirty-paper-coding method for transmissions over a broadcast channel. However, applying it to the individual user messages the final transmit power may vary from the sum of single user powers. This paper investigates the required transmit power for Tomlinson-Harashima-precoding in two user broadcast systems. Additionally, upper and lower bounds on the required power are derived. First our considerations are done in scalar channels. Then we extend the results to MIMO systems. In order to judge the performance of THP applied as broadcast technique we finally determine the achievable user rates.

The capacity of the semi-deterministic cognitive interference channel with a common cognitive message and approximate capacity for the Gaussian case

In this paper we study the cognitive interference channel with a common message, a variation of the classical cognitive interference channel in which the cognitive message is decoded at both receivers. We derive the capacity for the semi-deterministic model, a class of channels in which the output at the cognitive decoder is a deterministic function of the channel inputs. We also show capacity to within a constant gap and a constant factor for the Gaussian channel. Most of these results are shown using an interesting transmission scheme in which the cognitive message, decoded at both receivers, is also pre-coded against the interference experienced at the cognitive receiver. The pre-coding of the cognitive message does not allow the primary decoder to reconstruct the interfering signal; the cognitive message acts instead as a side information at the primary receiver when decoding its intended message.

Opportunistic resource allocation in mimo cognitive systems with multiple users

We propose opportunistic transmission methods exploiting spatial holes in multi antenna systems with a cognitive multiple access or broadcast channel. The proposed methods determine a precoding matrix for each cognitive transmitter as well as a post processing matrix for each cognitive receiver. For both, the pre as well as the post processing step different strategies are given. With the help of these matrices the cognitive system can be described by an equivalent non cognitive multiple user system. This equivalent systems reveals the multiplexing gain of the considered cognitive scheme and allows for determining the achievable sum rate of the cognitive users by applying multi user waterfilling.

Resource Allocation for OFDM Broadcast Channels Allowing User-Wise Coding

Several existing resource allocation algorithms for broadcast OFDM channels require separate codewords for each user in each carrier. In channels with short blocks of constant fading values codewords that strech across several carriers are desirable. Thus, we propose an algorithm with low complexity where user-wise coding is possible by avoiding irresolvable decoding dependencies. We show by means of computer simulations that this algorithm outperforms a pure scheduling approach and still exploits significant amounts of the broadcast gain.

On beamforming for overlay cognitive multiple antenna systems

We determine the maximum achievable rate in a cognitive multiple antenna system based on the overlay approach. The primary system is assumed to be oblivious to the cognitive system and thus non-adaptive. Furthermore, a primary receiver with a single antenna is considered. Thus, the covariance matrices for the primary messages are chosen according to beamforming strategies. We establish an iterative method for determining the optimum covariance matrix for the transmit strategy considered here. The optimum solution allows for the evaluation of suboptimum strategies. We investigate two suboptimum methods for determining the covariance matrix with the advantage that closed form solutions are available. The results reveal that the suboptimum solutions achieve performances close to the optimum.

The wideband slope region of BPSK and QPSK for broadcast channels in the low-power regime

We consider a multi-user scenario, where the users are operating at a low SNR level. The low SNR level entails that the energy per bit of each user is close to the minimum that allows for reliable transmission. In existing work of the last years, the wideband slope was introduced, which describes the behavior of capacity as SNR tends to zero and which thus is an adequate quality criterion for channels with low SNR. It has already been shown that in single-user channels, QPSK achieves the optimum slope reached by Gaussian alphabets, while BPSK only achieves half the slope. In this paper, we extend this result to two-user broadcast channels and show that the users achieve the optimum slope region if they use QPSK. As in the single-user case, each users slope attainable with QPSK is halved when using BPSK instead. Furthermore, we show that the suboptimality of the TDMA slope region, which was already shown for multi-user channels with Gaussian inputs, persists.

Heuristic resource allocation for sum rate optimization in MIMO-OFDM systems using eigenvalue updates

In this paper a heuristic resource allocation algorithm for sum rate maximization is presented. We consider an orthogonal frequency division multiplexing channel with multiple antennas at the base station as well as at the mobile stations. Our algorithm uses eigen-beamforming and dirty paper coding. The user interference is estimated and the eigenvalues of the affected beams are updated. Our algorithm is compared to the optimal but more complex solution. It is shown that the performance of the suggested algorithm is very close to the optimum.

On Tomlinson-Harashima precoding in 2-user degraded Gaussian broadcast channels

Tomlinson-Harashima precoding (THP) can be used in degraded Gaussian broadcast channels in order to exploit the broadcast gain with low required complexity at the receivers. In this work we derive the achievable rates for this technique with QAM constellations as signaling alphabets. We compare the performance of THP with superposition modulation.

Resource allocation with minimum rates for OFDM broadcast channels

Downlink transmissions with minimum rate requirements over orthogonal frequency division multiplexing (OFDM) channels are commonly done by means of scheduling algorithms. However, regarding it from an information theoretical point of view, this is not optimal since broadcast techniques can achieve higher rates. The drawbacks of the optimum broadcast algorithm are that the signalling overhead is larger than for scheduling and also the computational complexity is much higher. In this paper we propose an algorithm which overcomes these points. This algorithm is a hybrid algorithm combining scheduling and broadcast approaches. Thus, it combines advantages of both methods. Furthermore, we present modifications to this algorithm to avoid irresolvable decoding dependencies. We show by means of simulation results that the proposed algorithm operates close to the optimum performance and that it outperforms a pure scheduling approach.