Multi-Antenna Multi-Layer Interleave-Division Multiple Access
Zusammenfassung der Projektergebnisse
By means of multiple access schemes, several users share a communication channel. Different multiple access schemes have been proposed in the past. In this project, focus is on int er leave-division multiple access (IDMA) proposed by Li Ping. IDMA is a special variant of code-division multiple access (CDMA). Currently, the most popular CDMA technique is direct-sequence CDMA (DS-CDMA). Different from DS-CDMA, which separates users by user-specific spreading sequence, IDMA separates users by unique interleaver patterns. In IDMA, the info bits are encoded (i.e., spread) before interleaving. IDMA outperforms DS-CDMA, because the encoded data streams are more random. As a consequence, IDMA is able to deliver a good power and bandwidth efficiency. Orthogonal frequency-division multiplexing (OFDM) and multiple-input multiple-output (MIMO) have become two key techniques in modern transmission systems. Spatial multiplexing enhances the data rate, whereas space-time coding provides robustness with respect to transmission errors. In IDMA systems, each antenna can transmit multiple streams. Thus, it has the potential to support a very high bandwidth efficiency. On the other hand, one stream can be transmitted through different layers or different antennas, which also provides multiple ways of exploiting the spatial diversity. The achievable bandwidth efficiency of MIMO-IDMA as well as the tradeoff between spatial multiplexing and spatial diversity have been investigated in this project. To handle a frequency-selective channel, OFDM is a natural choice due to its efficiency in cancelling intersymbol interference (ISI). Combining IDMA and OFDM, the resulting multi-carrier IDMA (MC-IDMA) manages to eliminate ISI. On the other side, ISI can also provide time diversity. If a system is able to resolve the time diversity gain, it will benefit from ISI. This motivates to compare single-carrier IDMA (SC-IDMA) with MC-IDMA. The results show that these two schemes have their own pros and cons in different scenarios. So far, receivers with the perfect channel knowledge are assumed. In practice, channel coefficients need to be estimated before data detection. Conventionally, training symbols are inserted as a midamble or a preamble. In IDMA, one of the superimposed layers can be used for training. In this project, both IDMA as well as the channel model are modelled by a factor graph. Factor graphs are an extremely helpful tool. All scenarios discussed in the following can be represented by a factor graph. In order to deal with fast fading channels, a transfer node is proposed to enable soft channel extrinsic information exchange which outperforms conventional filter-based channel estimation methods in terms of the required number of training symbols. Besides its easy implementation with OFDM and MIMO, IDMA has some other nice features. After superposition, the signal constellation is nearly Gaussian distributed. From an information theoretical point of view, a signal is capacity achieving on the Gaussian channel if and only if the channel output is Gaussian distributed. This feature is useful in communication networks, both in uplink (=MAC channel) and downlink (=broadcast channel). Particularly, the application of IDMA for the purpose of relaying seems to be interesting for future work. As mentioned before, IDMA is a multiple access scheme. Channel coding, modulation and channel access are done jointly. A variant of IDMA is IDM. IDM is a multiplexing scheme, exploiting the same principle in order to separate data streams, but could also been treated as a coded modulation scheme. Hence, IDM is an alternative to other coded modulation schemes like multilevel coding, trellis-coded modulation or bit-interleaved coded modulation. Many results obtained in this project are directly applicable to IDM as well. IDM has a wider range of possible applications than IDMA. The core part of IDMA and IDM is the superposition operation. In ongoing work, we extract the superposition operation and use it as a non-bijective modulation scheme called superposition modulation (SM). In conjunction with proper channel coding, SM is capacity achieving. Details are provided in the dissertation by Tianbin Wo, more results will be published soon. Like IDM, SM has a wide range of possible applications. IDMA, IDM and SM are applicable, for example, in future wireless radio systems. Graphbased detection is also applicable in current standards.
Projektbezogene Publikationen (Auswahl)
- "Multilayer APP detection for IDM," Electronics Letters, vol. 46, no. 1, pp. 96-97, Jan. 2010
M. Noemm, T. Wo, and P. A. Hoeher
- Kapitel 13.9 "Kombinierte Modulation und Kanalcodierung" in: Peter A. Höher, Grundlagen der digitalen Informationsübertragung, Wiesbaden: Vieweg + Teubner, 2011
- Superposition Mapping & Related Coding Technqiues, Dissertation, Christian-Albrechts-Universität zu Kiel, 2011
T. Wo
