Holger Jaekel

Occupation Measurements Supporting Dynamic Spectrum Allocation for Cognitive Radio Design

Efficient use of radio spectrum is a necessity for future wireless systems. This can, for example, be accomplished by cognitive radios. An important point for designing future systems is to gain knowledge about the occupation of the frequency bands which may be used by the cognitive radio. This includes the general characterization of the traffic density at the frequency bands as well as a detailed analysis of the temporal frequency occupation. Thus, this paper describes measurements taken in the GSM network at three different scenarios. Furthermore, a detailed analysis of the measurement results is presented including statistical as well as spectral occupation analysis.

Multi-agent radio resource allocation

This paper presents a spatially distributed and dynamic billing, pricing and allocation mechanism for which a user terminal requires Cognitive Radio abilities. That is, the Cognitive Radio abilities will be applied to the economical environment. The radio resource goods are allocated to the users by a multi-unit sealed-bid auction. Intelligent entities like the bidding strategy have to represent operator’s and users’ behavior and to make decisions for them in order to fulfill the preferences and QoS. The main functionalities, used to execute the dynamic auction sequence and located in the MAC are described, assuming the entities possess Cognitive Radio abilities.

General expression of outage probability in cooperative networks

A general expression of outage probability in a wireless relay network with an arbitrary number of relays and selection combining at the destination is derived. This formula is then adopted to independent but not identically distributed channel coefficients and lower and upper bounds are developed. We study two relay schemes: the one-relay case and the two-relay case. The latter can further be divided into a two-phase and a three-phase transmission scheme which depend on the cooperation among the relays. For all considered cases, outage probabilities in Rayleigh fading environments are derived and analysed and it is demonstrated by numerical simulations that all relay protocols achieve full order of diversity. We further elaborate the impact of relay locations on outage probability. Finally, we discuss the required signal-to-noise ratio in order to achieve a predefined outage probability. With respect to the mentioned evaluation metrics the two-phase relaying scheme achieves by far the best performance, however, with the drawback of increased receiver complexity. Copyright

Do decode-and-forward relaying protocols beat transmit diversity?

This paper addresses the question whether decode-and-forward relaying protocols can beat transmit diversity with respect to ε-outage capacity. We derive expressions of ε-outage capacity for transmit diversity, decode-and-forward, selective decode-and-forward, and incremental decode-and-forward. We demonstrate that the relaying protocols outperform transmit diversity in certain regions. Decode-and-forward is beneficial over transmit diversity when the relay is located close to the source. For selective decode-and-forward to beat transmit diversity, the relay should be placed between the source and the destination, but not too close to the source. The best performance is achieved by incremental decode-and-forward. The region in which this scheme is beneficial over transmit diversity is large. Moreover, as the region becomes smaller, the outage probability for the source-destination link becomes larger.

Outage regions and optimal power allocation for wireless relay networks

We study outage regions for energy-constrained multi-hop and adaptive multi-route networks with an arbitrary number of relay nodes. Optimal power allocation strategies in the sense that outage probability is minimized are derived depending on the distances between the transmit nodes. We further investigate the rate gain of adaptive multi-route and multi-hop over direct transmission. It is shown that a combined strategy of direct transmission and adaptive multi-route outperforms multi-hop for all values of rate R. It can be stated that cooperation strategies are beneficial for low-rate systems where the main goal is a very low outage probability of the network. As the rate is increased, direct transmission becomes more and more attractive.

Multi-Agent Wireless System for Dynamic and Local Combined Pricing, Allocation and Billing

In established communication systems prices are determined quasi-static and for a large area according to a fixed price model. Additionally, the main decision criteria of resource allocation is the arrival time of resource requests. This system approach is not able to respond to the spatial distributed and dynamic users’ demand and willingness-to-pay. In this paper a system will be proposed to combine pricing, allocation and billing in order to react dynamically and locally to the market assuming that the user terminal has cognitive radio abilities and multiple interfaces.

Comparing the Outage Capacity of Transmit Diversity and Incremental Relaying

We derive the -outage capacity of a wireless relay network where relaying only takes place if the desti- nation has not been able to decode the source message. We refer to this scheme as incremental relaying. The relays use space-time block coding and perform decode-and- forward. We compare incremental relaying to transmit diversity in terms of -outage capacity and signal-to-noise ratio gain. It is demonstrated that incremental relaying outperforms transmit diversity only in certain regions that determine the relay locations. Moreover, our results confirm the well-known fact that decode-and-forward is es- pecially preferable if the relays are located close to the source.

Signal-to-noise ratio gains for cooperative relaying strategies

We study signal-to-noise ratio (SNR) gains for two different cooperation strategies that are applied to networks consisting of one source, one destination, and one or two relays, respectively. The first strategy is called orthogonal decode-and-forward where all transmit terminals transmit interference free in orthogonal time slots. The second strategy is called nonorthogonal decode-and-forward where all relays transmit simultaneously by applying space-time block coding. We demonstrate that the SNR gain ¿¿ is proportional to K+1¿(1/pout K) with K representing the number of relays. We further define the SNR gain exponent as a new criterion to evaluate the asymptotic behavior of SNR gain for high-rate regimes. It will be demonstrated that the SNR gain exponent only depends on the number of transmission phases.

Outage regions of cooperative networks for different coding schemes

In this paper we present outage regions for two cooperation strategies, namely multi-routing and adaptive multi-routing, if two different coding schemes are used. The first coding scheme is repetition coding and the second one is parallel channel coding. We show that for one cooperation strategy, the outage region for parallel channel coding is a strict subset of the outage probability of repetition coding, thus, leading to a smaller outage probability. We demonstrate that adaptivemulti-routing in conjunctionwith parallel channel coding achieves by far the best performance.

Error performance of multi-hopping half-duplex relay networks

We derive a compact description of error performance of a multi-hopping half-duplex relay network, where each relay performs decoding. For this, we introduce the event of consecutive error which describes that a decoding error might not be corrected by chance at a subsequent relay node. Furthermore, two tight upper bounds are derived. The first one is based on the union bound method and the second one describes the probability that there occurs at least one decoding error in the network. General expressions on error probability in slow, flat fading Rayleigh channels are developed as well and adopted to the chosen transmission scenarios to ensure fair comparison with respect to the amount of transmitted data and system energy. It is shown and evaluated by simulations that relaying achieves better results on error performance even if no spatial diversity is employed. Gains are evaluated by a new metric called asymptotic probability gain.