Ramiro Samano-Robles

Wireless Networks With Retransmission Diversity and Carrier-Sense Multiple Access

Network diversity multiple access protocols (NDMA) have been shown to considerably outperform previously proposed algorithms. However, several issues in their analysis remain unsolved, particularly under the assumption of imperfect collision-multiplicity detection in asymmetrical system configurations (i.e., configurations where the users have different channel and queuing statistics). To partially fill this gap, this paper presents a detailed study on stability, throughput and delay properties of asymmetrical nonblind NDMA protocols under such imperfect detection assumption. Additionally, the protocol formulation includes a carrier-sense mechanism which is found to improve system performance under finite-SNR environments and which can be useful in assessing the benefits of retransmission diversity on IEEE 802.11 wireless local area networks (WLANs). New expressions for the parameters and performance metrics of the NDMA protocol under different assumptions are here provided. These include the stability condition of a system without feedback to acknowledge correct packet reception and the derivation of the boundaries of the throughput region using a joint cross-layer optimization with respect to the probabilities of transmission and false alarm. Simulation results are used to further assess the performance of the protocol and to confirm the analytic results.

An infinite user model for random access protocols assisted by multipacket reception and retransmission diversity

The study of random access protocols has recently regained attention due to new cross-layer schemes such as multipacket reception (MPR) systems and network diversity multiple access protocols (NDMA). Despite their relevance, these two systems have only been simultaneously studied employing finite user population models and considering perfect detection of the active users, which are assumptions only useful in scenarios with low numbers of users and high values of the SNR. The purpose of this paper is to introduce an infinite user population model, valid for scenarios with large numbers of users and finite traffic loads, which allows us to extend the available results on ALOHA MPR protocols to systems that use retransmission diversity (RD). Unlike existing approaches our model includes both the effects of packet decoding errors and the effects of imperfect detection of the active users, which considerably affect the performance of conventional NDMA systems in finite SNR environments. Additionally, the proposed model provides a better approximation to the queuing delay of NDMA protocols than the conventional formula of an M/G/1 queue with vacations. Finally, the proposed algorithm also represents an extension and generalization of contention binary tree algorithms assisted by signal processing tools such as SICTA (successive interference cancellation tree algorithm) and other algorithms assisted by source separation. The benefits of the proposed model are assessed using simulation and analytic results.

A slotted ALOHA protocol with cooperative diversity

Over the last few years, the area of cooperative communications has regained attention within the physical layer community. However, existing works on cooperative random access protocols are relatively scarce and biased towards their physical layer properties, thus leaving unattended important problems of the medium access control (MAC) sublayer such as calculation of the backlog delay, design of appropriate back-off retransmission strategies, and stability evaluation, among many others. This paper partially fills this gap by studying the general performance of a symmetrical Slotted ALOHA protocol in which a cooperative relaying phase is enabled in order to improve the decoding probability of collision-free transmissions. Infinite and finite user schemes are used, and for the latter, Bernoulli and Markov models are further employed to study the steady- and the dynamic-state properties of the protocol, respectively. A stochastic reception model is presented which fairly describes the underlying physical layer events from the perspective of the MAC sublayer, including correct packet decoding probabilities, relay node availability, and error detection capabilities. Important results regarding the boundaries for optimum performance of cooperative relaying schemes and useful guidelines for the design of optimum relaying strategies are here derived and discussed.

Low-Complexity User Selection for Rate Maximization in MIMO Broadcast Channels with Downlink Beamforming

We present in this work a low-complexity algorithm to solve the sum rate maximization problem in multiuser MIMO broadcast channels with downlink beamforming. Our approach decouples the user selection problem from the resource allocation problem and its main goal is to create a set of quasiorthogonal users. The proposed algorithm exploits physical metrics of the wireless channels that can be easily computed in such a way that a null space projection power can be approximated efficiently. Based on the derived metrics we present a mathematical model that describes the dynamics of the user selection process which renders the user selection problem into an integer linear program. Numerical results show that our approach is highly efficient to form groups of quasiorthogonal users when compared to previously proposed algorithms in the literature. Our user selection algorithm achieves a large portion of the optimum user selection sum rate (90%) for a moderate number of active users.

Quality of Service in Wireless Network Diversity Multiple Access Protocols Based on a Virtual Time-Slot Allocation

In this paper, we propose a new resource allocation mechanism which is designed to improve the multiuser detection of wireless network diversity multiple access (NDMA) protocols. The mechanism consists of allocating an average number of time- slots to the active user population according to a prescribed quality of service requirement. It is dubbed virtual because it does not rely on user scheduling over different time-slots, but instead it is controlled by adjusting the probability of false alarm of each user. The allocation mechanism improves the throughput of conventional NDMA protocols at the expense of both an access delay degradation and an increased system complexity. The system requires to recover the signals from a mixing system with more outputs (the collected network transmissions) than inputs (the collided packets). By setting the average allocated time-slots to remain constant over different traffic loads, both the optimum transmission probabilities and the stability region of the protocol are approximated by relevant closed-form expressions. Also, the proposed analytical formulation extends the expressions of conventional NDMA systems to the asymmetrical user case (i.e. users with different data rates and detection statistics). Finally, it is shown that under extreme traffic loads, multiuser detection conditions and quality of service requirements, the proposed system degrades into the equivalent of the traditional networking protocols TDMA (time division multiple access) and Slotted- ALOHA (S-ALOHA).

A cross-layer approach to the downlink performance analysis and optimization of distributed antenna systems in multi-cell environments

Distributed antenna systems (DAS) constitute one of the most attractive schemes to efficiently achieve the stringent quality of service demands of next generation wireless networks. However, existing works have been only focused on the analysis of the downlink of single-user DAS multi-cell networks with fixed node configurations and without considering the effects of upper layer algorithms. This paper partially fills this gap by presenting the performance analysis and optimization of the downlink of DAS systems in multi-cell environments with single and multiple users and with basic resource management algorithms. The cross-layer analysis and optimization is first carried out using a simplified system-level simulation tool which provides enough flexibility for the understanding of the basic rules of operation of the proposed algorithms. After this, the most promising algorithms are implemented in a full version of a system level simulator which has been especially developed for the accurate analysis and validation of 3G (third generation) and B3G (beyond 3G) radio access technologies. Our results show that under the appropriate allocation and management policies, distributed antenna systems efficiently counteract the main impairments of current cellular architectures.

A multiaccess protocol assisted by retransmission diversity and multipacket reception

In this paper we propose a new multiaccess protocol that combines the concept of splitting tree algorithms for collision resolution with two of the most relevant cross-layer technologies for random access: retransmission diversity (e.g., NDMA-Network Diversity Multiple Access) and multipacket reception (MPR). The proposed protocol is shown to outperform all the existing algorithms based on either NDMA or MPR. Additionally, the protocol formulation provides an important generalization of the model used for the analysis of solutions in these three fields. Unlike conventional NDMA protocols, in which all the colliding users are requested to immediately retransmit in the next time-slot, our proposed algorithm calculates the optimum set of users allowed to retransmit at each one of the following time-slots. The optimization is based on the previous collected transmissions and on the MPR and source separation (SS) probabilities, thus maximizing throughput and minimizing access-delay. Two possible suboptimal algorithms with simplified feedback assumption and hence suitable for distributed resolution are further derived from the original algorithm: an enhanced version of NDMA assisted by MPR and a fair splitting tree algorithm assisted by MPR and SS. The capacity/stability region of the protocol for several system configurations with two active users is employed to assess the benefits of the proposed algorithms.

Stability Properties of Network Diversity Multiple Access with Multiple-Antenna Reception and Imperfect Collision Multiplicity Estimation

In NDMA (network diversity multiple access), protocol-controlled retransmissions are used to create a virtual MIMO (multiple-input multiple-output) system, where collisions can be resolved via source separation. By using this retransmission diversity approach for collision resolution, NDMA is the family of random access protocols with the highest potential throughput. However, several issues remain open today in the modeling and design of this type of protocol, particularly in terms of dynamic stable performance and backlog delay. This paper attempts to partially fill this gap by proposing a Markov model for the study of the dynamic-stable performance of a symmetrical and non-blind NDMA protocol assisted by a multiple-antenna receiver. The model is useful in the study of stability aspects in terms of the backlog-user distribution and average backlog delay. It also allows for the investigation of the different states of the system and the transition probabilities between them. Unlike previous works, the proposed approach considers the imperfect estimation of the collision multiplicity, which is a crucial process to the performance of NDMA. The results suggest that NDMA improves not only the throughput performance over previous solutions, but also the average number of backlogged users, the average backlog delay and, in general, the stability of random access protocols. It is also shown that when multiuser detection conditions degrade, ALOHA-type backlog retransmission becomes relevant to the stable operation of NDMA.

Frequency-Reuse Planning of the Down-Link of Distributed Antenna Systems with Maximum-Ratio-Combining (MRC) Receivers

Distributed antenna systems (DAS) have been shown to considerably outperform conventional cellular systems in terms of capacity improvement and interference resilience. However, the influence of frequency reuse planning on the performance of DAS remains relatively unknown. To partially fill this gap, this paper presents a comparative analysis of the down-link of DAS versus conventional cellular systems using different values of frequency reuse factor. The analysis assumes Rayleigh fading channels and it also considers maximum-ratio-combining (MRC) receivers at the user terminals to exploit diversity both in the transmission and reception links. Numerical evaluation of the analytical expressions shows that, in general, for most of the cases DAS can achieve better performance figures than conventional cellular systems using considerably smaller values of frequency reuse factor. Conversely, DAS can significantly improve the throughput (2×–3×) and power consumption (6–10 dB) of conventional systems when using the same frequency reuse factor. An interesting result shows that in some particular cases DAS outperform conventional cellular systems no matter the frequency reuse factor used by the latter one, which indicates an effective capacity gain provided by the combined operation of DAS and MRC receivers.

A contention binary tree algorithm assisted by source separation

Inspired by the network diversity multiple access (NDMA) protocols and by the successive interference cancelation tree algorithm (SICTA), we propose a modification to the conventional contention binary tree protocol by incorporating signal separation resources. At each step of the algorithm, the system attempts an optional source separation or a successive interference cancelation using the stored collided packets. The system generalizes all the previously formulated protocols according to either the available system complexity (related to the maximum number of colliding users) or to the system separation capability (related to the structure of the mixing matrix and the probability of successful separation under finite SNR environments). The protocol reduces to a conventional standard tree algorithm (STA) with maximum stable throughput (MST) of 0.346 packets/time-slot for systems with no signal processing capabilities. It also reduces to the SICTA algorithm with an MST of 0.693 for systems with perfect successive interference cancelation. Finally, by adapting the splitting probability according to the source separation capabilities of the system, the algorithm reduces to the NDMA protocol with an MST=1 for a system with perfect separation capabilities and simplified feedback. The tree algorithm is also recognized here as a decentralized networking solution that helps in reducing the system complexity (in terms of maximum number of collided users) and in resolving lost packets from the source separation attempted in previous slots.