Progress In Electromagnetics Research Symposium Proceedings 1963 New Intuitive Metrics for Diversity Performance Evaluation of Multi-element Antenna Systems Vasileios C. Papamichael 1 and Petros Karadimas 2 1 Department of Electrical Engineering, Technological Educational Institute of Western Greece and Wireless Systems Performance (WiSPer), Patras 26441, Greece 2 Department of Computer Science and Technology, University of Bedfordshire, Luton LU1 3JU, UK Abstract— Diversity performance of multi-element antenna (MEA) systems is evaluated using several metrics. The most common are the diversity antenna gain (DAG), the effective diversity gain (EDG) and the actual diversity gain (ADG). These metrics calculate the performance by comparing the MEA system with the reference antenna at a fixed outage probability level, i.e., usually at 1% outage level. As fixed outage levels are just an indication of probability of occur- rence of specific signal levels, they provide limited insight into realistic deep-fade cases. Thus, we introduce three novel metrics, namely, the fading mitigation gain (FMG), the reliability per- centage (RP) and the generalized diversity antenna gain (GDAG) for characterizing diversity performance when deep fades take place in one or more of the diversity branches. Based on the aforementioned metrics, intuitive knowledge on the diversity performance of MEA systems is provided. 1. INTRODUCTION The mitigation of multipath fading is an issue of primary importance in wireless communications systems. Multi-element antenna (MEA) systems combat fading through diversity combining, which potentially increases the received power. Until now, several metrics for the diversity performance have been proposed, with the most important being the diversity antenna gain (DAG) [1], the effective diversity gain (EDG) [2] and the actual diversity gain (ADG) [3]. They are derived by comparing the cumulative distribution functions (CDFs) of the received signal-to-noise ratio (SNR) of the MEA and that of a reference antenna at a specific outage probability level (usually 1% outage probability level). Their only difference lies on the choice of the reference antenna. The DAG is based on comparing the MEA with an isotropic, dual polarized reference antenna and is thus a metric of universal applicability when deriving diversity performance. The EDG arises by comparing the MEA with one of its single elements having 100% efficiency and the ADG by comparing the MEA with an arbitrarily selected antenna. Thus, both EDG and ADG are not of such universal applicability as the DAG, as they are strongly dependent on the chosen reference antenna. As the CDF just presents probabilities of specific SNR levels, DAG, EDG and ADG do not account for performance provided by a MEA system when only deep fades take place in one or more of the diversity branches. Performance in such cases only would be a better indicator of how effectively the MEA system combats fading. Moreover, the estimated percentage of the cases where the MEA’s received SNR is greater than that of the reference antenna would also provide further insights into the performance of MEA systems. Thus, in this paper, three new intuitive metrics focusing on deep-fade cases are proposed. The first metric is called fading mitigation gain (FMG) and accounts for the performance when at least one of the diversity branches of the MEA system has 20dB less power with respect to the average received power of the reference antenna 1 . It is thus a random variable. The second is called reliability percentage (RP) and accounts for the percentage of the cases when the MEA system has greater received power compared with that of the reference antenna. The third is called generalized diversity antenna gain (GDAG) and arises after weighting the two DAGs, i.e., the DAG (weighted by RP) when the MEA received power is greater than that of the reference antenna and the DAG (weighted by 1-RP) when is lower. An isotropic dual polarized reference antenna is considered as in [1], thus, for consistency, we maintain the term DAG in GDAG. The MEA systems and propagation environments are jointly modeled by using the stochastic approach presented in [4]. The remainder of this paper is organized as follows: In Section 2, the three new metrics, i.e., the FRG, RP and GDAG are defined. In Section 3, relevant results are presented considering a two-port compact MEA system and Section 4 draws the conclusion. 1 Considering equal noise powers between the reference antenna and the MEA system, we can consider the received power instead of the SNR.