David W. Browne

Multi-antenna testbeds for research and education in wireless communications

Wireless communication systems present unique challenges and trade-offs at various levels of the system design process. Since a variety of performance measures are important in wireless communications, a family of testbeds becomes essential to validate the gains reported by the theory. Wireless testbeds also play a very important role in academia for training students and enabling research. In this article we discuss a classification scheme for wireless testbeds and present an example of the testbeds developed at UCLA for each of these cases. We present the unique capabilities of these testbeds, provide the results of the experiments, and discuss the role they play in an educational environment.

An open access wideband multiantenna wireless testbed with remote control capability

This paper introduces an open access wideband multiantenna wireless testbed. The testbed is configured as a four transmit antenna by four receive antenna system based on software defined radio technology. It operates in the 2.4 GHz ISM band and supports an RF bandwidth compatible to IEEE 802.11a/g standard. A robotic positioning system has been developed to automatically control the position and orientation of the antenna array, which makes automated testing possible and greatly reduces engineering time. The testbed is configured such that it can be controlled by any computer on the Internet. This makes remote debugging and testing possible. This characteristic of test automation and remote access through Internet was designed into the testbed to make it a valuable resource to the other researchers. A set of software has been developed to support research using the testbed, with the current focus on MIMO channel characterization and MIMO-OFDM packet communications. A data acquisition system and an offline processing system have been developed for the testbed and have been used for over the air testing to produce results for the ongoing IEEE 802.11n standardization activity.

A signaling scheme and estimation algorithm for characterizing frequency selective MIMO channels

This paper presents a signaling scheme and channel estimation algorithm designed to characterize frequency selective MIMO channels. This scheme has been implemented in a wideband MIMO radio testbed and used to measure the achievable channel capacity in an indoor office environment. Also, we report the achievable capacity for a real hardware platform at a particular combination of channel, SNR, antenna separation and location. This is repeated over a large number of spatially unique antenna placements and locations to give a statistically representative set measurements for the environment. The data sets generated in the measurement campaign gives an indication of the variability of achievable capacity and not just the average channel capacity. This allows future studies to investigate the characteristics of a typical poor channel and how they cause a degradation in system performance.

Experimental Validation of Capacity Preserving Design for MIMO Arrays

MIMO radio communications presents a paradigm shift for antenna array design. This is because system performance depends on how a pair of antenna arrays jointly affect the eigenvalue distributions of MIMO channels. Very little experimental work has been done to confirm what characteristics MIMO arrays must have so that channel capacity is preserved. This work presents field experiments with a MIMO testbed that test two alternative MIMO array design approaches. The first approach is to construct arrays whose elements each have omnidirectional radiation patterns. The second approach is to construct arrays whose elements each cover a spatial sector. Results show conclusively what some of the necessary characteristics are for capacity preserving MIMO arrays.

Antenna topology impacts on measured MIMO capacity

The achievable spectral efficiency of multiple antenna radios is greatly improved by Multiple Input Multiple Output (MIMO) signaling. However, the benefit of MIMO is lessened in practice due to mutual coupling in the antenna arrays. This problem is exacerbated by the size restrictions on antenna arrays in portable devices. We present a compact antenna array that is designed to limit mutual coupling and has good radiation efficiency. The array was fabricated and integrated into a MIMO radio testbed. Using this array, we evaluate the testbeds achievable capacity and compare it with the performance achieved using arrays of dipoles with varied spacing. The results provide valuable insight on the design requirements for compact antenna arrays in portable MIMO sytems. We find that, for rich multipath environments, MIMO transceivers with weak coupling-induced correlation at one end of the link and a moderate coupling-induced correlation at the other end have very little loss of system capacity compared to the uncorrelated MIMO case. However, although mutual coupling may not cause appreciable loss in system capacity, mutual coupling should be minimized in compact MIMO array designs if the radiating efficiency is to be maximized.

Design of integrated antenna arrays for MIMO enabled laptops

This paper presents a novel antenna array for MIMO enabled laptops that is designed to preserve channel capacity. The antenna-level performance of this array is compared with the performance of a simple microstrip patch antenna array. This comparison is made with both arrays fully integrated within a realistic laptop structure.

Performance of integrated antenna arrays for MIMO enabled laptops

This work compares the system-level performance of two laptop antenna array designs for MIMO communications. The arrays are integrated into laptop display housings and differ in that one array has elements with directional radiation patterns while the other has elements designed for omnidirectional radiation. The performance of each laptop array is tested using a pair of MIMO radios in a full range of indoor multipath scenarios. The performance of both arrays is then compared using MIMO capacity related metrics. It has been shown that antenna arrays must have omnidirectional elements and low mutual coupling to be optimal in the MIMO sense. The results presented in this study show how well this requirement is met by an array of integrated back-to-back elements.

Experiments in space-time modulation

This paper reports on some experiments with multiple antenna radios. Specifically this paper compares the performance between an experimental system with training and coherent demodulation and a system using noncoherent techniques of communication. The paper also examines robustness in performance across a variety of channel conditions.

Directional radio propagation measurements for near-ground peer-to-peer networks

This demonstration will present a radio testbed that is being used to characterize the quality of directional peer-to-peer wireless communications in wilderness environments.

Antenna actuation for radio telemetry in remote sensor networks

We present a radio telemetry solution for remote sensor networks that uses rotational actuation of directional antennas. The benefits of this solution over either omnidirectional or fixed directional antennas for intra-nodal communications include greater energy efficiency, sparse node deployment, and robustness to node failure. Four strategies for minimizing link acquisition time are presented. Each strategy uses a search algorithm tailored to a unique combination of radio hardware to provide nodes with coordination and/or geolocation capabilities. These strategies are implemented on a two-node testbed and their performance is measured in a realistic field trial. Results show that the use of secondary radios for search coordination and geolocation provide an order of magnitude reduction in search duration. In particular, the strategy using only search coordination yields performance that is independent of node separation while performance of the strategy using only geolocation is strongly dependent on range.