John C. Koshy

Joint Source and Relay Optimization for a Non-Regenerative MIMO Relay

We consider a non-regenerative MIMO relay system where the source, relay and destination are all equipped with multiple antennas. The relay does not decode the packets but performs a multi-dimensional amplify-and-forward function (a relay matrix) on the baseband signals. Under the condition that the source is white, the relay matrix that maximizes the capacity between the source and the destination has been previously found. In this paper, we show a new result on how the source covariance matrix and the relay matrix can be jointly optimized to maximize the source-destination capacity. It is shown that the optimal coordinate system governed by the previously discovered relay matrix is still valid under the joint optimization, and the joint optimization yields a further capacity gain when the SNR at the relay is low

Low complexity iterative MIMO receiver based on successive soft interference-cancellation and MMSE spatial-filtering

For MIMO systems, the ST-BICM approach using iterative processing has been recognized as a method for achieving near-capacity performance. However, the a posteriori probability calculator in the MIMO detector, relying on exhaustive or partial search of candidate bit vectors, is not amenable to practical implementation at high rates ( ≥ 16 raw bits per channel use) due its exponential complexity in rate. On the other hand, recently developed low complexity single stream demappers based on a parallel approach, while yielding comparable performance in ideal conditions, suffer significant performance loss in several practical scenarios. In this paper, we propose a novel demapper which closes the performance gap between the low complexity detectors based on single stream demapping and their exponentially complex counterparts.

Experimental results using a MIMO test bed for wideband, high spectral efficiency tactical communications

This paper presents results from a MIMO measurement campaign designed to demonstrate spectral efficiencies of 13.7-26.0 information bits per second per Hz (bps/Hz) using an 8times8 test bed over a range of outdoor wireless channels with low antenna heights at each end of the link. Links included line-of-sight and non-line-of-sight channels obstructed by foliage and buildings. Measurements were performed at 456, 904, and 2177 MHz using a variety of array configurations including multiple polarization configurations. The transmitted MIMO OFDM signal occupied a 5 MHz bandwidth and used several error control coding and interleaving strategies. Signal processing at the receiver employed iterative detection with spatial filtering, SISO demappers, BCJR-based decoding (for turbo codes), and soft-cancellation of self-interference. Results demonstrated successful link closure at spectral efficiencies up to 26 bps/Hz using real-world synchronization, channel estimation, and carrier-frequency offset tracking

A low complexity iterative receiver for high spectral-efficiency battlefield MIMO communications

For multiple-input multiple-output (MIMO) systems, space-time bit-interleaved coded modulation (STBICM) using iterative detection has been recognized as a method to achieve near-capacity performance. However, the a posteriori probability (APP) detector required for this near-optimal performance exhibits prohibitive implementation complexity at high rates (ges16 coded bps/Hz) due to its exponential complexity in rate. With a view to enabling practical implementations of high spectral-efficiency wireless communications, this paper introduces a single-stream detection approach using spatial-filtering (or nulling) and soft-cancellation. Relative to the conventional APP detector, the proposed method is shown to have a superior performance-complexity trade. This paper also demonstrates the robustness of the proposed single-stream detection approach in overloaded conditions where there are fewer receive antenna elements devoted to signal separation than there are transmitted streams

Measurements of Multiple-Input Multiple-Output (MIMO) performance under army operational conditions

This paper presents the results from a field measurement campaign that was conducted to provide an understanding of Multiple-Input Multiple-Output (MIMO) performance relative to that of a Single-Input Single-Output (SISO) system in military-type environments. The Space & Terrestrial Communications Directorate (STCD) MIMO system was used to conduct these experiments. The system has two antennas and operates at transmit frequencies of 430 and 1380MHz, which fall within military frequency bands. A variety of operational environments with many scenarios were considered. The experiments were conducted at C4ISR OTM testing facility, Fort Dix, New Jersey for transmissions along a wide road, a narrow road and through heavy foliage. These experiments measured the throughput gain of MIMO over SISO given the same transmit power and channel usage. The gain in throughput was corroborated by information-theoretic capacity calculations using channel estimates collected during the experimental campaign. In addition, the impact of the antenna spacing on throughput gain was also studied. Depending on the multipath-richness of the environment, the experimental results show that the 2-antenna system provides a throughput of 1.3 to 2.0 times that of a SISO system. On average, a range extension of 1.5 times could be realized for all the considered scenarios and transmit frequencies. The results suggest that antennas in MIMO systems should be placed at least a half of carrier wavelength apart, as indicated in open literature.

Distributed carrier synchronization for HF cooperative communication employing randomized space time block coding

The HF channel supports the wireless needs of expeditionary forces for both communication between nearby unit members (≪5km) and over-the-horizon reach-back to ship or rear base (OTH ≫ 500km). Cooperative communication combined with MIMO methods employing randomized space-time block coding (RST-BC) promise significant gain for over-the-horizon transmission which allows lower transmit power and improved warfighter mobility. However, RST-BC requires tight control of RF carrier frequency to be effective, without reliance on GPS or centralized timing protocols. We examined consensus timing created by a distributed first-order frequency locked PLL (FLL) comprised of the ad-hoc network of cooperating nodes. The necessary carrier synchronization can be achieved with distributed first-order FLL using only locally derived carrier frequency offset information. We report the performance of a fast carrier recovery method for the BPSK recruitment link with measured HF groundwave signals. A network simulation of the proposed distributed synchronization method is presented based on measured recruitment link behavior. Fast settling time is demonstrated with adequate tolerance to failure of individual nodes. The method is sufficient to meet the demanding requirements of RST-BC.

Performance Prediction in Adaptive Mimo Systems

In Multiple Input Multiple Output (MIMO) systems, predicting performance prior to data transmission remains a difficult problem. This paper proposes an algorithm to predict the average bit error rate (BER) and packet error rate (PER) of a MIMO system, after decoding the error correction code (ECC). Since the algorithm makes no a-priori assumptions about the channel, it is applicable to a wide variety of random channels. In our results, we show a good match between the predicted BER and PER, and the BER, PER from simulated data transmission.

A new low-complexity demapper for high-performance iterative MIMO: information-theoretic and BER analyses

Space-time bit interleaved coded modulation with iterative detection has been recognized as a method for achieving near-capacity performance using multiple-input multiple-output (MIMO) systems. However, the maximum-likelihood/joint-detection, required for this method, exhibits prohibitive implementation complexity at high rates (/spl ges/15 bps/Hz). With a view to enabling practical implementations of high spectral-efficiency wireless communications, this paper introduces a new single-stream detection approach using spatial-filtering and soft-cancellation. This method also exhibits superior performance relative to the conventional multi-stream detection approach as long as there are as many receive elements as there are transmit streams. The proposed method is evaluated in terms of bit error rate performance and information-theoretic considerations using an EXIT (extrinsic information transfer) chart method.

Performance Prediction for Turbo Codes in MIMO Wireless Systems

In multiple-input multiple-output (MIMO) systems, accurately predicting performance prior to data transmission remains a difficult problem. This paper proposes an algorithm to predict the bit error rate (BER) and packet error rate (PER) of a MIMO system with turbo coding. First, we modify the union bound (UB) with the help of Craigs formula to make it more amenable to random channels. Second, we extend a method to compute turbo code distance spectrums to accommodate puncturing and code termination. We apply this version of the UB and the distance spectrums to a MIMO system, and produce an algorithm to predict performance. In our results, we show tight bounds at lower, and practical, BERs and PERs.

Iterative MIMO detector using a group-wise approach

For multiple-input multiple-output (MIMO) systems, space-time bit-interleaved coded modulation (ST-BICM) using iterative detection has been recognized as a method for achieving near-capacity performance. However, the a posteriori probability detector required for the conventional ST-BICM receiver is not amenable to practical implementation for high-rate MIMO due to its exponential complexity in rate. This motivated the development of a single-stream detector exhibiting polynomial complexity and capable of delivering performance comparable to the conventional detector under near-ideal conditions. However, under realistic operating scenarios, with possibly low-rank channel conditions and/or employing fewer receive antennas than transmitted streams, the performance gap between the single-stream detector and the conventional detector widens. With a view to enabling practical implementations of high-rate near-capacity MIMO, we propose a novel detector based on a group-wise approach that can be flexibly configured to deliver the best possible combination of performance and complexity over a wide range of realistic operating conditions.