ACRi’s Dr. Okamoto is the Principal Investigator for the project that is featured in this article, entitled “Robots huddle together to boost communication in urban disaster areas”. The article discusses how equipping a group of robots with Dr. Okamoto’s “sparse antenna” technology improves communications by extending range while mitigating interference. The article mentions that this could be useful for first responders to communicate with robots working in hard-to-reach environments, such as urban disaster areas.

Link to article

Advertisements

This article was published on July 28 in an article entitled: “Team-working robots huddle together to boost comms”. The article discusses how ACRi’s “sparse antenna” technology eliminates dead spots and minimizes interference from multipath or jamming signals. This technology makes it easier to communicate with robots working in challenging communications environments, such as urban disaster areas.

http://www.newscientist.com/article/mg20727715.600-teamworking-robots-huddle-together-to-boost-comms.html


This article discusses ACRi’s latest NSF-funded project. ACRi’s Dr. Okamoto is the Principal Investigator for this project. Chris Kitts and his students at Santa Clara University do a wonderful job with their robotic control algorithms:

Article link


ACRi will present three papers at the SPIE Conference on Defense, Security, and Sensing held from April 5-8 in Orlando, Florida. Those attending the conference are encouraged to contact Dr. Okamoto if they wish to schedule meetings during or after the conference. The three papers that ACRi is presenting are:

High-Power Interference Suppression Via Reduced Complexity Adaptive Blind Beamforming
Authors: Garret Okamoto and Chih-Wei Chen

Abstract: This paper evaluates an adaptive beamforming solution which addresses the significant problem caused by high-power transmitters located in close proximity to users. Current solutions are overwhelmed by the rapid increase in number and variety of strong interference sources. This smart antenna blind beamforming algorithm requires less computational complexity than standard algorithms, making it feasible to be added to current and next-generation systems, and provides a highly adaptive and reliable interference-resistant communications environment. Simulations show that ACRi’s high-power interference mitigation solution automatically nulls jamming signals that are 20 dB to 40 dB stronger than the user signal. The results show that ACRi’s new beamformer achieves close to the theoretically best performance obtained by algorithms such as MVDR that assume the spatial information of the user and interference signals are known (which may not be feasible when high-power interference is present and the user is mobile), which is excellent because ACRi’s algorithm requires significantly less computational complexity and does not require the spatial information to be known in advance for the user or the interference signals. Systems with a limited number of antennas are evaluated because legacy and current generation systems have as little as two antennas.

Tracking and Interference Suppression Performance for the Minimal Computational Complexity Non-Eigen Decomposition Beamformer
Authors: Garret Okamoto and Chih-Wei Chen

Abstract: This paper evaluates tracking and interference suppression performance for the ultra low complexity Non-Eigen Decomposition (NED) blind beamforming algorithm. Current blind beamforming algorithms require too much computational resources for them to be used by ground and air robotic systems and other systems with limited available computational power. This paper focused on the ultra low complexity NED beamforming algorithm for adaptive interference mitigation. NED does not rely on the eignenvalues and eigenvectors used by conventional algorithms and requires significantly less computations, with a total computational load of O(4M-4) per snapshot for a system with M receiving antennas by approximating the cross correlation vector of the received signals in the reference and other antennas. This technique requires neither a training sequence nor an assumption of incoherency among impinging signals. By significantly reduces the computational requirements for beamforming, NED makes it possible for robotics and other systems to achieve the advantages of beamforming—such as enhanced reliability, increased range, resistance to co-channel interference signals, increased throughput and capacity, and extended battery life. Tracking ability determines what types of applications and scenarios a technique is applicable for, particularly important in mobile wireless systems such as the target applications identified for NED (ground robots, unmanned aerial vehicles, and mobile phone handsets). Simulation results show that the DOA estimated by NED is almost identical to the true DOA even when the user moves extremely quickly, with NED’s fast convergence rate (only 4 iterations needed in the highest mobility case) also evident. The high angular movement between samples corresponds to user speeds ranging from 31 km/h to 124,000 km/h if the distance between the mobile user and transmitter is 1 km.

Beamforming Performance for a Reconfigurable Sparse Array Smart Antenna System via Multi-Mobile Robot Cluster Control
Authors: Garret Okamoto, Chih-Wei Chen, and Christopher Kitts

Abstract: This paper describes and evaluates the beamforming performance for a flexible sparse array smart antenna system that can be reconfigured through the use of multiple mobile robots. Current robotic systems are limited because they cannot utilize beamforming due to their limited number of antennas and the high computational requirement of beamformers. The beamforming techniques used in this paper are unique because unlike current beamformers, the antennas in the sparse array are not connected together but instead each robot has a single antenna. This work is made possible through breakthroughs by the authors on ultralow computational complexity beamforming and multi-mobile robot cluster control. This new beamforming paradigm provides spatial reconfigurability of the array to control its location, size, inter-antenna spacing and geometry via multi-robot collaborative communications. Simulation results evaluate the effectiveness of various beamforming techniques when 1, 2, 3, 4, and 8 robots are utilized. The simulation results are also evaluated for multiple geometric configurations for the robots, evaluating whether or not different geometric shapes may provide greater range or interference mitigation performance for different communications scenarios. Preliminary over-the-air measurement results are provided via ACRi’s flexible SDR communications hardware platform that will be integrated with the individual robotic systems.


ACRi is happy to announce that it has been awarded a $150k NSF SBIR Phase I Contract. This award is for research on “Reconfigurable Sparse Array Smart Antenna System via Multi-Robot Control”.

http://www.nsf.gov/awardsearch/showAward.do?AwardNumber=0946027

The abstract of the project can be found at the above link to the official NSF announcement and is also included below:

This Small Business Innovation Research (SBIR) Phase I project develops and evaluates a flexible sparse array smart antenna system that can be reconfigured through the use of multiple mobile robots. Current robotic systems are limited because they cannot utilize beamforming due to their limited number of antennas and the high computational requirement of beamformers. This pioneering research is made possible through recent breakthroughs for ultralow computational complexity beamforming and multi-mobile robot cluster control. Unlike current beamformers, the antennas in the sparse array will not be physically connected together but instead each robot will have a single antenna. By developing new signal processing and robotic control techniques, robotic communications will be enabled where impossible today due to range, dead spots, or interference. Over-the-air measurements will make it possible to finally evaluate how key issues (distance between robots, geometric shape of the sparse array, etc.) affects system performance.

The broader impact/commercial potential of this project is that it can revolutionize commercial robotic systems and other applications in the wireless industry. Enabling multi-robot collaborative communications makes reliable communications possible in worst-case environments. Performance evaluation of sparse arrays will provide valuable insight for collaborative communications for other applications such as distributed sensor networks while the beamformer?s ultralow computational requirement makes it feasible to be added to current and future wireless systems. Creation of a new class of robotic communications will enable robots to be more effective in current applications and create new markets for the robotic sector. The use of robots has increased exponentially with robots increasingly relied upon for defense, law enforcement, and manufacturing, but communication limitations prevent robots from being effective in many situations. Preventing this critical loss of communications for robots searching for people trapped in collapsed buildings or while on scout missions can save lives and have a great societal impact. This research will foster new fields of scientific and technological understanding by enabling Academia and Industry researchers to evaluate the advances made through this pioneering research, which will enable performance optimization for smart antenna systems whether the antennas are physically connected or at different locations.


The January 2009 issue of National Defense Magazine features on article on ACRi’s research advances. This article spotlights how ACRi’s Non-Eigen Decomposition algorithm can enable UAVs and ground robotic systems to utilize beamforming to gain the advantages of communications reliability, range increase, battery life extension, interference mitigation, and LPD/AJ. The article is entitled “Software Improves Connections to Robots.”

This issue of National Defense Magazine won’t be in stores until next week, but you can read the article online at:
http://www.nationaldefensemagazine.org/archive/2009/January/Pages/SoftwareImprovesConnectionstoRobots.aspx


ACRi members Garret Okamoto and Chih-Wei Chen had a paper published at the 2008 IEEE MILCOM conference. The paper was entitled “Minimal Complexity Blind Interference Mitigation via Non-Eigen Decomposition Beamforming.”

Dr. Okamoto gave a presentation at MILCOM based on that paper on November 18, 2008.

The Abstract of that paper is below:

This paper evaluates the ultra low complexity Non-Eigen Decomposition (NED) algorithm for adaptive interference mitigation. Current blind beamforming algorithms require computational complexity too high for many target applications. NED does not rely on the eignenvalues and eigenvectors used by conventional algorithms and requires significantly less computations, with a total computational load of O(4M-4) per snapshot for a system with M receiving antennas by approximating the cross correlation vector of the received signals in the reference and other antennas. The weight vector is a function of only the cross correlation vector and initial guess and does not require a step size. This technique requires neither a training sequence nor an assumption of incoherency among impinging signals. Simulations show that NED achieves comparable performance as other blind beamforming algorithms in nulling interference sources The smart antenna structure for NED implementation in spread spectrum and OFDM system will also be discussed.