John N. Sahalos

Increased Range Bistatic Scatter Radio

Scatter radio achieves communication by reflection and requires low-cost and low-power RF front-ends. However, its use in wireless sensor networks (WSNs) is limited, since commercial scatter radio (e.g. RFID) offers short ranges of a few tens of meters. This work redesigns scatter radio systems and maximizes range through non-classic bistatic architectures: the carrier emitter is detached from the reader. It is shown that conventional radio receivers may show a potential 3dB performance loss, since they do not exploit the correct signal model for scatter radio links. Receivers for on-off-keying (OOK) and frequency-shift keying (FSK) that overcome the frequency offset between the carrier emitter and the reader are presented. Additionally, non-coherent designs are also offered. This work emphasizes that sensor tag design should accompany receiver design. Impact of important parameters such as the antenna structural mode are presented through bit error rate (BER) results. Experimental measurements corroborate the long-range ability of bistatic radio; ranges of up to 130 meters with 20 milliwatts of carrier power are experimentally demonstrated, with commodity software radio and no directional antennas. Therefore, bistatic scatter radio may be viewed as a key enabling technology for large-scale, low-cost and low-power WSNs.

Sparse Linear Array Synthesis With Multiple Constraints Using Differential Evolution With Strategy Adaptation

This letter addresses the problem of designing sparse linear arrays with multiple constraints. The constraints could include the minimum and maximum distance between two adjacent elements, the total array length, the sidelobe level suppression in specified angular intervals, the main-lobe beamwidth, and the predefined number of elements. Our design method is based on differential evolution (DE) with strategy adaptation. We apply a DE algorithm (SaDE) that uses previous experience in both trial vector generation strategies and control parameter tuning. Design cases found in the literature are compared to those found by SaDE and other DE algorithms. The results show that fewer objective-function evaluations are required than those reported in the literature to obtain better designs. SaDE also outperforms the other DE algorithms in terms of statistical results.

An evaluation of cloud-based mobile services with limited capacity: a linear approach

Mobile computing is pervading networks at an increasing speed as mobile devices are used with diverse forms of wireless technologies to access data. This paper evaluates different cloud-supported mobile services subject to limited capacity, as the selection of a service may introduce additional costs, such as those that derive from the additional amount of memory required for processing. In this context, a novel linear model and algorithm in the mobile cloud computing environment are proposed from the service capacity perspective, considering the cost that derives from the unused capacity. The probability of overutilization or underutilization of the selected service is also researched, once a linear growth in the number of users occurs. To further make effective and strategic investment decisions when selecting the appropriate cloud-based mobile service to lease off, the model formulation is based on a cost–benefit appraisal. The proposed quantification approach is evaluated with respect to four different case scenarios, exploiting a web tool that has been developed as a proof of concept and implementing the algorithm to calculate and compare the benefits and costs in the mobile cloud-based service level.

On the Design, Installation, and Evaluation of a Radio-Frequency Identification System for Healthcare Applications [Wireless Corner]

In this paper, we present the design, implementation, and testing of a radio-frequency identification (RFID) system for healthcare applications. The constantly growing passive RFID technology at ultra-high frequencies (UHF), in conjunction with current state-of-the-art information and communication technologies (ICTs), was used for the system design. The end product was installed at an oncology hospital in Cyprus, where it was thoroughly evaluated by medical staff and hospital administrators. This pilot project had three main objectives: a) automatic and error-free patient identification of in-hospital patients using RFID-enabled cards or wristbands; b) Real-time location service (RTLS) for locating and tracking medical assets and high-value equipment in the hospital ward; c) quick and hassle-free drug inventory management through the use of inexpensive smart labels and cost-effective stationary readers. Here, we present a detailed description of the three major subsystems of the pilot project, emphasizing the main features and capabilities of the system, important design and implementation issues, as well as system evaluation and testing. During the design stage of the project, special emphasis was placed on user friendliness, system capabilities, adequate coverage and tag readability, privacy and security of sensitive patient data, system reliability, and the daily practices of medical personnel and hospital administrators.

Bistatic backscatter radio for power-limited sensor networks

For applications that require large numbers of wireless sensors spread in a field, backscatter radio can be utilized to minimize the monetary and energy cost of each sensor. Commercial backscatter systems such as those in radio frequency identification (RFID), utilize modulation designed for the bandwidth limited regime, and require medium access control (MAC) protocols for multiple access. High tag/sensor bitrate and monostatic reader architectures result in communication range reduction. In sharp contrast, sensing applications typically require the opposite: extended communication ranges that could be achieved with bitrate reduction and bistatic reader architectures. This work presents non-coherent frequency shift keying (FSK) for bistatic backscatter radio; FSK is appropriate for the power limited regime and also allows many RF tags/sensors to convey information to a central reader simultaneously with simple frequency division multiplexing (FDM). However, classic non-coherent FSK receivers are not directly applicable in bistatic backscatter radio. This work a) carefully derives the complete signal model for bistatic backscatter radio, b) describes the details of backscatter modulation with emphasis on FSK and its corresponding receiver, c) proposes techniques to overcome the difficulties introduced by the utilization of bistatic architectures, such as the carrier frequency offset (CFO), and d) presents bit error rate (BER) performance for the proposed receiver and carrier recovery techniques.

Bistatic backscatter radio for tag read-range extension

This work examines the idea of dislocating the carrier transmission from the tag-modulated carrier reception, i.e. bi-static rather than mono-static backscatter radio. In that way, more than one carrier transmitters can be distributed in a given geographical area and illuminate a set of RF tags/sensors that modulate and scatter the received carrier towards a single software-defined receiver. The increased number of carrier transmitters and their distributed nature assists tags to be potentially located closer to one carrier transmitter and thus, improves the power of the scattered signals towards the receiver. Specifically, this work a) carefully derives near-optimal detectors for bi-static backscatter radio and on/off keying (OOK) tag modulation (which is widely used in commercial tags), b) analytically calculates their bit error rate (BER) performance, and c) experimentally tests them in practice with a custom bistatic backscatter radio link. As a collateral dividend, it is shown that the non-linear processing of the proposed receivers requires certain attention on the utilized tag design principles, commonly overlooked in the literature, validating recently reported theoretical results on the microwave domain.

Room-Coverage Improvements in UHF RFID with Commodity Hardware [Wireless Corner]

This work studies the three-dimensional (3D) identification performance of UHF RFID systems with commodity hardware. Detailed three-dimensional propagation modeling is developed, with ray-tracing that allows examination of tag- as well as reader-antenna diversity. It is shown that multipath can enhance identification performance compared to free-space conditions. Furthermore, it is found that tag diversity can enhance identification performance on the order of 10%. Reader-antenna diversity becomes beneficial only when special attention is given to controlling the destructive summation of the transmitted fields and simple, general antenna-installation rules are provided. Performance can be further enhanced with the introduction of a phase shifter or appropriate transmission scheduling, and various examples are discussed. Measurements inside a room with a dense three-dimensional grid of passive RFID tags confirmed the results. Finally, a method to perform power-measurements with commodity RFID hardware that exploits the sensitivity-during-read threshold of each tag is put forth.

A Site-Specific Stochastic Propagation Model for Passive UHF RFID

This letter proposes a computationally inexpensive, site-specific propagation model, useful in coverage evaluation of radio frequency identification (RFID) networks with passive tags. The parameters of a Rice distribution for each point in the volume of interest are computed using site-specific approximations that address overall space geometry, materials, and polarization. The probability of successful identification of passive RFID tags is calculated. Coverage statistics and performance evaluation of complex RFID networks can be quickly conducted. Experimental results at the UHF regime corroborate the accuracy of the model.

Multiple Signal Direction of Arrival (DoA) Estimation for a Switched-Beam System Using Neural Networks

A new Direction of Arrival (DoA) estimation method based on Neural Networks (NNs) is presented. The proposed NN-DoA procedure is especially designed for a Switched-Beam System (SBS), whose basic component is an 8 £ 8 Butler Matrix (BM). The technique is simple and appropriate for real time applications. Simulations of DoA estimation tests show accurate results even for a big set of simultaneously incident signals.

Theoretical investigation of measurement procedures for the quality assurance of superficial hyperthermia applicators

This work presents the results obtained from simple numerical models concerning the measurement uncertainty with thermographic techniques used for the evaluation of superficial hyperthermia applicators. Based upon the calculations performed, it is shown that, when using a thermographic technique to measure the SAR distribution of an applicator, heating times from 60-120s and measuring times of 10s are acceptable for an accurate assessment of the half-width at half power (HWHP) of an applicator (error less than 2%) with an expected HWHP larger than 2.5cm. Only when the HWHP is expected to be less than 2.5cm does the heating time need to be adapted to obtain an accuracy of 2% or better. For the assessment of the maximum SAR, the situation is worse. Even with a careful experimental design, it is difficult to measure the maximum SAR with an error less than 7%.