Rationale for the development of the UK defence standards forsafety-critical computer software

Changes in the procurement environment and developments in technology that will require the adoption of new development certification procedures within the next few years are examined. Characteristics of safety-critical computer systems and the safety problems posed by digital computers are described. The way in which the changes have influenced the preparation of the new draft defence standards offered for public debate in the UK prior to their formal adoption by the Ministry of Defence is discussed. Principle features of the future safety-critical systems policy are outlined. The use of Ada for safety-critical software is considered     

Improved tracking of maneuvering targets: The use of turn-ratedistributions for acceleration modeling

Tactically maneuvering targets are difficult to track since acceleration cannot be observed directly and the accelerations are induced by human control or an autonomous guidance system therefore they are not subject to deterministic models. A common tracking system is the two-state Kalman filter with a Singer maneuver model where the second-order statistics of acceleration is the same as a first-order Markov process. The Singer model assumes a uniform probability distribution on the targets acceleration which is independent of the x and y direction. In practice, it is expected that targets have constant forward speed and an acceleration vector normal to the velocity vector, a condition not present in the Singer model. The work of Singer is extended by presenting a maneuver model which assumes constant forward speed and a probability distribution on the targets turn-rate. Details of the model are presented along with sample simulation results     

High Dynamic GPS Receiver Using Maximum Likelihood Estimationand Frequency Tracking

A new high dynamic global positioning system (GPS) receiver ispresented and its performance characterized by analysis,simulation, and demonstration. The demonstration receiver is abreadboard model capable of tracking a single simulated satellitesignal in pseudorange and range rate. Pseudorange and range rateestimates are made once every 20 ms, using a maximum likelihoodestimator, and are tracked by means of a third-order fadingmemory filter in a feedback configuration. The receiver trackspseudorange with rms errors of under 1 m when subjected tosimulated 50 g, 40 g/s circular trajectories. The tracking thresholdis approximately 28 dB·Hz, which provides 12 dB margin relativethe the minimum specified signal strength, assuming 3.5 dB systemnoise figure and 0 dB antenna gain.     

Towards a coherent remote sensing data policy

Access to space-based remote sensing data is critical for Earth science and the study of global change. This article summarizes a variety of US government Earth science data policies and problems. The authors examine current efforts to develop data policies for the next generation of US remote sensing programmes, noting likely problems based on past experiences. They argue that the goal of US Earth science data policy should be to provide the widest possible dissemination of data. Setting such a goal permits the development of a simple, coherent data policy that serves scientific, commercial and US government interests.     

A SAR processor based on two-dimensional FFT codes

A synthetic aperture radar (SAR) processor approach based on two-dimensional fast Fourier transform (FFT) codes coupled with an asymptotic evaluation of the unit response function is presented. For the latter, no approximation is made to the distance function, so that the full range of geometric aberrations is analytically considered, enabling an effective reference filter to be designed. The two-dimensional FFTs were designed as to run on computers of very limited memory: the required FFT is computed by means of FFTs of lower order. Two FFT codes were considered: one is faster and allows full or reduced (quick look or multilook) resolution performance to be obtained easily; the second is slower but allows the use of a space-varying filter and/or investigations on limited portions (zoom) of the image. Both codes are suited to parallel processing, e.g. by a transputer net. A full discussion on computer memory and time requirements is presented as well as first examples of image processing results     

GPS based onboard and onground orbit operations for small satellites

In extension to common applications such as groundtrack displays and antenna steering, the SGP4 orbit model is proposed for operational orbit determination in small satellite missions. SGP4 is an analytical orbit model for Low-Earth orbiting satellites that is widely used for the propagation of NORAD twoline elements. Twoline elements may hence be generated completely independent of NORAD. Their use as exclusive source of orbital information simplifies the operations concept and reduces mission costs through the extensive use of existing low-cost mission support software. Due to small computer resource requirements of 8–10kByte, the SGP4 model may also be applied for onboard orbit computations making use of e.g. a 80186 processor, thus ensuring full compatibility of ground-based and onboard operations. The proposed approach is particularly suited in combination with a space-borne GPS receiver, were the C/A-code navigation solutions are treated as measurements that are adjusted in a least-squares sense using the SGP4 model. As consequence, inherent drawbacks of the pure navigation solutions such as data gaps and scatter as well as limited velocity accuracy are avoided, while the operational navigation activities are kept at a minimum. The feasibility of the concept is illustrated based on real GPS navigation data from the TOPEX/Poseidon and the MIR space station with an inherent data quality of 50–100 m. It is shown that 3 hours of data within a 4 day period are sufficient to keep the position error within 4 km, that is considered sufficient for most applications.     

Century scale solar variability imprinted in the 44Ti activity in meteorites.

The measurements of gamma-activity of the 44Ti (T1/2 = 66.6 years) produced by spallation reaction of galactic cosmic rays (GCR) in Alfianello, Olivenza, Rio Negro, Dhajala and Torino meteorites, which fell in the time interval 1883 AD (Alfianello)-1988 AD (Torino), show a century scale modulation, connected to long-term solar-wind flux variations in the interplanetary space. The variation of the 44Ti activity with the time of fall of meteorites is qualitatively consistent with the Gleissberg solar cycle, but the amplitude is three-four times higher than expected (about 5%) for GCR flux modulated by solar activity, as determined solely by the sunspot number. The cosmogenic 44Ti is a suitable radioisotope for this investigation, but its activity in meteorites is very low. We performed these measurements in the underground low level counting station of Monte dei Cappuccini in Torino, by means of a big hyperpure Ge crystal (approximately 2 kg) in selective coincidence with a heavy NaI (T1) scintillation detector (approximately 28 kg). This system allows a reliable and non destructive measurement of 44Ti (44Sc) in meteorites of 200-1200 g in weight. The background is about 1 count per day in the gamma-peak at 1157 keV of 44Sc in equilibrium with its parent 44Ti. The high stability of the performance allows long-lasting runs (approximately 10(7) s) in order to attain results with a standard deviation up to about 10%.     

Family of constrained signal detectors for hyperspectral imagery

Generalized likelihood ratio tests (GLRTs) are derived for the problem of detecting targets in hyperspectral images. These detectors are derived under the assumptions that the signals from the materials in the image mix linearly and that the noise in the system is Gaussian. It is also assumed that the abundances of the signals from the various materials in a pixel must sum to one. This constraint models the fact that the material abundances in a pixel are just the fraction of the pixel that they occupy. Under these assumptions, detectors are derived which outperform the detectors derived without the sum-to-one constraint.