and step-function biases. We confirm previous results that positioning errors affect SAR image formation through an increase in peak and/or integrated sidelobe errors, while bi- ases due to satellite outages and/or IMU integration error affect geolocation accuracy. We conclude that the wave- length is long enough at UHF radar frequencies to enable a tactical grade IMU to be substituted for the navigation grade IMU typically employed in higher frequency mi- crowave SAR systems.

Compatibility between the Mars Pathflnder spacecraft-generated (15,1/6) con- volutional code and the DSN Block III maximum-likelihood convolutional decoder has been demonstrated in the laboratory at all spacecraft data rates. Laboratory tests of the (15,1/6) code (unconcatenated) have found it to be superior to that of the (7,1/2) convolutional code. At 20 bps, the performance gain of the (15,1/6) code over that of the (7,1/2) code is 1.5 dB, and at 1185 to 11,060 bps the per- formance gain is over 2 dB at a threshold of 5£ 10¡3 bit-error rate. In addition, the Mars Pathflnder project has successfully completed a test campaign to measure the in-∞ight performance of the (15,1/6) convolutional code concatenated with the Reed{Solomon code. It has been found to match the laboratory tests performed prior to launch.

We have performed a study on telecommunication systems for a hypothetical mission to Mars. The objective of the study was to evaluate and compare the benefits that microwave-X-band (8.4 GHz) and Ka-band (32 GHz) - and optical communications technologies a afford to future missions. The telecommunication systems were required to return data after launch and in orbit at 2.7 AU with daily data volumes of 0.1, 1.0, or 10.0 Gbits (Gb). Spacecraft terminals capable of delivering each of the three data volumes were proposed and characterized in terms of mass, power consumption, size, and cost. The estimated parameters for X-band, Ka-band, and optical frequencies are compared and presented here. For all cases, the optical light terminal exhibits about 60 percent of the mass of the corresponding radio frequency (RF) subsystem. Power consumption is comparable for all three technologies at a 0.1 Gb/day data volume, but the power required at either Ka-band or optical is less than half of the X-band requirement at 10 Gb/day. These benefits can be obtained only with a suitable investment in reception facilities for Ka-band or optical frequencies.

In this part-II of the advanced communications benefits study, tow critical metrics for comparing the benefits of utilizing X-band, Ka-band and Optical frequencies for supporting generic classes of Martian exploration missions have been evaluated. The first of these is the overall equivalent communications system mass on the spacecraft. The second comparison metric is the overall cost impact. This 'overall' cost assessment has considered the costs for both the spacecraft end of the link and the ground end. In both cases the metrics indicate that higher frequency communication bands have favorable mass and cost, particularly at higher data volumes transmitted daily to the earth. The same metrics are also applied to telecommunication for a hypothetical Neptune mission, extrapolating from the designs for the Mars case.

Compatibility between the Mars Pathflnder spacecraft-generated (15,1/6) convolutional code and the DSN Block III maximum-likelihood convolutional decoder has been demonstrated in the laboratory at all spacecraft data rates. Laboratory tests of the (15,1/6) code (unconcatenated) have found it to be superior to that of the (7,1/2) convolutional code. At 20 bps, the performance gain of the (15,1/6) code over that of the (7,1/2) code is 1.5 dB, and at 1185 to 11,060 bps the performance gain is over 2 dB at a threshold of 5£ 10 i3 bit-error rate. In addition, the Mars Pathflnder project has successfully completed a test campaign to measure the in-∞ight performance of the (15,1/6) convolutional code concatenated with the Reed{Solomon code. It has been found to match the laboratory tests performed prior to launch.

Ka-band (32-GHz) radio communication links are more adversely affected by atmospheric noise temperature than X-band (8.4-GHz) or S-band (2.9-GHz) links. As a result, the strategy for telecommunications between a spacecraft employing Ka-band and the Deep Space Network is necessarily different from that employed at the lower frequencies. To aid in the development of new communication strategies, several years of 31.4-GHz water vapor radiometer data from the Goldstone and Madrid sites were reduced. Renewal theory was employed to examine the failure and recovery statistics of these Ka-band weather-dependent links.