Sirit, (formerly SAMSys), Durham, NC

Typical Responsibilities -

    • Complete RFID DSP code rewrite.
    • Cleaning up TX RF PSD.
    • RX performance improvements.
    • Strategic customer support.
    • System definition.
    • Coordination with control processor team.
    • System integration with strategic customers' systems.

Skills Attained -

    • AMD's VisualDSP++ Blackfin code development environment.
    • RFID technology - design and testing.
    • RF antenna parameters and selection.

General Observations -

    • SAMSys, later Sirit, was very typical of a small development company - each member needing to be able to wear a number of hats, and the company having to deal with a relatively narrow portfolio of products. SAMSys's strategic plan was highly focused on the widely predicted success of the RFID industry, which didn't even approach the anticipated levels of industry acceptance. RFID companies were folding right and left like books in a library. SAMSys/Sirit was one of the few which did survive, in large part due to the painful cutbacks they had to impose upon themselves. The SAMSys contingient shrank from a roster of over forty individuals to less than a dozen in a matter of months. Only keeping those who had been key members to the new product line's development team was the only chance of survival. As described above, Jerry's responsibilities were with the legacy product line. His work there was for the most part completed. With these thoughts in mind, Jerry understood Sirit's decision to bring him into their last rounds of layoff.

Notable Accomplishments -

    • Jerry joined SAMSys inheriting a severe crisis in the release of their newly miniaturized RFID module. It was already months behind schedule, and was performing extremely poorly. With a nearly complete restructuring and rewrite of the DSP code, Jerry was able to resurrect the product's performance to acceptable levels. That is, successful read rates were brought from an embarrassing ~50% to ~99%. Jerry completed this work in about 8 weeks of joining the company.
    • In RFID, the transceiver both transmits and receives simultaneously on the same frequency. In such a system, it is extremely difficult to eliminate RF leakage of the transmit signal into the receive section. SAMSys struggled with this problem with their legacy line of products. Jerry conceived of a concept to drastically reduce this leakage to acceptable levels. The system involved a self-directed closed-loop feedback servo to drive the leakage to nearly non-existant levels. Since there was doubt in the engineering management, and a prediction of failure from the team's engineering PhD, Jerry decided to simulate it. Jerry bet the PhD a beer that it would work. In a two-week period, Jerry learned to use Simulink for the first time, built a simulation of the system, including all known parasitics, simulated it and proved that it would work. The system was adopted very successfully in the company's new INfinityTM 510 product line, greatly enhancing the receiver's sensitivity. Regretfully, the PhD never did make good on the beer bet.